PDA

View Full Version : CONVERSION: Bridgeport Manual to CNC conversion...



Pages : [1] 2

Davek0974
10-05-2016, 04:21 PM
Anyone here converted a manual Bridgeport mill to CNC?


After building my plasma table and my mini-mill, both of which are working lovely, I WANT MORE CNC ;) ;)


My old 9x48 Bridgeport is looking for some love but is it worth it?


It would need a ball-screw conversion kit, these are available, plus I am guessing servo drives on X, Y and Z so I have encoder display, the knee doesn't seem to get driven on the few vids i have watched so far.


On the plus side, it's already powered via a VFD :D and i could likely recoup some cash by selling the working power feeds, control box and new DRO i fitted last year.


Worth a look or too expensive/not good enough...

komatias
10-05-2016, 05:34 PM
Yes, done it to something similar.

https://www.youtube.com/watch?v=M_VAXZbv6hc

has cost me around 3000 in total by now but only using steppers plus the cost of the machine. 1500 for the ballscrews as I went full hog with high preload and C3 accuracy. Also went with 300 worth of angular contact bearings to suit. Would have gone servo's if funds had allowed then.

Do not sell the DRO, I find it invaluable as you want to be able to switch between cnc and manual and also helps calibrate steps per unit.

From what I read the kit ballscrews are not as good as you are made to think, hence I went with my own ballscrews.

John S
10-05-2016, 06:06 PM
Don't bother.
Sell the manual and buy a dead CNC.

You have a machine that's currently worth about 900 ? you throw close to 2K at it and have a machine worth 800
Look for a BOSS 1 to 4 or an MDI machine. These were made as CNC, ballscrews, pressure oiling, a really nice quill where the ball screw is actually around the quill and not an add on that just wants to lever the quill sideways.

Downside of these machine which is in your favour is the electrics were crap, sorry make that $hit. Result is the iron is very good as ost have stood the last 15 years acting a shelf somewhere.

Beaver V5 here sold for 650

http://www.ebay.co.uk/itm/Beaver-V5-CNC-milling-machine-/182111147591

TOS CNC mill sold for 460 OK not working but that doesn't matter for a refit.

http://www.ebay.co.uk/itm/Tos-CNC-Milling-Machine-Heidenhain-360-Controller-/252353975743

You could have got 1/2 of that back reselling the controller.

Davek0974
10-05-2016, 07:45 PM
I think you might be right there John, it's being in the right place at the right time to catch the machines when they pop up, plus getting some transport arranged.

Will keep looking.

Davek0974
12-05-2016, 12:33 PM
Am I right in my belief that if i want real-time display of position in Mach3, I need to use servo's and encoders?

komatias
12-05-2016, 01:00 PM
Am I right in my belief that if i want real-time display of position in Mach3, I need to use servo's and encoders?

No, you can use optical encoders too. There is a screen out there that allows you to have 2 DRO's. However, you cannot use these as positional feedback in Mach3

Davek0974
12-05-2016, 04:25 PM
Hmm, just had a thought...

Instead of dropping around 2k on building a mini-mill/router, what I should have done is use that cash to convert the Bridgeport and fit an adaptor mount that enables me to hold the 24k water-cooled spindle - best of both worlds??

I could even mount a complete Z-axis on the rear of the ram where the slotting head should fit!

The BP has a lovely 48" bed, ok maybe it only has 9" of Y axis but thats still not too shabby I think, certainly I've done a hell of a lot of big jobs on the old girl.

So, what would the drawbacks be here?

The new spindle has the speed I need, I think a BP conversion can reach 4-5000 mm/min rapids which is where the mini-mill is set so thats no issue, ballscrews would provide the same accuracy as the mini-mill has presumably.

Hmmm......

komatias
12-05-2016, 04:33 PM
You lose Z travel when you convert the bridgeport's quill and another bit more when you add the adapter to hold the high speed spindle. Apart from that, I cannot see any issues except the speed at which your motors are able to move the bed. Good lubrication is key here. 4-5000 is way to quick for stepper conversions but possible with good servos. I have restriced mine to 3000mm/min actually.

Davek0974
12-05-2016, 04:40 PM
Ok, but by building a custom Z-axis and mounting it on the rear of the ram I would retain whatever travel I build into the new axis, it means swivelling the ram round to change from heavy to light milling but thats about it, plus would need two z-axis drives - heavy and light.

The Z on the heavy side is a pain, a couple that i have seen today - one replaced the quill lever with a toothed pulley and drove that via a planetary gearbox reduction - seemed ok, another replaced the fine-feed hand wheel and drove that, the ones that mount to the front of the quill I am not too impressed with, have read a few tales of quills being trashed by the single bolt pulling out.

As for speed, so far most of my mini-mill work ranges from 600mm/min to 150mm/min - hardly blistering, I'm sure the old girl can match that ;)

I think this may be an interesting project.

komatias
12-05-2016, 05:20 PM
Yes, that would do it. I have gone with direct drive on the quill and do not over drive it. The fine feed goes through a clutch while the backlash on the lever is too much for CNC.

Davek0974
12-05-2016, 06:06 PM
Good points, thanks.

By direct, I gather you have the screw nut coupled to the front where the depth-stop used to sit?

JAZZCNC
12-05-2016, 06:51 PM
It's slippery slope and quickly your into it $$$ before too far to stop.!! . . . Don't Do it Listen to John S.

Davek0974
12-05-2016, 07:08 PM
awww, bum, damn that cold hard logic JAZZ ;)

In my defence, it would pretty much fit in with exactly what i want though!

So, the alternatives then -

Find a donor machine and hopefully just fit the electrics?

Downsides I can see...

space - nowhere to put it.
would like a big table again - this seems to mean a big machine (see point above)
transport costs
unknown wear & tear
different tooling probably

This seems not too bad, but its bed is quite small, although it does have an ATC - would that be the sort of stuff to search for???

http://www.ebay.co.uk/itm/182121347063?_trksid=p2055119.m1438.l2648&ssPageName=STRK%3AMEBIDX%3AIT

JAZZCNC
12-05-2016, 07:17 PM
Yep I'd take that every time over converting BP.

Davek0974
12-05-2016, 07:38 PM
Hmm, looks like a no-go then - it weighs 4,500kg and has a 7.5kw spindle - i could not get that sort of power in the shop and anything much heavier than BP would be a bitch to get into place.

The bed is poor too - 500x600 not much bigger than my mini-mill!

JAZZCNC
12-05-2016, 07:56 PM
Hmm, looks like a no-go then - it weighs 4,500kg and has a 7.5kw spindle

Can't beat fat lass with bit of muscle.!!! . . . . But ye got to be practicle. .:thumbdown:

Davek0974
12-05-2016, 08:14 PM
That's the thing - I know mass rocks when it comes to machines, and also see the point of having a pre-ballscrewed donor but as you say - I need to be practical if I bother to do this.

It would have to be viewed as what it is - a computer controlled Bridgeport Knee Mill NOT a VMC wannabe. I really can't squeeze anything much bigger than a BP in and have no expansion options as we are residential so have to be careful a bit ;) There is no garden left to build a bigger shop.

In the past I have had bigger stuff - a Colchester Student lathe, a large horizontal mill and a vertical slotting machine - all the stuff i used when building traction engines for fun. Back then I would have killed for a CNC bridgeport but had neither the cash or the knowledge. Using a boring head to machine a 25mm hole out to 100mm and only being able to take small cuts gets old real fast - that job would be seconds with CNC ability.

I'm still not convinced and a bit pissed that the machine I linked to above costs less than the ballscrew kit alone :(

jamesgates1000
12-05-2016, 09:22 PM
Why don't you trade your Bridgeport in for a Bridgeport interact and convert that to Mach3?

Davek0974
12-05-2016, 10:04 PM
Why don't you trade your Bridgeport in for a Bridgeport interact and convert that to Mach3?

Something like this???

http://www.ebay.co.uk/itm/Bridgeport-Interact-Series-2-Milling-Machine-/141981320255?hash=item210ebf043f:g:oZsAAOSwYmZXD2I q

jamesgates1000
13-05-2016, 10:51 AM
Looks like a pretty good deal, I have used them before with the heidenhain control and they are good machines. Best wait for some advice from the others regarding suitability for Mach3 conversion.

Davek0974
13-05-2016, 10:54 AM
Luckily, I got the tape measure out last night.

I can't fit anything bigger than the standard Bridgeport in! That only fits because the head is between the roof trusses, so that rules out practically all others I think as even an Interact is 90" tall which i think is 12" more than a standard machine.

So my options now seem limited to converting my own BP, buying another BP and converting that then selling my on on or not bothering.

Davek0974
13-05-2016, 12:58 PM
Just had a lengthy chat with my supplier of bits and they seem to favour heavily towards 8NM stepper-servo's for this task. The specs read well, basically a closed loop stepper and DSP drive with variable current etc. Price was about 250 per axis for motor and drive with cables.

Being stepper also means i can use my existing knowledge of BOB's and USB motion controller etc.


More options to throw in the pot.

Davek0974
14-05-2016, 12:13 PM
Knowledge gap...

Stepper-servo motors, I know I need the drive plus a BOB plus a motion control - done that before and it works.

AC servo motors, I know I need the drive amp plus ? - what goes between the drive unit and the PC?

Just trying to price up, the price is identical between 8Nm stepper-servo motors and drives OR 750W servo motors and drives - I have no idea how to choose between one or the other, one I know the other I do not.

I am advised that 750W is ideal for the job so thats one decision, also the 8Nm step-servo is ideal it seems.

Anyone want to fill the gap?

I know the servo is rated to 3000rpm and the stepper to maybe 1000rpm?

The stepper torque is way down where it's needed but the servo is where, linear spread?

Looks like a ballscrew kit is going to be around 1300 so thats a fair chunk of cash in one hit, the rest I need to get right first time.

JAZZCNC
14-05-2016, 02:16 PM
Ok well in theory there's no difference in requiremnets between Servo and stepper for connecting to PC. However in practise there is specification difference.

Servo's require much higher Pulse frequency and much cleaner signals. Now your average 5 Bob doesn't cut it and the Parallel port is just doesn't cut the mustard.

The speed and resolution you get from Servo is dependant on the encoder fitted. The encoder also determines the pulse frequency required.
For instance most Servo come fitted with 2500 count encoders which are quadrature encoders so this number is *4 so 10,000 pulses required for one revolution at full speed. If 3000Rpm that's 30,000,000 pulses in 60s. Break this down to seconds and that's 500,000 pulse per second or 500Khz required to achieve full motor RPM.
Now PP is only 25Khz or at realistic best 45khz 45,000 pulses. At best this means you'll only get 270rpm out of your 3000rpm motor.

Now most servo drives provide a pulse multiplier so one pulse becomes 2, 4 etc depending on amount set. This however comes at the cost of Encoder resolution which gets reduced by the same amount. IE: 1:2 Multiplier 10,000 = 5000ppr 1:4 2500ppr etc.
Not problem your thinking I don't need all that resolution anyway but it's not just resolution that suffers, so does smoothness of the motor and really the whole point of servos is for there superior Resolution and smoothness along with linear torque.

So bottom line is if using servos then really you need an external motion control card that gives you required pulses of good quality.

On more positive note there are other advantages to servos than just resolution and linear torque. 3000Rpm often is too fast so you'll use ratio which increases torque so smaller cheaper servo can be used.
Torque is linear and the rated servo torque is when at full motor speed unlike steppers which is when at stand still and drops quickly when rpm rise. Large steppers drop very quickly and not much use after 900-1000rpm.
Servo drives are powered directly from mains so NO PSU required.

If I was converting a mill then I'd use servos because they are much stronger than Steppers and smoother in operation. Torque is linear so much stronger when your pushing hard and deep.

Davek0974
14-05-2016, 03:23 PM
Thanks JAZZ,

The UC300 USB motion controller goes to 100kHz so that seems ok, I only suggest that one because I have used the smaller UC100 with total success and the price is pretty good too. However I do not know yet if that interfaces direct with the servo drive unit?

Not having a high power PSU is a big plus as well and as there is no price difference it looks like servos have won the toss :)

Would you go with 750W motors or smaller??

2:1 reduction for resolution & torque increase, more?

Servo drives seem to have lots of settings, what is used - speed & torque, positioning..... All new to me but i have the manual for the drives now.

JAZZCNC
14-05-2016, 04:44 PM
The UC300 USB motion controller goes to 100kHz so that seems ok, I only suggest that one because I have used the smaller UC100 with total success and the price is pretty good too. However I do not know yet if that interfaces direct with the servo drive unit?

Well not really 100khz is still 400khz short of full speed/res. Or 1:5 multiplier so only getting 5th of resolution. Plus it's USB rather than ethernet so wouldn't be my first choice.
Really Ethernet Smooth stepper is what you want or feeling really flush then go for Cslabs Controller.


Would you go with 750W motors or smaller??

Probably 750W but really you should do some calcs to get better idea of size, inertia ratio etc. Like steppers Bigger isn't always better.


2:1 reduction for resolution & torque increase, more?
Again depends on other things like screw pitch and what kind of speed/res/torque your looking for. Thou 2:1 with 5 or 4mm pitch screw sounds about right. My Big Semco CNC Mill uses 5mm pitch with 2:1 ratio and 1.8Kw servo's but it Dwarfs BP.
Matt (kingcreaky) on the Forum used 750W servos with 2:1 ratio on BP Interact sized mill and was happy with them. (He's got a video some where on this site)


Servo drives seem to have lots of settings, what is used - speed & torque, positioning..... All new to me but i have the manual for the drives now.

If using Step Direction then you use Positioning. If using Analog +/-10v control then would use Speed. You'll be using Step & Dir so positioning.

Difference between Step/Dir and Analog is mostly how they are controlled and how the Closed loop works. With S/D the loop is closed only between drive and Motor. Analog the Loop is extended between Controller drive and Motor.
The result is that Analog feeds the Encoder position back to the controller, which it must do to work, which then checks this position with the commanded position and if wrong sends corrective signals.
The upshot is that the DRO's always display the actual position of the table, no matter if mach or controller as commanded this or you whind it there by hand. In affect works just like Manual DRO's.

Step/Dir on the other don't do this. It sends commanded position to the drives and takes it on faith that it arrived there. The drive then takes control of getting to commanded position and any corrections needed.
So the DRO's only show the commanded position and if it's wrong or if for instance something is loose you'll never know your out of position. Essentially works just like your steppers now with the exception the drive is checking the motor actually moved the distance told.

Bottom line is S/D is easier to setup. Analog gives more feed back but takes more setting up and tuning of drives.

See this video I made showing 1.8Kw servos controlled via Analog.


https://www.youtube.com/watch?v=tW8knRAOr7s

Davek0974
14-05-2016, 05:09 PM
Nice video, that would be what i would like ideally - feedback.

So The option is now ESS at about 150 and no feedback or CSMIO/A at about 490 with all the bells and whistles.

The motors I was looking at are from the same guy as in your video, drives the same too.

The 750W suggestion comes from the CNCZone where a builder mentions doing 30+ conversions of this type and always uses 750W, 1kw was too large and 400W too small.

He uses 1:1 or 2:1 but I favour 2:1 as I want to get the most torque I can and the machine won't handle massive rapids.

Ballscrew kits seem to be 5tpi so 5.08mm pitch

Question is - is the benefit of real-time position DRO worth the extra cash?

JAZZCNC
14-05-2016, 05:25 PM
Question is - is the benefit of real-time position DRO worth the extra cash?

Yes IMO but with the Cslabs controller there's more to it than just this. You also have very accurate Homing using the Index pulse of the Encoder. You can also Fit a Servo to the Spindle and control that just like an axis so can index to position or do things like ridged tapping etc.

Then you have 24V I/O etc not to mention you don't need a BOB. This Alone is BIG Plus, esp when using servo's. Don't confuse the IP-S with ESS they are completely differant animals.
While they seem expensive they are not actually that bad when you consider that it's one complete unit nothing else is required. It's also Industrial spec and high quality.
By the Time you have Bought ESS and Decent quality BOB like PMDX126, which about the only one I'd use with servo's plus Spindle speed Control board then your not long way behind the IP-S or IP-A.
They really do not compare other than both use Ethernet and provide 4Mhz after that it's all down hill for the ESS.!!


The motors I was looking at are from the same guy as in your video, drives the same too.

Think you'll find 750W will be most likely be 80 frame(like Nema34) not 110 which will probably be better for you. Drives may be smaller too.?

Davek0974
14-05-2016, 06:06 PM
Ok JAZZ, i'm sold on the CSMIO/A unit, does it have PWM output or is that an add-on? Controlling the VFD would be nice.

Will be reading up on that tonight.

Rigid tapping is a bugger because you need to get an encoder on the spindle - this is a nightmare the Bridgeport manual machines. Possible later add-on though.

Yes I think those motors will be Nema34 size.

JAZZCNC
14-05-2016, 06:23 PM
Ok JAZZ, i'm sold on the CSMIO/A unit, does it have PWM output or is that an add-on? Controlling the VFD would be nice.

Yes it as Analog 0-10V Output for spindle Control built into it. Also has few Analog Inputs so can use Pot to override spindle speed and feedrate if desired.

m_c
14-05-2016, 09:09 PM
A couple things to remember with the CS-Lab controllers, is if you want rigid tapping, you need to add the ENC/threading module, although not really an issue if you're happy to use a tapping head or thread mill.
Also, if you want to add a MPG, you need to buy the MPG module.
I would of suggested a KFlop+Kanalog (+ a Konnect if you need more than 8 In or 16 Out) as they'll do whatever you program them to do without any additional modules, however you need to know at least a little bit about C programming to get them working.

I was advised for my big mill, that the larger motors would be a better match. I was looking at 1000W 80 frame motors, and 1050W 110 frame motors, although I think I'll have to get 1200W to get the 3000RPM now. However I do think you could probably get away with the smaller frame motors on a standard Bridgeport, as it's not that heavy a table.

Davek0974
14-05-2016, 11:11 PM
fitting an encoder to a BP is going to be hard i think, not many places to access the spindle, I have tapping heads so not really a concern yet.

I'm fairly well recommended from previous converters that 750w is a good match for this job, 3:1 reduction should help too.

I really don't want to get into C programming ;)

dodgygeeza
15-05-2016, 10:24 AM
Then put an AC servo on the Spindle? I have done this on my MDI.
Its 3KW and 4KRPM. Bit small but i will see how it goes.

Davek0974
15-05-2016, 10:26 AM
Control options...

Not sure we need a alpha keyboard for a mill, hows about a touch-screen monitor to do away with the mouse and a small numeric USB keypad?

Or

normal monitor (cheaper/stronger) and a track-ball and numeric keypad?

Looking at controls..
X/Y joystick
Z joystick
Knee up/down buttons
Speed override knob
Feed override knob
Cycle start
Feed-hold
Stop


Spindle motor...
This is currently run from a VFD but only at 50Hz, the head is the vari-speed unit with back-gear so i have 50-3000rpm but in two ranges and manual speed control. Whats a good option here - using the VFD won't address the variable torque ability of the vari-speed unit so can't just lock it in one spot.

Any cool options here to play with?

With a bit of fiddling the vari-speed changer could probably be motorised with a stepper and gearbox, not sure it's worth it?

Mach can obviously take over the Fwd/Rev function - still using the old switch but I wired it to the VFD so I can plug-reverse for tapping;) That was before i got the tapping head.

Any cool ideas?

Davek0974
15-05-2016, 03:15 PM
Z axis...

It looks like I have four options here...

1 - Drive the Quill direct from the bolt point where the depth gauge mounts at the front.
2 - Drive the manual lever shaft on the right side
3 - Drive the power-feed clutch shaft on the left side
4 - Drive the knee lift

Points -
Option 1 seems to be the most common but adds a lot of bulk to the front of the head and costs a little Z axis travel.

Options 2 & 3 are similar, both will suffer from the rack backlash but option 3 is the neatest one, backlash can be reduced with the clock-spring.

Option 4 is a contender I think, it gives a massive 300mm Z travel, but suffers from stickiness and has backlash in the bevel gears. The force is reasonable - I measured my one and if my figures are correct, it needs approx 1Nm to lift the knee based on the handle length of 160mm and 6kg force on that handle 6kg / (1000/160) = 0.96Nm

The backlash is a factor but the Z is usually used to plunge or position - both of these are done in the positive direction so maybe backlash does not matter here?

Thoughts?

Davek0974
16-05-2016, 04:43 PM
Based on advice, 750w AC servos and 2:1 is ideal on a bridgeport so that is whats going in, totally overkill for the Z axis but the price of smaller motors is higher so i cant see the point really, keep it all the same.

Got a few parts on the way - stainless panel mount keyboard, stainless panel mount trackball, some bits from the box at work - nice 10-turn pots and dials for FRO & SPO controls, a few spare relays and other bits and bobs.

Just about got a source lined up for a ballscrew kit - this bit really hurts the wallet :(

Does anyone have a drawing or sketch that would give me the dimensions of the right table (X) bearing bracket and the front (Y) bearing plate??

Mine will be missing as I have the original 6F power-feeds which will be taken off when i convert.

I can get the parts but the prices seem silly and they surely must be an easy enough part to make up from steel or aluminium?

Thanks

Davek0974
18-05-2016, 09:03 AM
Drive belts - HTD5mm x 15mm wide big enough?

JAZZCNC
18-05-2016, 10:16 AM
Drive belts - HTD5mm x 15mm wide big enough?

They'll work ok but i'd be more inclined to go with 25mm for Mill with heavy bed.

Davek0974
18-05-2016, 10:24 AM
They'll work ok but i'd be more inclined to go with 25mm for Mill with heavy bed.

Hmm, will have to check pulley dimensions will fit machining on shafts, there is a conversion kit which uses 1/2" belts, I thought that was a little weak so went for the 15mm size as i used before.

Will look into the 25mm size

Davek0974
18-05-2016, 02:33 PM
Drive belts, just had a chat with a very knowledgeable guy and it seems HTD is the best option here, GT is better but more expensive all round. The backlash concern is valid but equates to approx 0.25mm linear which if i have my sums right on 5mm pitch screws and a 75mm pulley will give me an error of 0.0053mm - I think thats acceptable on a 30+ year old machine ;)


He did however strongly recommend going for 8mm pitch and 20-30mm wide belts.

Can't use 8mm pitch as you can't get the needed ratio in as the pulleys are so much bigger so will have to be 5mm pitch.

Davek0974
18-05-2016, 08:36 PM
CSMIO unit on order, motors on order, ball-screws ordered, wallet crying :(

Panel controls...

With the Bridgeport you have to run in reverse when in low range, is there a way to have this done in Mach3 or will it still be a Hi/Low selector switch on the panel?

The VFD for main spindle, with the vari-speed mechanical system I'm not sure if i should try using the vfd as variable or leave it fixed at 50Hz, the top speed is limited to 3500rpm absolute max as it tends to wreck the spindle bearings above that and low down it probably would not have the torque needed, so stick with mechanical ?

Main spindle on/off manual control - would these buttons control Mach3 or would they control the vfd directly?

Just looking for ideas here.

Davek0974
20-05-2016, 02:01 PM
Food for thought...

While waiting for the screws & motors to arrive, should I strip her down to the bones and clean / paint the old girl ;)

Would take a bit of effort as it would need space clearing to get the engine hoist in to lift the head, ram, collar, bed and knee off.

Could be useful as it would allow me to check for worn out bits, split lube pipes etc.. Would also make the build and wiring a cleaner job by getting rid of the grime and muck.

Hmm....

komatias
20-05-2016, 10:05 PM
Food for thought...

While waiting for the screws & motors to arrive, should I strip her down to the bones and clean / paint the old girl ;)

Would take a bit of effort as it would need space clearing to get the engine hoist in to lift the head, ram, collar, bed and knee off.

Could be useful as it would allow me to check for worn out bits, split lube pipes etc.. Would also make the build and wiring a cleaner job by getting rid of the grime and muck.

Hmm....


Yes and get the knee saddle and table reground and turcited. Speak to slideway services, they quoted me by far the best price

Davek0974
21-05-2016, 06:22 PM
Looking at replacing the mechanical speed control and back-gear with an AC servo motor, something like this...

http://www.aliexpress.com/store/product/New-5N-M-1-5KW-3000RPM-110ST-AC-Servo-Motor-110ST-M05030-Matched-Servo-Driver/314742_760988112.html

Should give full power across the board I think, also no loss in the drive belt and no changing gear for low speeds.

komatias
21-05-2016, 08:40 PM
Bingo! Much better than your VFD approach. Have a look at the ones rated for spindle usage. They will work better at higher rpms. i have specced a GSK one on my new machine. Now you do not really need to have it go full servo unless you want to do rigid tapping.

JAZZCNC
22-05-2016, 12:46 PM
Bingo! Much better than your VFD approach. Have a look at the ones rated for spindle usage. They will work better at higher rpms.

I agree George the AC Inverter Spindle Motors are the correct approach but they are still often controlled via VFD or Inverter. My Semco Mill as 5Kw Inverter Spindle Motor and that is controlled via ABB inverter.
Difference between ordinery motor being will spin to 7500rpm and still provides lots of torque.

The latest AC Asynchronous Servo Spindle Motors are great but would be completely OTT for Old BP like this and cost more than the whole setup.

Personaly for the price of cheap VFD I'd use what you have. If the BP motor is upto it and you have tapping head don't see the point.?
Edit: I might be tempted to remove the Variable drive setup and go direct drive but wouldn't bother with Servo unless need the Indexing etc it provides. Even then this is possible to lower degree by fitting an encoder to vfd provided it supports this option.

18486

Davek0974
22-05-2016, 02:14 PM
Hmm, got the head apart now, to refit it will need new bushes in the sheaves plus a new drive belt, probably 100 odd? I already have a VFD on it but only use it as a converter at 50Hz.

OR

I can pull the lot, fit a 1.8kw AC servo rated to 3000rpm, 1:1 poly-v drive and along with the back-gear if ever needed would give me full control from G-code - total cost about 400

So we are looking at about 300 difference for luxury servo drive??

What to do ;)

komatias
22-05-2016, 05:26 PM
Personaly for the price of cheap VFD I'd use what you have. If the BP motor is upto it and you have tapping head don't see the point.?
Edit: I might be tempted to remove the Variable drive setup and go direct drive but wouldn't bother with Servo unless need the Indexing etc it provides. Even then this is possible to lower degree by fitting an encoder to vfd provided it supports this option.



Davek,

see what the man says above. no point in going servo on a bridgeport.

Either way, do look at getting a power drawbar and some quickchange TTS type tooling

JAZZCNC
22-05-2016, 07:11 PM
I can pull the lot, fit a 1.8kw AC servo rated to 3000rpm, 1:1 poly-v drive and along with the back-gear if ever needed would give me full control from G-code - total cost about 400

So we are looking at about 300 difference for luxury servo drive??

What to do ;)

Thats 300 extra for what.? Ridgid tapping that you don't need.! Indexed postioning you don't need because no ATC.!
Controller will give you every thing you need regards speed control etc from G-code using VFD just like it would if Servo so don't see the point.
Only advantage I see is the indexing and possibly extra torque at lower speeds, but you have back gear so not problem.

Save the Dosh and spend it else where.!

Davek0974
22-05-2016, 10:06 PM
Yep, it was a nice thought but had a good think over dinner and using the existing motor and VFD is the sensible option. As said, rigid tapping is not needed and using the variable mechanical drive with the VFD as well will give full spread of speed and torque from 40 to 3000rpm if using back-gear or 400 to 3000rpm in high alone. My tweaked macro will tell me where the mechanical dial needs to be so problem really.

It was a nice thought but a luxury and cash will be needed for fitting out the cabinet etc, when i find one big enough.

Davek0974
23-05-2016, 11:20 AM
CSMIO, servo motors and drives arrived a minute ago, all looks very heavy-duty indeed, just ordered a DB25 break-out block for the MPG unit and one of the neat little power divider blocks from CS-Labs :)

Got the mill head stripped, paint on order, removed all the surplus gears and junk from the head.

Wondering about doing the neat Z axis fit where the ball-screw takes the place of the feed-stop screw and is a 1605 unit OR doing the more common fit where the ball-screw sits in front of the original feed-stop screw position???

The only downside I see with the first option is that the pulley sits at the bottom of the screw right in front of the spindle nose.

komatias
23-05-2016, 01:14 PM
I will send you a photo of my setup later on. I have made parts to allow me to use the manual quill with the removal of a single screw.

Also look on CNCzone for SCzEngrgGroup's conversion of the bridgy.

Davek0974
24-05-2016, 01:38 PM
Have motion on the test motor now :)

Needs tuning i am told but there does not seem to be a tuning option in the drive menu??

I did an auto-tune in CSMIO but that only goes so far, even the manual tells me it needs to have the drive velocity tuned in the driver before auto-tuning or it will fail.

lucan07
24-05-2016, 02:30 PM
Sounds like a question for Jazz, write that you've decided to use a 6560 and he will appear:triumphant:

Davek0974
24-05-2016, 07:21 PM
It seems the drive is pre-set when enter the motor code number so no tuning in the amp.

Have finally got the head apart on the mill - a tiny broken shaft was holding me up - an absolute bugger to get out.

18494

18495

Now thats off it should move a bit more rapidly. The plan is to do a complete strip, wire-brush the poor repaint job it has off then sand and repaint in "Bridgeport Grey" engine enamel and reassemble to a bare machine then fit the CNC conversion.

The DRO and Original, working power-feeds will be sold off to raise funds.

Space is the biggest problem, will probably have to erect a shelter and do most of the prep in the garden.

Davek0974
25-05-2016, 04:26 PM
Thoughts over coffee - Limitations of the system?

When i was building the large scale traction engines, everything was made with a tolerance of only 0.1mm or better - easily obtainable on a hand controlled Bridgeport with a three digit DRO. It really was not needed to go tighter on these engines and many say that if too tight they never worked properly due to expansion when running etc.

Anyway, the third digit on the DRO was pointless as you could never set that precision by hand and merely looking at the mill out of the corner of your eye made it wobble more than that!

So, now (when done) I will have ball-screws and servo's with Mach3 acting as my DRO display - this is all running on my bench very nicely now.

But - the encoder/motor is connected to the screw via a toothed belt - surely this must introduce an error possibility, especially with a 100kg sitting on the bed??

My main point here is - If i want to use the mill manually, say to drill some holes without cam-ing the job, I can put Mach into reset, this drops out the servo motors and i can spin the handles like before and Mach tracks the position on screen - I am hoping here that this is a workable situation with decent accuracy when positioned manually??

Davek0974
25-05-2016, 04:28 PM
Yes and get the knee saddle and table reground and turcited. Speak to slideway services, they quoted me by far the best price


At 950 + VAT + transport there and back this will have to wait i'm afraid. I have chrome ways so that put the price up as well.

JAZZCNC
25-05-2016, 08:11 PM
Sounds like a question for Jazz, write that you've decided to use a 6560 and he will appear:triumphant:

No I won't .!! . . . . Because I already told him via Private bongo's.:monkey:

JAZZCNC
25-05-2016, 08:16 PM
But - the encoder/motor is connected to the screw via a toothed belt - surely this must introduce an error possibility, especially with a 100kg sitting on the bed??

Not really if the belts are tensioned correctly and decent quality. Lots of very expensive VMC have belts on the screws.

Regards Manual operation then it will Display position just like Manual DRO's would.

Davek0974
25-05-2016, 09:33 PM
Was playing around with PID tuning today, reading the manual on how to do it manually (ooh-err) It was not happy and no amount of fiddling got it better, long positioning time, not sitting at position etc. That was until i stuck my finger on the end of the shaft and gave it a tiny amount of resistance - instant improvement rapid stopping on position etc. Lesson learnt, seems unloaded motors don't like tuning ;)

I have ramped it up to 3000rpm, been deliberately messing about with rapids for several metres each way plus millimetres each way, then put mach into reset and jerked the motor around, reset and did a go-to-zero and the damn thing snapped straight back to my zero mark - I was most impressed with this feat ;)

JAZZ - how did you wire the enable line to the drive - there is a signal mis-match here as CSMIO puts out 24v but the drive needs to be pulled low - I stuck a relay in which works but....

Also, what size fuse would be good for a drive and 750W servo?? Thinking 5A slow-blow?

JAZZCNC
26-05-2016, 02:24 PM
JAZZ - how did you wire the enable line to the drive - there is a signal mis-match here as CSMIO puts out 24v but the drive needs to be pulled low - I stuck a relay in which works but....

I do it with a relay tied to the HV option on IP-A that enables all the drives.

Davek0974
26-05-2016, 02:40 PM
Thanks, if it really is relays then thats fine, I have that working on the bench setup, just wondering if that was the only option here :)

Davek0974
26-05-2016, 03:29 PM
Ok, where did I slip up...

Previously had the motor tuning set as book, 2000 steps per (10,000 / 5mm pitch)

When I fit this on the machine it will be at 2:1 ratio so i set steps per to 4000 (10,000 * 2 / 5mm pitch) and set velocity to 7500 (3000rpm / 2 * 5mm pitch) acceleration is at 750mm/s/s

Works ok in jog mode, works ok in code mode at speeds up to 7500mm/min

If i do a "Goto zero" it runs for a while then stops with ePid fault.

It also allows me to set feed higher than it can handle and then goes into ePid fault.

Is there a reason here?

Davek0974
26-05-2016, 04:54 PM
Ok, found it - it seems it does not like running right at the limit of max speed - if kept just below max it all works perfect.


If I set the max at 7500 and command a feed at 7500 it faults, if I set 7400 it works perfectly in all modes.

Won't get t hat high in real life as the bed is not really big enough i think for a 7500mm rapid ;)

lucan07
26-05-2016, 10:23 PM
Just to throw a spanner in the works may be completely irrelevant but on my travels today saw a bridgeport (couldn't tell you the model) converted to CNC, that used original and a 24k spindle slightly different setup to your plan.

The 24k spindle was mounted literally beneath the original in a frame so when changing between spinles he just had a Z offset to consider but apparently with Knee or second Z thats not a problem, frame actually locked into a collet then connected to frame, his spindle was shorter than mine and he lost a fair bit of Z travel but this still left a good amount of travel on Z for what he used the 24k for.

Just thought I would share unfortunately I was out on the road bike (lycra man) to escape idiots for a while so no phonewith me to grab a picture.

Davek0974
28-05-2016, 12:41 PM
Should these collars just slide off, it feels like they are hitting a hard stop?

1852018521

Probably surplus on a CNC machine anyway but don't know yet.

lucan07
28-05-2016, 02:28 PM
Just a thought, have you tried tapping them round had similar collars in a different setting but they were threaded.

Davek0974
28-05-2016, 04:25 PM
No, one of them slides about 3mm and then stops, the other is stuck fast

Davek0974
28-05-2016, 06:23 PM
Can't go much further ;)

18522
18523

Stripped most of the poor paint job off now, will be sanding and painting this weekend, not filling the dents in as it's a machine not decoration ;)

Davek0974
29-05-2016, 11:23 AM
Now for the spare parts :(

The knee lift-screw bearing is dead, it looks like a 3306-2RS Double row angular contact 2 rubber shield item - 30mm ID, 72mm OD, 30mm W

Can anyone confirm that please?

Also is there a good source for the tubes and fittings for the lube system in the UK, might replace the lot as it looks like some of the points on the knee lift ways were doing nothing, luckily i always backed up with the oil can;)

lucan07
29-05-2016, 12:21 PM
Off subject just making sure you have this came across it looking for something else looks interesting

http://neme-s.org/Shaper%20Books/bridgeport_j1_head_2_0_01%20Rebuild.pdf

m_c
29-05-2016, 01:07 PM
I got replacement Bijur metering units for my mill from Lubetec. They do have a web shop, but it's just a generic skinned airline/pneumatics one used by lots of similar suppliers, and doesn't contain any of the lubrication stuff.

They do however have their own catalogue online, which contains all the centralised lubrication stuff.

I did consider the Arc Euro parts, however the threads were wrong and the genuine Bijur ones weren't that much more expensive. Just checked and they were 5.54+VAT each a few years ago, so not worth the hassle of retapping threads and changing fittings.

You'll need to get the flow rate of the existing metering units, as well as the thread/style/direction of flow if you need new ones.
I removed all the metering units from my mill, connected them up, and then left the lube system running overnight (it's an electric motorised one designed to lube every few hours). Any that were still dry in the morning, I replaced as they're pretty much impossible to clean properly once gummed up, and a few metering units is far cheaper than the potential damage from lack of lube.
One trick I did use, was to fill the system with ATF fluid and leave it running for a day. You can then check that all the slides have a nice coating of red oil, which tells you the system is working and getting oil to where it should.

For replacement pipe/fittings, I just used whoever could supply them. My mill originally had aluminium distribution pipe, which had corroded where it had been sitting in coolant, so I replaced it with the same sized copper pipe.

Davek0974
29-05-2016, 01:14 PM
Thanks I'll have to test them all first then.

The plumbing is all nylon so that should be cheap enough with some olives.

Mine is a manual pull-pump by interlube.

Davek0974
29-05-2016, 05:17 PM
Looking better...

18524

Got the knee painted too.

Hit a block now - turns out the knee-lift screw bearing is dead, got the deadly silver paste coming out of it and feels like a bag of nuts. Will have to order one from somewhere but cant reassemble until i get it :(

Davek0974
30-05-2016, 03:52 PM
Got most of the big stuff stripped , sanded and painted now. Started putting a few little bits back on but cant do much until i get the new knee lift bearing.

Found an issue with old lube system - this is the feed to the saddle...
18527
Olive was cocked and tube crimped, probably wrong size olive for the pipe :(

Moving on....

Motor mount plates for X & Y axes.

Is it better to hang the plates from the four bracket mount screws or the three bearing retainer screws?

18528

Using the four screws would mean a much larger lump of metal is needed to span the width, I won't be refitting the dials so using the three mount screws probably won't affect anything.

Any thoughts?

komatias
01-06-2016, 10:07 AM
Go off the plate:


https://lh3.googleusercontent.com/-uOJ0NtEx5Vs/U35VNfN7iMI/AAAAAAAAH9g/xaAdmzZi5gcDHhAlKHsxiCRioMH__aL3gCCo/s800/mounts%2B-%2B4
https://lh3.googleusercontent.com/-RrU1ehdec0w/U35VNblnaCI/AAAAAAAAH9o/qEohKw1nKvoReFsiwozPGQwT1-9wkpwsgCCo/s800/mounts%2B-%2B5


I made a spacer and clamped it all there.

My Y axis one looks like this:

https://lh3.googleusercontent.com/-aXZpJW_Oc8Y/U35VNcgBpgI/AAAAAAAAH9g/QuQjTMt2YoAip7lB-U9MgM0pO1efhwXJgCCo/s800/mounts%2B-%2B3
https://lh3.googleusercontent.com/-vPrW2JFgWTA/U35VNexEzII/AAAAAAAAH9g/LdAVWXtQZ-41DAXP0zVf6liC4ksVNBZUACCo/s800/mounts%2B-%2B7


I bought proper angular high spec contact bearings for the screws (+250) and some very fancy face locking lock nuts from Germany.

Davek0974
01-06-2016, 10:17 AM
Very nice :)

Have you any more pics, looking for ideas on homing/limit switches etc

I had not thought of putting the Y motor to the rear, all the builds i have seen have it hanging down low beneath the handle, it will have to stick out though as my motor is too long to tuck under.

komatias
01-06-2016, 10:32 AM
Dave, I have mounted my limits off of the newall DRO scales but not happy with them. Going to be printing some parts shortly to fix the issues and will post.

Davek0974
01-06-2016, 10:40 AM
Ok, I am only using the Mach DRO's - my Sino DRO kit will be sold to raise funds, can't really see the need to have two DRO systems.

komatias
01-06-2016, 10:56 AM
Yes, with the CSMio you can get away from the extra DRO. I have it becuase I am using steppers and a smoothstepper but also have a very cunning plan there too.

What kind of switches are you planning on using?

Davek0974
01-06-2016, 11:00 AM
What kind of switches are you planning on using?

I have no idea yet but they will need to be very heavy duty to put up with the mess.

lucan07
01-06-2016, 11:45 AM
These are good quality strong, adjustable and waterproof and come in a variety of designs

http://www.aliexpress.com/item/Waterproof-ME-8108-Momentary-Rotary-AC-Limit-Switch-For-CNC-Mill-Laser-Plasma-47942/1707482221.html

Davek0974
01-06-2016, 12:08 PM
Yes, that would my choice, there are a lot of head options - roller, plunger, lever etc with that switch type.

Need to choose where to mount them sensibly too.

Davek0974
02-06-2016, 08:03 PM
Got the knee back on tonight, new lift bearing makes it silky smooth after years of grinding up an down...

18542

Anyone know how to adjust gibs properly? The books just say tighten until you feel resistance but on the knee that could be very tight?

The vise-grips are only there to crimp off the feed to the saddle so i can prime the way lube pipes ;)

m_c
02-06-2016, 10:23 PM
Dial/DTI gauge, magnetic stand, and a big lever.

You don't want to adjust it up to the point there's no measurable play as you risk it binding and no room being left for oil, but you also don't want it flapping about.
When adjusting slides that I can't try moving by hand to test resistance, I usually aim for around 0.02mm of movement on a DTI. On older machines, you may need to allow for worn areas, so you ideally want to adjust things at the least worn part of the travel.

Davek0974
02-06-2016, 10:29 PM
How is it measured though - the distance the measurement is taken over will vary the reading etc, is it measured up/down or left/right??

m_c
02-06-2016, 10:39 PM
You want to be measuring the lift in the slide (forward/backward movement on a knee), however you need to judge the best way to achieve that for each machine.

For a knee, I'd probably mount a DTI on the side of the body pushing against a feature/block mounted on the side of the knee, and try rocking the knee side to side, and pivot it up/down. You obviously don't want to be measuring the play in the knee nut/screw mounting, so you may need to try levering a few different ways to work out what will work best.

njhussey
03-06-2016, 12:34 PM
You can also get them from Denco Lubrication (part of the Bijur group of companies) in Hereford (or Cougar Lubrication if you're in the South of the UK), shame I wasn't working there still else I'm sure I could have helped out;)

Nice refurb log Dave, following with interest. I was looking at doing this for work but we're still undecided on retrofit or purpose built....funds will decide!

Davek0974
03-06-2016, 12:52 PM
Got the lube parts from Lubetec, very fast delivery.

Started sketching out my control circuits, change-over from main spindle to bolt-on aux spindle will be done via the respective Mach3 profiles.

18546
Luckily the CSMIO has two 0-10v outputs and stacks of digital I/O so i can connect both VFD's up and use mach to decide which motor to start or which limits to look at.

lucan07
03-06-2016, 12:57 PM
Not being funny (well perhaps just a little) but I would not let mach decide I would definitely tell it which ones to use, the thought of mach going eenie meenie minie mo doesn't inspire me with confidence.. Sorry couldnt resist that!

njhussey
03-06-2016, 01:02 PM
Got the lube parts from Lubetec, very fast delivery.

They're stockists for the Bijur range so should be :)

One of the (many) things I've still to do is finish off the control/safety circuits...I've forgotten what I was going to do so probably going to start from scratch again and try to get my head round the subject (I hate electrickery stuff!!) again :) The CSMIO products are good, I'm glad I went with them.

Davek0974
03-06-2016, 01:06 PM
Not being funny (well perhaps just a little) but I would not let mach decide I would definitely tell it which ones to use, the thought of mach going eenie meenie minie mo doesn't inspire me with confidence.. Sorry couldnt resist that!

Let me expand...

I will have two Mach profiles - std Bridgeport & engraving

In profile 1 I have say output 7 defined as active low and no other connection
In profile 2 I have say output 8 defined as active low and no other connection

On the CSMIO each one of these outputs controls a respective VFD these outputs, being active low will come on as soon as Mach has started up and will stay on regardless of Mach reset condition.

So to do engraving I would fit the aux spindle and tell Mach to use profile 2 and so on.

It doesn't actually decide what to do, it just chooses the correct VFD for the profile I click on.

Davek0974
03-06-2016, 01:10 PM
They're stockists for the Bijur range so should be :)

One of the (many) things I've still to do is finish off the control/safety circuits...I've forgotten what I was going to do so probably going to start from scratch again and try to get my head round the subject (I hate electrickery stuff!!) again :) The CSMIO products are good, I'm glad I went with them.

I love the electrickery stuff - I have been cussed many times on here because I try to go too far with system controls etc - I am used to using PLC's and pneumatics at work and it tends to spread to my stuff ;)

njhussey
03-06-2016, 01:21 PM
I love the electrickery stuff - I have been cussed many times on here because I try to go too far with system controls etc - I am used to using PLC's and pneumatics at work and it tends to spread to my stuff ;)

I'll come to you then for double checking my stuff then, I did some programming, PLC and PIC stuff on my degree....but that was 15 years ago and I've forgotten more than I knew...which wasn't much in the first place!!

Clive S
03-06-2016, 01:45 PM
I will have two Mach profiles - std Bridgeport & engraving
I am doing something similar but with linuxcnc and mesa combo 5i25 + 7i76 so that one profile controls the lathe and the other the mill with one control box

Davek0974
03-06-2016, 01:49 PM
Sounds good, two machines with one system does seem to make sense when most of the I/O is the same.

Davek0974
04-06-2016, 09:13 PM
Got the ram and knuckle on this morning, also stripped and painted the saddle and head.

18561

Also fitted a spacer here as the gear mesh was terrible, way too wide. For the first time in all the years i've owned her, the knee now glides up and down silently

18562

Lee Roberts
05-06-2016, 01:47 AM
Looking lovely Dave, looks so much nicer cleaned up and painted, should turn out to be a nice conversion/restoration.

.Me

Davek0974
05-06-2016, 04:35 PM
Got the head on, spindle in and new oil pipes plumbed in and tested.

18580

Also stripped and painted the lower drive housings and scraped out the wax that used to be grease ;)

lucan07
05-06-2016, 05:00 PM
Looking good Dave, are you fitting new ballscrews you ordered as you re assemble or are they fairly easy to access/fit without breaking down the whole thing again?

Davek0974
05-06-2016, 05:04 PM
Yes, i'm at the stage now where I can't progress until I get the screws. Hopefully they will be here this week before we leave on Thursday for a couple of weeks holiday.

Davek0974
06-06-2016, 08:53 PM
Ball-screws arrived today :)

Managed to start fitting the Y axis one but where the oil feed drilling is in the nut, the pipe hits the knee chip guards so it'll last about 5 minutes :(

I have emailed the suppliers as it looks wrong unless i have it fitted wrong of course but there is only one oil drilling.

Davek0974
07-06-2016, 06:34 PM
How smart am I :)

Not very it seems!

I ordered a ball-screw kit for a Bridgeport with a 9" x 48" table, however, after trying to fit the Y axis again tonight it turns out I have the extended knee model with a 12" Y travel not 9" like normal :livid: :livid:

Still, I've only had it 11 years, never noticed.

The suppliers are being good so far and are looking into getting the bigger version swapped out with my one while i am on holiday.

Aaaargh!

Also ordered up another three replacement bearings for the head - the old girl was in a worse condition than i thought.

Davek0974
08-06-2016, 10:03 AM
Bigger version screw is on the way :)

Note to remember - measure twice, order once ;)

komatias
08-06-2016, 10:07 AM
Dave,

out of interest: what paint did you use?

Davek0974
08-06-2016, 10:15 AM
It's from a UK company called Craftmaster Paints, it's their "Engine Enamel" range in Bridgeport Grey.

Its the same company I used for painting the traction engines - the best paint I've ever used :) Not cheap it cost 30 for the 1ltr tin inc postage but well worth it.

Davek0974
24-06-2016, 05:40 PM
Hi all, back again after the holiday :)


Wasted no time in getting out to the shop, started catching up on customer orders and did some work on the mill.


Seems I have ordered a screw set for the Z axis that is far too long - bum :thumbdown::thumbdown:
Will eBay that one for a few quid - its a 20/05 zero backlash unit 360mm long if anyone wants it.


Then i discovered the motor bearings are half-shot :sorrow::sorrow:, a bit of slack and grumble when spun. It will work ok but not sure having come this far if i should refit or replace with something else - options would be repair (expensive), fit a new vfd style motor with shaft extended (more expensive) or throw on a servo motor (a bit more expensive).

Having replaced nearly all the bearings it seems daft to refit the duff motor now - time to think here.

Boyan Silyavski
24-06-2016, 05:55 PM
How much could be that motor bearings? it seems cheaper to replace them, right? Aren't they some standard size?

Having done all that work til now, the logic is not to leave anything unrepaired.

Davek0974
24-06-2016, 06:26 PM
I have read some tales as its a US motor and a bit awkward to dismantle, to find bearings size i would have to strip down which risks disturbing old and fragile varnished wiring etc. Not a big issue but could end up in a trashed motor.

The extended/custom shaft makes replacements a total bugger - practically impossible to fit a new motor and re-use the Reeves vari-speed drive system as new motor shafts are far too short.

This was the reason Bridgeport motors are stupid expensive.

So, options - strip and attempt repair.

Fit a servo and dump the Reeves drive but retain back-gear ?

The budget for this job went silly a long time ago - i was not intending on replacing all the ball bearings etc and some of them are so it seems now that an extra 300 or so on a 3kw servo/drive is pretty irrelevant overall.

The servo would likely be the easiest option - just a new mount plate and two pulleys, the repair on the bearings could be the cheapest option but would still leave an inferior, power-sucking drive system on what is now a damn expensive machine.

I'll work some prices out.

Boyan Silyavski
24-06-2016, 06:53 PM
What about repairing it on place with graphite or bronze filled epoxy ,Moglice and others? Why dont you check if that's a possibility. It would be dirt cheap

Davek0974
24-06-2016, 06:55 PM
These would be ball bearing races so not repairable in that way.

Boyan Silyavski
24-06-2016, 07:57 PM
Ain't a ball screw a ball bearing that could rotate as high as 3000rpm?

Davek0974
24-06-2016, 08:08 PM
I'm losing the plot here Boyan, I'm talking about the bearings inside the 1-1/2Hp spindle drive motor are buggered. ;)

Davek0974
24-06-2016, 08:35 PM
Drop-plates for X & Y servo motors - 10mm, 12mm, 15mm thick ally??

m_c
24-06-2016, 09:43 PM
Drop-plates for X & Y servo motors - 10mm, 12mm, 15mm thick ally??
Thicker will be better, but it will depend on servo shaft length/pulley thickness as to what's practical. Some form of box section would be better/stiffer.

Regarding the motor bearings, unless the motor has some odd wiring/end cap configuration and/or you damage the windings trying to lever the end caps off, stripping it and replacing the bearings shouldn't be a problem.

Boyan Silyavski
24-06-2016, 10:00 PM
I'm losing the plot here Boyan, I'm talking about the bearings inside the 1-1/2Hp spindle drive motor are buggered. ;)


I was saying if it works for removing ball screw backslash and the ball screw is in fact a more complex ball bearing , its worth a try at the cheap price of it. How fast this motor moves? 1200rpm? Same as ball screw.


I see as a better idea changing to cheap Dc motor or better servo motor. And why not a cheap modern normal motor , disassemble, weld some shaft and machine it on place. That would be super simple. And a bracket adaptor

Davek0974
24-06-2016, 10:04 PM
I'm going to have a go at dismantling i think, try that first.

A 2.0kw AC servo would be best but a toss-up between money and easy-fit;)

Davek0974
25-06-2016, 02:25 PM
Well, I spent a good few hours trying to dismantle this motor while being very careful not to risk damage, I could not even get the drive disc off the long shaft, seized well on. There is also a pressed-on fan at the top to shift which would likely bend/break so i carefully replaced what i had done and left it.

So my choice is now to build the cabinet and wire for the VFD / motor OR throw one of these in...

http://www.aliexpress.com/store/product/NEW-1-8kw-AC-Servo-Motor-and-Drive-in-6N-M-1800W-3000RPM-with-Servo-Cable/314742_2037588952.html

I have asked the guy for a 2kw or 2.2kw 3000rpm unit but not had a reply yet, seems the bigger servos don't have the speed needed.

Fitting would be easy, but the only benefit i would really see is the easy and accurate speed control - I don't need/want rigid tapping although i think it would be useful for smaller taps maybe - what is needed to gain rigid tapping?? Is it just a setup option or do i need more CSMIO modules??

Davek0974
25-06-2016, 05:50 PM
Pulleys/belt - 44t HTD8 x 30mm should handle it ok.

m_c
26-06-2016, 11:30 AM
You'd need an ENC module if you wanted to do rigid tapping.

The motor will most likely run for a long time before the bearings get bad enough to cause issues. If you could get some fresh grease into them, it would most likely make them last even longer and quieten them down.

Davek0974
26-06-2016, 11:57 AM
I guess if i do fit the servo, adding the enc module is a no-brainer really, daft not to fit it.

The bearings are sealed and hidden completely in this motor, i could likely get some WD40 in there but no grease. My main issues with fitting it are that it will fail some time, probably when i need it, and will then need my cabinet pulled apart to change from VFD to serve drive etc.

It sort of becomes a "might as well fix it now" option i think, then i can benefit immediately from the improved speed control etc. If i sell my tapping heads they would likely fetch enough to pay for the servo drive anyway :)

Davek0974
26-06-2016, 03:09 PM
Decided to exercise my minimal CAD/CAM knowledge and plunge into a slab of 15mm ally for the Y axis motor plate.

18792

Took 44 minutes on the mini-mill, used a single 5mm tool for the whole job and pocketed all the waste out ;)

Fitted perfectly...

18793

Seemed a better place than hanging down below I think.

Davek0974
27-06-2016, 07:29 PM
Decision is now leaning towards throwing the old motor back on!
Why?
One reason was that i don't have any idea if a 1.8kw servo will be a good replacement and 1.8kw seems to be the only (affordable) size that runs to 3000rpm, also, fitting an encoder for the rigid tapping module would be a total ball-ache on a Bridgeport - there is no clear access to the spindle shaft and I wrongly thought it would use the motor encoder signals which it does not.

So, back to plan 1 until the bearings fail 100% is to use the standard motor with my VFD and use the variable mechanical speed to give me two electrically variable speed ranges that overlap.

Boyan Silyavski
27-06-2016, 08:16 PM
But your motor now has half the rating of this servo RPM right? 1.8kw servo has much more usable power in all revolutions than a standard motor. And higher momentary power.

I think you are mistaken a bit. I don't understand sth. The servo has an encoder. why you need a separate encoder for the rigid tapping? The servo drive surely has output of signals so what additional encoder? Read servo drive manual and you will see. My servos for example have encoder outputs and much more. Check manual.




It will be difficult for the bearing to fail 100%. It will make noise and so but i doubt it would fail if it has some grease in it.

Davek0974
27-06-2016, 08:26 PM
Its all to do with power and torque, from another forum on this setup.....

Quote..

Well the 1.5Hp induction motor with variable mechanical drive would have (basing torque on 1.2Kw)

2880RPM 0.8kW and 3.94 Nm

1440RPM 0.8Kw and 7.9 Nm

720 RPM 0.8Kw and 15.78Nm

The 1.8kW Servo would have,

3000RPM 1.8kW and 5.7Nm

1500RPM 0.9Kw and 5.7 Nm

750 RPM 0.45Kw and 5.7Nm


As you can see, I would be losing a lot of power in the lower ranges but retaining constant torque, what is needed is both torque AND power.

It can be done but i would need to fit a 3.8kW servo to match the existing motor without losing performance, this is too expensive and they only run to 2500rpm.

I have just been advised that the encoder module CAN use the motor encoder ok, as long as the motor and spindle are at 1:1 ratio.

Will refit the original motor.

Davek0974
29-06-2016, 07:44 PM
Finally got the replacement Y axis screw :rolleyes: Got it fitted and got the table back on at last...
18822

Not quite plug-n-play though - the new end bracket does not fit properly and will need machining, will have to get this done elsewhere as my mill not usable ;)
1882318824
Can't simply file the holes longer as the counterbores need moving too. I had to get the new bracket as the original machine had power-feeds.

Davek0974
01-07-2016, 08:40 PM
Found a local company to move the bores and counter-bores on the Y axis bracket - so annoying having to get stuff done outside when it would have taken me about 15mins to do, I had to leave it most of the day, make two 30min trips and stump up 40 :(

Anyway, done now and fits nicely, also mounted the motor plate using some spacers to pick up on the bottom of the counter-bores at the top face is not flat. Had to make a shim for the bearings to give a bit of pre-load but not bind up. Next job is to turn up a spacer tube for the pulley-to-bearing gap and then fit the belt - one axis finished.

Will also be planning limit switches this weekend, metal has not arrived so can't make the X axis plate yet :(

John S
01-07-2016, 09:05 PM
Still got this step speed head unit with a brand new 3Hp 1425 rev motor and conversion plate, you just need a VFD to run it.
300 to get it out of my way ?

Davek0974
01-07-2016, 09:10 PM
Hang on to it a little longer, lets see if this one will survive, if not i'll be in touch ;)

Davek0974
02-07-2016, 04:25 PM
Got X-axis motor plate fully fitted...

18849

Limit switches next, Y-Axis Limits and home switches...

18848

From the left we have Y--, Y-Home, Y++.
The home switch remains tripped once it trips so that i can home even if the switch is triggered, a home-offset value will take care of the switch position.

Then the X-axis, a bit harder this one, first idea...

18847

From the right we have X-home then a dual-direction X-- & X++ switch. The picture is not quite right - imagine the right-hand switch is another roller type like the left one, I have one somewhere, just got to find it ;)

The only down-side I can see is that it blocks the chances of using flood coolant again as the switches are in the way of the left drain hole.

Still thinking on this one.

Davek0974
03-07-2016, 06:59 PM
I hooked the rebuilt head up to my VFD today, sounds a hell of a lot better, for some unknown reason it runs a bit warm around the top bearing plate though.

Its not quite right though yet as the vari-speed system seems to do nothing in the higher end when in back-gear, but is ok in normal gear??

Davek0974
06-07-2016, 08:12 PM
Nearly got the x-axis motor fitted, once that's on I can lash-up the drive and controller and do motion testing :)

18867

Davek0974
10-07-2016, 08:10 PM
We have motion :witless: In 2 axes at least. Done some basic servo tuning, cut the top speed to 5000mm/min as 7500 was just too scary. Got homing and soft limits working nicely. Homing is silly-accurate with index-homing turned on - about +/1 0.01mm or better as far as i can measure.

Next I fitted up the Spindle VFD, expecting serious issues with my complex manual/auto speed control idea. I was totally amazed that I have perfect control of speed from 40rpm to 3000rpm within about 25rpm tolerance! It tells me to choose back-gear or high-gear and high or low speed-pulley setting depending on speed chosen by code.

The only issue it does have is that if it is set for say 100rpm in back-gear low range, and the code calls for 2000rpm in high-gear high range, it will spin up the spindle and then tell me to change gear - this cannot be done without smashing the drive.

I have all the code in SpindleSpeed.m1s - can it also be made to stop the spindle and restart it after pressing enter without wrecking the code-flow etc ??

My bench-top lash-up, well it's not a bench but the plasma-cutter but here you go :adoration:...

18871

Going to need a pretty big cabinet.

JAZZCNC
11-07-2016, 02:21 PM
I have all the code in SpindleSpeed.m1s - can it also be made to stop the spindle and restart it after pressing enter without wrecking the code-flow etc ?

Can't see any reason why not. Post the file your using and I'll take look.

Davek0974
11-07-2016, 03:39 PM
I had a second thought last night - it's not as simple as i thought as not only does it need to stop the spindle and restart it, it has to restart it in the same direction as it was running previously because when in low range/back-gear the motor has to run in reverse.

Not so sure its possible now??

I'll get the code up though soon.

Thanks

JAZZCNC
11-07-2016, 03:48 PM
Not so sure its possible now??

Anythings possible just changes how hard it is to achieve.!

Explain a little about how it's working now.? How do you run motor direction now directly via VFD or M3/M4.

Davek0974
11-07-2016, 03:54 PM
At present, it runs via VFD and 0-10v signal from CSMIO, the motor has back-gear for speeds below 400rpm and also has a variable speed mechanical control that i am using in conjunction with the VFD to enable maximum torque to be obtained at all speeds. The VFD runs from 25Hz to 75Hz so keeps the motor in the power band.

In SpindleSpeed.m1s macro i have some code that examines the requested speed and compares the current pulley range - if it is out of range it warns me to set high/low speed on the mechanical dial and possibly set the back-gear or release it if needed.

Mach is set up with 4 pulleys.

I will go get the macro and pulley data.....

Davek0974
11-07-2016, 04:02 PM
Here we go...

Spindlespeed macro
18872

Pulleys are set in Mach as follows...
1 - 40 to 126rpm & motor reversed
2 - 127 to 366rpm & motor reversed
3 - 367 to 1031rpm
4 - 1032 to 3000rpm


So 1&2 are back-gear and 3&4 are high-gear.

I was amazed at how closely the speed was set via Mach, works really well.

Davek0974
14-07-2016, 04:41 PM
VFD control ?

I currently have my fwd and rev signals wired through two relays using the built-in VFD control supply to control itself.

Is this worth it or should I bin the relays and connect the VFD ground to the CSMIO control ground and go direct from CSMIO output to VFD input?

Its a little layer of isolation but also mechanical complexity if not needed.


Safety relays....
The CSMIO manual shows a PILZ safety unit being used, I gather others are just as good? PILZ are damn expensive but other like Telemechanique items are better priced.

I gather they are used because they monitor the main control contactor/relay and also offer forced contact seperation etc internally, any other reasons? Any types to look out for?

Davek0974
16-07-2016, 05:07 PM
Z-axis is coming along nicely - got all the parts machined on the mini-mill ;)

Last step is to make the link from the nut to quill.

My first idea is to plasma-cut a part from 10 or 12mm steel plate, then fit it up and weld it to original bridgeport part. This would make assembly easy as i can fit the parts up on the mill then clamp together and remove to make the weld.

Sound like a plan?
18883

18882

komatias
17-07-2016, 10:02 PM
Hi Dave,

I have taken photos as part of my advert so thought I would share just to give you ideas.

Started off with a plate that I flattened and cut out holes to allow access to the tramming bolts and the quill front. Used the two steps on the front of the head as a means of alignment but found they were about 0.5mm off plain. So I have shimmed the one of the flats.

Behind the ballnut you will make out a socket cap screw that engages with the actual casting where I drilled and tapped an M8. The box above I made using some scrap piece of 6mm plate and practiced my aluminium welding.

https://lh3.googleusercontent.com/-9lOnLj-6Oe4/V4vgGS1jfpI/AAAAAAAAI5w/Hhee1ofBPk8_JGNOZ8OIqmgDWD4N7n6GwCCo/s640/IMG_20160717_203248.jpg


The ball nut mount is made up of two parts that sort of dove tail into each other. This part is made of mild steel and I was well chuffed to get the clearances between the parts to 0.05mm. The two parts are held together by the M8 socket cap just behind the ballnut. I could thus home the Z, undo the screw and use the quill manually.

https://lh3.googleusercontent.com/-PTIYA1i49HU/V4vgGSNAhNI/AAAAAAAAI5w/a2p541pPsS4IUNPOraQ4vpeXFu_3JmOCgCCo/s640/IMG_20160717_203301.jpg


https://lh3.googleusercontent.com/-3Nm4DazWrIs/V4vgGcysxeI/AAAAAAAAI5w/TeDWhTui5ecVXmSQooxZw9uBUjMdQnHOQCCo/s640/IMG_20160717_203526.jpg

The mounting plate is secured on the bottom through the old feed hole in the front using a top hat boss which allows me to not lose too much quill travel.

https://lh3.googleusercontent.com/-R4aUe052fPA/V4vgGe00rjI/AAAAAAAAI5w/PHjTz2peBQgOjzRFgKjpnBu_gNuynZj8wCCo/s640/IMG_20160717_203723.jpg
https://lh3.googleusercontent.com/--23abQq8L4A/V4vgGS4PuNI/AAAAAAAAI5w/aaV5xnLj-_QSdW-txWWN4Csd6lHDQtrxQCCo/s640/IMG_20160717_203729.jpg

On the side, since the quill powerfeed is not used any more, I have used the holes to mount a side brace. If you notice there is a small M4 screw just under the elbow of the pressure regulator, this is to tension the belt of the motor. I have built this kind of tensioner in all the axes of my machine.
https://lh3.googleusercontent.com/-PYbTU8yTrNw/V4uKZV5CQTI/AAAAAAAAI4M/skahgnLnN_oAF9Zoq6ksk6wYIxP1xnQCgCCo/s640/DSC_0884.JPG


Yours is bound to look prettier of course, hope you got some ideas though.

Davek0974
17-07-2016, 10:14 PM
Its all about function not pretty :) And that looks nicely functional, thanks for sharing.

20mm screw?

Air collet closer too?

komatias
17-07-2016, 10:21 PM
Thanks! It is pneumatic impact wrench type drawbar in a nice package. There are plans for one on the net too. You press the green button a bit and it unscrews a specially ground R8 collet.



You may have seen I use (and sell) the repeatable Z offset ER collet chucks which makes life infinitely easier when running multiple toolchanges. Measure the tools beforehand, enter them into the tool table and swap them out when needed. plenty of videos of them on youtube (search TTS tooling).

R's

Screw is 1605

Davek0974
17-07-2016, 10:49 PM
TTS looks useful, will look into that.

Davek0974
19-07-2016, 07:32 PM
Threw some panel buttons into the mix today, also fitted LED's for little bling ;)

Video here...

https://youtu.be/5Yz0CloE7TA

JAZZCNC
19-07-2016, 07:52 PM
Why do the other lights come on when press cycle start.? Presume they are Feed hold and Stop.!

Are you turning lights on using Outputs and Macro pump.?

Davek0974
19-07-2016, 08:03 PM
Yes my thoughts were that when idle the Cycle start and Spindle Toggle would be active and when active the Feed-Hold and stop would be active. The logic is that the lights indicate rapidly what a possible action would be.

I could go one step further and use the lamp supply as the switch source so only the lit buttons would do anything but i think Mach has that pretty much covered anyway.

Yes, I have a small bit of code in Macropump.m1s and an output feeding a relay, the lamps run from the N/c and N/o contacts on that relay.

The macropump code checks two omelette's for run or feed-hold.

Edit..
Frigging autocorrect - that last line should read....
The macropump code checks two oemled's for run or feed-hold.

JAZZCNC
19-07-2016, 10:26 PM
I do similair thing with the control panles on my routers but it just checks Mach led's and turns on/off outputs accordingly to whats active. I use 24V led's which the Outputs turn on/off directly.

18900

Davek0974
19-07-2016, 11:03 PM
Nice,

I'm only using the relay as the two lamp patterns are directly opposite each other so i just use both N/o and N/c on the relay. LEDs are 24v but this seemed easier and only used one output;)

Davek0974
21-07-2016, 08:21 PM
The Z-axis is doing my head in :concern:

My idea of welding it together is doomed to failure I think - the original bridgeport quill stop is porous and soaked in oil, any attempt to weld it simply creates a pool of oil and crap welds. I managed to tack it up but that only got me to my next headache....

Alignment is being a bitch - I can't seem to find a position where all the parts just play nicely, there is a fair amount of stress and strain between the screw and the quill, it can be moved up and down by hand but I'm not happy with it from an engineering point of view.

Not sure how to sort this at the moment, a right PITA.

I might just bolt the two parts together and see how it goes, that will allow me to shim things here and there for alignment, seeing the one from Komatias above being held by an M8 screw makes me think this might work.

Davek0974
21-07-2016, 11:24 PM
Ok, sorted it i think :chuncky:

Couldn't rest without a fix so a couple of hours fiddling and a few thin shims later we have smooth motion from top to bottom, its pretty easy to reverse-drive it by moving the quill lever :yahoo:

The bracket that holds the nut was slightly twisted and the nut hole was too tight, fixing both of those cured the problem.

Now i need to fix up some limit switches and get the motor hooked up.

18920

Davek0974
23-07-2016, 06:06 PM
Got the switches fitted and have finally achieved 3-axis motion :)

18924

A short video showing basic path with two tool-changes...


https://youtu.be/mJUxtS997MQ

Davek0974
24-07-2016, 10:12 PM
Scary stuff - she has cut her first steel :stung:

Spent the morning putting stuff back - trammed the head, fitted the vise and trammed that in, fitted a floating Z touch-plate and modified the touch-off macro so it did what i wanted.

Threw a small lump of 20mm scrap in, a duff 1/2" four-flute cutter and programmed a facing cut in Aspire of 0.5mm, seemed way slower than i would have moved if doing it manually - may not be a bad thing. It gave me a speed of 140mm/min at 650rpm

Picture below - looks worse than it is, pretty good finish as it goes considering the state of the tool etc. and dry cutting plus unknown grade of steel.

18928

Next I tried spot-drilling four places and followed up with through-drilling with a 6mm drill, all went perfectly.

I really need to sort the speed control out though - it tells me when to put it in back-gear or high/low speed so the VFD can give me the final speed BUT it only does this when it gets the S command from the code - it shows the message on screen but the carries on cutting at the wrong speed. What i need is for it to stop the spindle and show the message, wait for me to set the right conditions and the restart the spindle and get to work.

Boyan Silyavski
26-07-2016, 06:09 PM
All looks fine now. So whats next? Mist cooling, right?

Davek0974
26-07-2016, 08:27 PM
Ha, I already have that built, it's fitted on the mini-mill.

The next step is to weld up a suitable control cabinet, this is needing a big cab - about 850 tall x 700 wide and 270 deep, commercial ones are +250 in that size so will plasma cut and weld up from 3mm sheet steel parts i think.

Then i have to strip the rats-nest down and start the final build.

I've been running various cutters through HSMadvisor and it seems i've been pushing them quite hard when doing it manually - all the feeds seem very relaxed ;) I tended to run the HSS so the chips were straw coloured and the carbide insert ones just flat out and hard - dark blue chips that melted your shirt ;)

Davek0974
31-07-2016, 05:22 PM
Got the control cabinet built today, absolute monster and a two-person lift :)

18967

Fitted the mounting bars too.

18966

Next job is to strip the test setup down and start final fitting.

Clive S
31-07-2016, 05:56 PM
Got the control cabinet built today, absolute monster and a two-person lift :)Well it will be a three person lift when you fill it:joker:

Davek0974
31-07-2016, 07:13 PM
your'e not wrong there, i have all the guts temporarily mounted on a pallet top and i can just about lift it all, not quite sure how i'll lift the steel panel into the cabinet yet;)

Davek0974
05-08-2016, 01:57 PM
Back to my dual-spindle attachment, yes its been done before, found two people that have done a great job and turned out some amazing stuff on them.

I was looking at a single thick clamp ,made from maybe 25-30mm alu, but this idea seems far better....

18989

It uses a two-layer clamp, bridged in the middle firmly and then clamped on the quill plus clamped to a fixture held in a collet in the spindle. This seems like it would be very stable plus use up less of my valuable Z axis travel.

Material is probably only 15-20mm thick so i would lose no more than 20mm of travel in Z, leaving me a healthy 100mm to play with. The wide spaced clamps should also grip the high speed spindle better too.

Not sure why he has made the bottom clamp like it is at the spindle nose, I would go with a solid area and a 20mm shaft fitted for a 20mm R8 collet - the attachment would then be slid upwards into the collet and onto the quill.

komatias
05-08-2016, 04:01 PM
Why not just get a quillmaster or a speeder? Somthing that is inline with the spindle in the firstplace.

Davek0974
05-08-2016, 04:08 PM
Why not just get a quillmaster or a speeder? Somthing that is inline with the spindle in the firstplace.

A quill master only gives 2:1 speed up so would give me 6000rpm, no idea of what a speeder is?

My high speed spindle gets me to 24,000 rpm, still low but very usable on aluminium.

Davek0974
06-08-2016, 06:15 PM
Some more progress...

Got the cabinet mounted on the side of the mill...

18992

Also cut and welded up the pendant workstation and trial-fitted it on its suspension arm...

18991

Not all the holes will have buttons but thought it best to cut spares and fit blanking plugs.

Slightly concerned about the strain on the 10mm stud holding the arm on the rear of the ram but it seems ok.

komatias
06-08-2016, 08:52 PM
Dave,

truly inspiring!

You thinking of making a flood enclosure for the mill? It made my life so much better when I did.

Davek0974
06-08-2016, 08:58 PM
I think it will need a chip control enclosure but I'm hoping to run mist coolant only.

Davek0974
07-08-2016, 04:00 PM
Started panel wiring today, one of my favourite parts of a build :)


Got the grounds and most of the neutrals in, plus the motor supply from the VFD to the terminal bar.

18993

The space at top-right is for the high-speed spindle VFD - can't fit that yet as its still on the mini-mill.

John S
07-08-2016, 04:53 PM
You are not putting the output from the VFD onto 3 separate switches are you ?
If one trips you will fry the VFD.
VFD has to go to motor with no breaks.

Davek0974
07-08-2016, 04:59 PM
No, no, no ;)

The output from the vfd goes direct to the terminal bar and then to the motor.

The three contactors are:- VFD 1 supply, VFD 2 supply and Servo's supply.

The VFD's are for main motor spindle and aux high-speed spindle, only one will run at any time.

I think having a terminal bar in between motor and VFD is acceptable.

John S
07-08-2016, 05:09 PM
Yes that's fine Dave, hard to tell what someone envisages just from a picture

Davek0974
07-08-2016, 05:18 PM
Of course, no worries ;)

A rough idea of the safety/control side of the wiring...
18994

John S
07-08-2016, 05:31 PM
Sir,

Sir,
Query please?

You show pin 14 from the XPS as going to relay 1, relay 2 and the coil of K3 but you are also showing K3 across Y1 and Y2 ?

Davek0974
07-08-2016, 05:46 PM
K3 controls power to the servo drives, the N/C contact on the XPS monitors the K3 unit, if it is jammed or fails to open, the XPS does not pull in and Mach will stay in reset as well as no spindle relay/contactor supply.

If K3 jams on despite pressing the E-stop or limits, it will not latch-in again.

It was a reasonable safety enhancement i think.

As with all systems, a catastrophic failure - in this case K3 welding shut on one or more lines will still leave a danger but the chances of a contact weld at these low currents is pretty rare i think?

m_c
07-08-2016, 08:19 PM
You do realise that a standard contactor is designed so that if one contact welds shut, the others will still open?

Davek0974
07-08-2016, 08:33 PM
Interesting, would the carrier still lift the aux contact block though? On these Sirius range of contactors, the monitoring/feedback contacts are by way of an aux contact block mounted on top.

The XPS will not work without an N/C pair on one of the output channels.

m_c
07-08-2016, 08:51 PM
I never realised the XPS was through a NC contact.
It'll depend on the spec. Some will be designed so the contact closes regardless, others will hold it open. On larger contactors, you can normally wedge a small screwdriver in and close a single contact, which would then let you see if the NC contact opens or not, however smaller contactors are usually far more enclosed.

Having said that, at the currents you're working with, the contactor should never weld shut, so I wouldn't worry about it. Even if it was to stick and maintain full power, the power being maintained should not be the sole reason if the machine wasn't to stop during an e-stop.

Davek0974
07-08-2016, 09:00 PM
I just like having things covered as far as practically possible - in this case if the e-stop is pressed, the CSMIO is told to stop as well as power being removed from all servo's and VFD's, both via separate channels. I think that should be ok, the XPS contacts are forced apart so pretty much guaranteed to open.

BTW, what happens if a servo encoder fails - does the motor runaway or stop??

m_c
07-08-2016, 09:05 PM
On an AC/Brushless motor, the drive should detect a fault.
With a brushed DC motor, potentially run it into the limits at great speed.

John S
07-08-2016, 09:22 PM
BTW, what happens if a servo encoder fails - does the motor runaway or stop??

You need clean underpants - don't ask...........................

Davek0974
13-08-2016, 04:28 PM
Got the panel finished, bench-tested, all passed 100% :)

Got the pendant rough-wired so i could bench-test the panel, seems functional.

All axes homed and jogged perfectly, no surprises, e-stop system seems to work ok.

Managed to lug the panel into the cabinet, nearly ruptured myself but it's in now and bolted down.

Tomorrow will be final wiring on the switches and limits etc, plus re-rigging the pendant for the next power-up test. Then I'll probably strip the pendant and send it out for powder-coating, or just spray it if impatience takes over ;)

19009

19010

Clive S
13-08-2016, 06:56 PM
Nice job Dave:encouragement:

Davek0974
14-08-2016, 09:55 PM
Got the panel wired to the machine today, most of the big jobs done now, its running again, ready for testing.

19037
19036
19038

Small jobs to do like secure some of the cables when i work out where they need to be, make the belt covers, cover the Z axis drive, fit the light, then figure out how to run her ;)

Lee Roberts
14-08-2016, 11:59 PM
Bravo!

.Me

Davek0974
20-08-2016, 07:47 PM
Fitted some chip covers for the ways, nitrile rubber, tomorrow will be making the belt covers for X&Y drives.

Lowest and most forward position...
19054

The front cover will just hang over the front like many of the real CNC Bridgeports do.

Davek0974
22-08-2016, 10:18 PM
Got the X&Y axis chip covers fitted today...

1906419065

Also found this on the welding bench....

19063

Should help a bit :)

Boyan Silyavski
23-08-2016, 10:06 AM
Also found this on the welding bench....

19063

Should help a bit :)


Its good to have a plasma machine at workshop

Davek0974
28-08-2016, 01:19 PM
Made some button tags today, i've also had the console powder-coated :)

19095

Coming along nicely

Clive S
28-08-2016, 03:42 PM
Very nice Dave and pro.:encouragement:

routercnc
28-08-2016, 05:45 PM
Looking great. Did you buy the blank tag material already made up or did you have to stick layers together? Can you let me know what the material is and where you got it? Thanks

Davek0974
28-08-2016, 07:34 PM
I got it from here https://hindleys.com/index.php/ Its 1.5mm black-white-black engraving plastic, 600x300 sheet, 14 delivered.

First time i've used it and came out perfect, used a 30deg engraving tool, 0.2mm tip, 24000rpm, 900mm/min, air blast only and used an Aspire V-Carve path. Outer edges were cut with a 2mm single flute carbide bit with 6mm x 0.6mm tabs.

routercnc
29-08-2016, 03:31 PM
I got it from here https://hindleys.com/index.php/ Its 1.5mm black-white-black engraving plastic, 600x300 sheet, 14 delivered.

First time i've used it and came out perfect, used a 30deg engraving tool, 0.2mm tip, 24000rpm, 900mm/min, air blast only and used an Aspire V-Carve path. Outer edges were cut with a 2mm single flute carbide bit with 6mm x 0.6mm tabs.

Thanks Dave one project to add to my list !

Neale
29-08-2016, 04:48 PM
I'm going to need some of that plastic soon - good to see a source of smaller pieces at not-too-unreasonable prices. An 8x4 sheet would last me quite a long time...

Clive S
29-08-2016, 06:28 PM
I got it from here https://hindleys.com/index.php/ Its 1.5mm black-white-black engraving plastic, 600x300 sheet, 14 delivered.

First time i've used it and came out perfect, used a 30deg engraving tool, 0.2mm tip, 24000rpm, 900mm/min, air blast only and used an Aspire V-Carve path. Outer edges were cut with a 2mm single flute carbide bit with 6mm x 0.6mm tabs.Dave very nice, what font did you use. IN fact could you post the dxf up.:barbershop_quartet_

Davek0974
29-08-2016, 06:49 PM
DXF attached for a square button tag, the font is Arial BTW

19102

Davek0974
01-09-2016, 10:30 PM
Got the panel finished, just need to find a few blanking plugs :thumsup:

19119

JAZZCNC
01-09-2016, 11:24 PM
Got the panel finished, just need to find a few blanking plugs :thumsup:

19119

Ah ah just bought 2 of those KB of flee bay.

Clive S
01-09-2016, 11:27 PM
Very nice again Dave. How come you did not put some grad marks on the feed and speed:congratulatory:

I have made a few for mine but cut them out just with the 60' V cutter. Saves changing tools.

Did you use Bold Arial

Davek0974
02-09-2016, 01:23 PM
:)
Grad marks would be meaningless i thought as its based on the screen display not any real position so i didn't bother, i can always redo them i guess one day ;)

Yes it was Arial Bold and a v-carve path

Davek0974
05-09-2016, 01:35 PM
Anyone have any views on what sort of speed and acceleration i should be aiming for?

Bearing in mind it has a relatively low spindle speed, even in aluminium, the feed rates are pretty low, I don't think it's worth going much above 2500mm/min?

I have read that there is a point in acceleration where Mach starts rounding corners etc, is this a known point or just something that can happen at some setting value?

Its currently set for 1000mm/min but that seems quite harsh when you do a rapid, it makes the machine shake a little at start/stop of the move.

Boyan Silyavski
05-09-2016, 05:10 PM
Judging from power servo feeds made for that machine, ~900mm/min is the maximum. i think a machine like that could even reach a maximum of 5000mm/min but with slow acceleration. So if servo motors could do the trick of moving it fast, its up to you to decide the balance between speed and acceleration, depending on what you would be doing.

I have to ask Jeff who has a servo driven similar machine, but as far as i remember once he told me, he doesn't push it more than 500mm/min ever.

So no point of making it run away from your workshop if not needed. I would leave it like that / ~900mm/min / and with normal acceleration. Then with the time you know it better you could fine tune it.

Whats your acceleration set at?

Davek0974
05-09-2016, 06:31 PM
It needs to be a bit faster, i have tool paths that run 1150mm/min, i think 2500 may be a good compromise. The servos themselves will easily push it to 7500mm/min if i let them but it is very scary.

I will try my usual 1/10 rule - 2500mm/min and 250 acceleration and see how it goes i think.

Boyan Silyavski
05-09-2016, 08:10 PM
250 acceleration is not much, so even if scary at first, i think 2500mm/min will not be a problem. Dave, i have set 3000 acceleration on my machine, now that's scary, when you see 200kg gantry flying around with that speed.

Davek0974
05-09-2016, 09:59 PM
Ok set speed for 2500 and acceleration for 400, moving right along - servo tuning.

Have done nothing on these apart from an auto-tune in CSMIO setup. I thought the results were ok, but having spent a few hours today scouring the 'net it seems they were pretty bad really. The first video here shows one axis after an auto-tune. Now, I have zero experience with servo's, being stepper builds up to this build, so I stand to be fully corrected but, i think the graph shows pretty bad over-shoot or oscillation?


https://youtu.be/IOAS4LMJzcg

So, throwing caution to the wind, i dived in and removed all of the "I" and "D" parameters as well as the "ff" or fudge-factor i think its called. That seemed to improve things so i started increasing the gain in steps of 1,000 and the graph seems better to my untrained eye?

The following error is lower, dead-band is reached faster and the dials on the machine no longer overshoot and slowly wind back to a stop - this was a concern as i could see it affecting cut quality etc.

This video was shot after my messing around...


https://youtu.be/Yr4joMqo61Q

Now, please feel free to rain on my parade, but does the last video look happier than the first?
Is there any issue running with zero "I". "D" & "ff" values?
I can't hear any grumbling from the motors, there is no visible 'hunting' at idle, following errors are down and the dials now stop on a dime as shown.

The Z axis would not tolerate as much gain and did go into permanent oscillation once so i backed off a fair bit but still no "I", "D" & "ff" values.

Suggestions?

Davek0974
09-09-2016, 08:24 AM
Some short videos of the conversion in operation, 15mm Alu plate, full recommended speeds and feeds in use...

2d pocket clearing...

https://youtu.be/gpFq3sJfYVc

2d Adaptive clearing...

https://youtu.be/FBWLnouPMiA
I have massively increased the non-cut return speed since this video was taken :)

So far its working well.

njhussey
09-09-2016, 09:11 AM
Looking good Dave, always good to see chips flying!!!

Davek0974
09-09-2016, 09:35 AM
Thanks, its good to watch:)

USB Issues?

I need USB in the control pendant, USB keyboard/trackball and USB socket for code loading.
Currently i have a cheap 4-way hub and extension but windows is not happy and i get multiple "new device found" chimes every time i start up or plug in a USB stick

Is it best to try a powered hub or run separate extensions to the PC??

The length needs to be 5m to reach.

Clive S
09-09-2016, 10:06 AM
Dave can you not transfer files over Ethernet

Davek0974
09-09-2016, 10:18 AM
I don't really transfer, the code runs direct from the stick and the rest is just keyboard/trackball

Davek0974
10-09-2016, 04:02 PM
USB Issues solved - it works well without the hub and just having two extensions, i have ordered another extension and some clip-on ferrite blocks to suppress any interference.

Also, at last - the adaptor I have mucking about with for the last couple of months is finally done :tranquillity: Admittedly it took so long as i was building the machine to make it and figuring out how to operate it but in the end, after ironing out all (so far) of the bugs in tooling, offsets, post-processing etc it really only took a couple of hours.

It's also the first part designed in Fusion 360, that was a learning curve!

It fits the mill snugly and as long as the high-speed spindle motor fits ok, should work beautifully, still got some jobs to clear before i can strip that down though.

Total cost in Z axis travel is 22mm - not bad i thought - that leaves me about 95mm to play with but on a small spindle with tiny tooling that should be ok i think.

Thanks for all help guys :wink::wink:

19170

19169

19168

Davek0974
12-09-2016, 10:51 AM
I have a tapping head, TapMatic SPD5 - http://www.tapmatic.co.uk/products/p_12.php

It's a self-reversing unit in mint condition.

But it was for a manual machine - you feed in to a quill depth stop, the thread pulls the tap down which disengages the clutch then you pull up which engages reverse and the tap screws out. You need to pull up slightly more than the thread pitch in order to keep reverse engaged.

Now, I don't have rigid tapping, probably never will on this machine, so is it possible to create the feeds/retracts to use this device in CNC or not?

Or, other options?

komatias
12-09-2016, 11:06 AM
I have a tapping head, TapMatic SPD5 - http://www.tapmatic.co.uk/products/p_12.php

It's a self-reversing unit in mint condition.

But it was for a manual machine - you feed in to a quill depth stop, the thread pulls the tap down which disengages the clutch then you pull up which engages reverse and the tap screws out. You need to pull up slightly more than the thread pitch in order to keep reverse engaged.

Now, I don't have rigid tapping, probably never will on this machine, so is it possible to create the feeds/retracts to use this device in CNC or not?

Or, other options?



Yes you can, but you need to figure out the depth you are going to end with. I had trouble with Vertex job.

Personally I prefer the tread milling approach. Quicker and cleaner.

Davek0974
12-09-2016, 11:13 AM
I have a lot of M6 holes to thread - can you thread-mill that small?

m_c
12-09-2016, 11:26 AM
I have a lot of M6 holes to thread - can you thread-mill that small?

Cutwel list cutters down to M3, but you might need to sit down when reading the price.
The most basic 1mm pitch cutter has a list price of 60.66 (Order code - L1211310)

komatias
12-09-2016, 11:27 AM
Yes it is possible.

I bought a single form one to do my grid of M12's from here: http://www.drill-service.co.uk/Product.asp?Parent=180380020000&Tool=799

You would use the 6mm one with a max cutting OD of 4.8 so you can get into the 5mm pilot holes.

If you want to go quicker, you can get the multipitch endmills that can make the thread in a single pass but they do become more expensive.

Have a look on NYCCNC's videos for how to set it all up in Fusion 360.

Regards

Davek0974
12-09-2016, 11:36 AM
Ouch! thats a bit spendy for a thread :)

Just found a Torch video and they have a similar head to my Tapmatic - theirs is Procunier but the action is the same it seems.

The only benefit they have is that its all pre-programmed canned cycle stuff. I don't have that luxury but at least it proves it can work.

I will have a look at the specs on the head for tap-travel below the stop point and see if i can come up with a method of programming :) From memory it was 5mm below the stop point to where the feed disengaged so for a 10mm deep thread you would stop feeding at 5mm, wait for the clutch to disengage then feed out.

Davek0974
12-09-2016, 11:59 AM
Hmm, thought i could get smart and do it in Fusion360 which comes up with this code...
(1)
(T1 D=6. CR=0. - ZMIN=-10. - RIGHT HAND TAP)
G90 G94 G91.1 G40 G49 G17
G21
G28 G91 Z0.
G90


(DRILL2)
M5
M9
T1 M6
S1000 M3
G54
M7
G0 X-5. Y0.
G43 Z15. H1
Z5.
M29 S1000
G98 G84 X-5. Y0. Z-10. R4. F1000.
G80
Z15.


M9
G28 G91 Z0.
M30

This should be a M6 thread 10mm deep with a 4 second pause, But it seems Mach3 does not recognise G84 so thats out, whatever G84 really is :)

Boyan Silyavski
12-09-2016, 12:01 PM
ebay , 1 (http://www.ebay.co.uk/itm/GROOVING-AND-THREADING-TOOL-MICRO-CARBIDE-approx-010-Width-1-4-Shank-/381136824268?hash=item58bd8667cc:g:ibYAAOSwa39UxXq k) and 2 (http://www.ebay.co.uk/itm/GROOVING-AND-THREADING-TOOL-MICRO-CARBIDE-approx-012-Width-1-4-Shank-/381136836082?hash=item58bd8695f2:g:woYAAOSwk5FUxYE z) . I have been looking to grab a couple . 5$ instead of 50$. What you say? I think these are ideal for various inside threads

John S
12-09-2016, 12:26 PM
G84 is rigid tap cycle.

You should have bought one of these.

http://www.stevenson-engineers.co.uk/files/4axis%20mill.jpg


450 landed in UK.

No computer, no monitor, no windows license, no Mach licence, No CSIO box of tricks.

Does rigid tapping, tool changer, pendant, all built in.

These are going to be a game changer in the future.

Boyan Silyavski
12-09-2016, 12:39 PM
G84 is rigid tap cycle.

You should have bought one of these.




450 landed in UK.

No computer, no monitor, no windows license, no Mach licence, No CSIO box of tricks.

Does rigid tapping, tool changer, pendant, all built in.

These are going to be a game changer in the future.


You are reading my mind, i have difficulty expressing how i feel about each one of the BOBs available now on market, or it would be expensive, or 5v,or not shielded, or will not do this or that correctly, or ...http://www.sherv.net/cm/emoticons/guns/shooting-guns.gif

Davek0974
12-09-2016, 12:40 PM
Ouch, nice box. Didn't know about it then though, I also like the more graphical nature of Mach3, tweaks, tricks, customisations etc :)

Davek0974
12-09-2016, 01:01 PM
This might be getting there, treat it as a drill, chip-break operation, full retract, set peck depth to thread depth minus the 5mm for tapping clutch, pause before retract to allow clutch to disengage then retract.

I se for a 6mm x 1mm thread so i chose 500 rpm and 500mm/min feed should match the pitch rate of 1mm/rev ??

Only issue i see here is that it rapids out as G0, will look at that.


(1)(T10 D=6.1 CR=0. TAPER=118DEG - ZMIN=-5. - DRILL)
G90 G94 G91.1 G40 G49 G17
G21
G28 G91 Z0.
G90


(DRILL2)
M5
M9
T10 M6
S500 M3
G54
M7
G0 X0. Y0.
G43 Z16. H10
Z2.
G1 Z-5. F500.
G4 P4.
G0 Z6.
Z16.


M9
G28 G91 Z0.
M30

Davek0974
13-09-2016, 12:33 PM
Sorted, the guys on the Fusion360 forum edited a post-processor so it can feed in, dwell and then feed out.

Should have it sorted nicely :)

Davek0974
13-09-2016, 04:32 PM
Sub-table for the engraving spindle...

19180

Holes A are for M10 or M12 CSK socket-head screws to bolt into the Bridgeport T-Slots, holes B are two locating dowels on the rear face to sit in the middle slot on the main table, keep it all square etc :)

25mm eco-cast tooling plate.

John S
13-09-2016, 06:06 PM
Dave, once setup drop a couple of shallow keyways in the plate both vertically and horizontally so you can drop a piece of key steel in to act as back and / or side stops to get parts square

Davek0974
19-09-2016, 08:28 PM
Sub-table is coming along nicely, can just be seen in the bottom of the picture, just the matrix holes to drill & tap now...

19237

The new spindle motor fitted my adaptor perfectly, got the VFD dialled in nicely, just the water cooling to sort out.

Davek0974
25-09-2016, 01:31 PM
Job done, turned out very nice.

100 holes, spotted, drilled, countersunk, blind-tapped M6

19282

Surface was parallel to within 0.03mm max so i'm not going to bother running a facing cut.

Colin Barron
26-09-2016, 09:14 PM
G84 is rigid tap cycle.

You should have bought one of these.

http://www.stevenson-engineers.co.uk/files/4axis%20mill.jpg


450 landed in UK.

No computer, no monitor, no windows license, no Mach licence, No CSIO box of tricks.

Does rigid tapping, tool changer, pendant, all built in.

These are going to be a game changer in the future.

Is this going back to the old style controllers?
Are the inputs manual or autocad? Does it connect directly to servo drivers?
Thanks
Colin

John S
26-09-2016, 09:56 PM
Takes standard Fanuc G-code.

Plugs and sockets on the back to go direct to drivers, tool changer, VFD for spindle, encoder for threading or rigid tapping, probe and pendant.

komatias
27-09-2016, 09:50 AM
Is this going back to the old style controllers?
Are the inputs manual or autocad? Does it connect directly to servo drivers?
Thanks
Colin


I would not say old style controllers, I would say professional. There is an element of setup but not as much as you would think. You will also need to change your habits if you are used to Mach or EMC, for example: just zeroing your part may not actually write the current offest to G54.


Some standalones, like my GSK980MDc also allow you access to the ladder logic so you can tweak it to your hearts content. Been trying to learn this myself but it is proving a tad difficult since there is very little to go on except some cryptic manuals.

Another element I have seen is that some controllers will accept DXF file direct input for drilled holes. You set each hole size in a different coloured layer and do all the coding on the console.

What I really like is that it moves away from the classic argument of EMC vs Mach, parallel vs motion controller etc. It just sort of works.

Davek0974
02-10-2016, 09:20 AM
To get back on track....

Have decided to replace the little Bridgeport motor and simple V/Hz VFD with a 3Hp inverter rated motor, sensorless-vector vfd, timing belt drive and encoder - this should give me usable torque and power from about 300rpm to 3000rpm and also semi-closed-loop speed regulation as Mach and the CSMIO controller will maintain the speed as load changes, up to the point of tripping the vfd :)

It means i can finally get rid of the clattering vari-speed pulleys and power-sapping drive and not have to mess with setting the speed manually. I also get actual speed display on-screen.

Also got the braking resistor so stopping time should be about 1s or less.

John S
02-10-2016, 11:32 AM
Told you ages ago you should have bought that step speed head with brand new 3Hp motor already fitted ( sigh )

Davek0974
02-10-2016, 11:37 AM
Yeah i know, live and learn :)

I think this this will be a step ahead of even the step-pulley though as it will be timing belt drive and no step-pulleys at all ;)

Needs the timing belt as i'm mounting the encoder on the motor shaft at 1:1 drive

Davek0974
02-10-2016, 04:03 PM
Any ideas for encoder mounting?

I was thinking of coming off the end of the motor shaft - there is an M10 threaded hole in it, then connecting the encoder to the shaft. I'm not sure if a rigid connection is a good idea though, is a coupling of some sort better here??

The other option is a 36t HTD5 pulley bored out to 35mm to fit the spindle shaft and mount this between the spindle drive pulley and the old lower vari-disk/brake disk in the head then mount the encoder out the front where the old speed control used to live using another 36t pulley. Not as cosmetic as the motor shaft idea but??

Any thoughts?

Boyan Silyavski
02-10-2016, 04:54 PM
I remember when once was looking into that, that there were encoders that could count straight from the HTD pulley teeth. though there was some limit on the speed, but i think you are in. Though cant remember more for them right now

Davek0974
02-10-2016, 05:45 PM
Probably, i already have the encoder though ;)

m_c
03-10-2016, 02:05 PM
Read the installation requirements for the encoder.
Depending on the style of encoder, there are different installation requirements.

Davek0974
03-10-2016, 02:37 PM
Thanks, was just looking for ideas - the encoder is a rugged little unit with self-supporting shaft capable of something like 15N radial loading which is pretty tight for a belt.

I'll try the motor shaft idea first as i have the bits for that, if i need pulleys i'll have to order some up.

m_c
03-10-2016, 02:52 PM
You'll probably want some kind of flexible coupler for that style of encoder, as unless you can guarantee very accurate alignment, you'll likely kill the encoder pretty quickly.

An encoder with a separate disc/reading head would of been a better choice for mounting directly onto a shaft.

Davek0974
03-10-2016, 03:28 PM
Thanks,

I have seen the actual encoder body mounted on a spring plate presumably to allow for small misalignment

I will get a flex coupling in there.

Davek0974
04-10-2016, 07:23 PM
Well, the motor was fitted, the pulleys wouldn't turn so something was wrong, then i discovered the rib that was cast into the top housing cover was rubbing badly on the top of the belt edge :(

Off it all comes again, rib was duly altered with a grinder ;) all back together again and nice smooth movement this time, my measurements and belt length calculations were spot-on and the adjuster slot on the motor plate was mid-way when tight. :)

Lugging a 3Hp motor up top on a Bridgeport on your own with only a pair of steps is no fun, doing it twice was pushing the limits a bit ;)

Just got to fit the VFD, wire up and we should be good.

Davek0974
05-10-2016, 10:24 AM
Something that rarely gets mentioned...Power required...

I have a 3ph 2.2kW spindle motor on a vfd, 700w X servo, 700w Y servo, 400w Z servo, maybe 100w for pc and controller on this machine.

How do i calculate the supply capacity allowing for diversity - it will not be running 100% flat out all the time :)

I know i need to uprate the MCB's and supply cables as the build was originally for the small 1.5hp spindle motor.

Adding up the maximum ratings for all drives gives a figure way more than i can supply :)

m_c
05-10-2016, 11:27 AM
A quick mental calc, and I'd say 16A would cover you.
2200+700+700+400 = 4000.
Divide that by 220 gives 18A, or 250 gives 16A.

Power up surges of the drives are likely to give a bigger problem than too much current draw during use.

Davek0974
05-10-2016, 12:07 PM
Thanks, so there is no difference in current draw if a 3ph motor is running on 3ph or on converted single to 3ph via a vfd (apart from the internal vfd usage) ??

Davek0974
05-10-2016, 02:26 PM
Just got a reply from my supplier, in a flat-out full-load situation, the draw on the supply will be 22.5A, above that then the inverter will start reducing speed, prolonged load will trigger a shutdown.

Question is - can a small Bridgeport actually USE 3Hp/15Nm at the spindle :)

m_c
05-10-2016, 07:29 PM
Thanks, so there is no difference in current draw if a 3ph motor is running on 3ph or on converted single to 3ph via a vfd (apart from the internal vfd usage) ??
That depends on where you're measuring the current, which is why you use the KW rating to work it out.
I was going to mention just how much the VFD will allow the motor to run above it's rated power, but I see the supplier has answered that.
Although the motor is rated at 3HP/2.2Kw, that'll be at 50Hz speed rating. When you increase speed, you can think of AC induction motors just like a steppers, in that the key factor is how much current you're putting through the windings.
So double the speed, and you can double the power, which means your 2.2kw motor running at double the 50Hz speed could produce well over 4kw.
Off course as with steppers, you need to monitor temperature, and if running at low speed a lot, you may want to consider getting a forced induction kit for the motor (aka a cooling fan that bolts onto the motor) to avoid overheating the motor.

Davek0974
05-10-2016, 08:11 PM
Thanks, i had it running tonight and it draws about 3A idling.

It has a PTC built in for the temperature and that will be wired in to the VFD in due course.

Looks like i have exactly 10:1 speed ratio from 10 to 100Hz which figures, works out nicely at 300 to 3000rpm.

Stopping in 1s is pretty impressive :)

Now to configure it for cnc control...

Colin Barron
08-10-2016, 04:14 AM
Thanks, i had it running tonight and it draws about 3A idling.

It has a PTC built in for the temperature and that will be wired in to the VFD in due course.

Looks like i have exactly 10:1 speed ratio from 10 to 100Hz which figures, works out nicely at 300 to 3000rpm.

Stopping in 1s is pretty impressive :)

Now to configure it for cnc control...

Will the momentum of the spindle and cutting tool not damage the belt if you continually stop quickly?

John S
08-10-2016, 07:42 AM
Probably not the belt but it will shag the keys and keyways up in the back gear unit.

That and the varispeed sheaves which Dave has got rid of are the weak points in these heads.
No idea why anyone wants to stop that quickly anyway, it's not like a Bridgy is a production machine and saving 1 second on 15 auto toolchangers is a game changer

Davek0974
08-10-2016, 09:30 AM
Yes, its hard on the head and it sounds hard on the head, thats my reason to slow it down a bit, there really is no point in sudden stop on a machine with splined quill and all that other crap up top ;)

The chances are that most of the dog-clutch wear was caused by my using it for running a 12-1/2" dia fly-cutter when cutting the saddles on the traction engine cylinder block and chimney base and perch-bracket - these would take many hours of running in back-gear and really worked the old girl. It was a job for a HBM really but when you have a Bridgeport you can do pretty much anything;)

The tiny splines and those dodgy dogs are certainly a weak spot, rarely fail but wear a lot and get very vocal, you can wreck the dogs by being a jack-the-lad and changing gear while the spindle is running, would never do this but have seen the carnage caused.

The belt i have fitted is good for something like 60-70hp so i doubt that will be bothered ;)

Davek0974
10-10-2016, 08:33 PM
Been messing with rigid tapping - looks like it will be good:)

Also have live spindle speed on screen by changing the DRO from 202 to 39

Have set the vfd display to show load in kW - i could mount the display remotely or fit an analogue 0-10v meter which would show 0-150% of motor power. So far i have not made it go over 600W :)

Getting there.

Davek0974
16-10-2016, 02:09 PM
Fun little Sunday morning project to aid my education in Fusion, CAM etc...

Support bracket for the high speed spindle when not in use :)

Junk bits of 15mm ally plate, rigid tapping, chamfers the works ;)

19463

19462

Davek0974
23-10-2016, 05:07 PM
Future thoughts...

Machine is working very nicely so far. However I still have big doubts about the weakest link in the system - the Z-axis drive, its a pretty tortured setup, the connection to the quill is poor being just one bolt, the block face is only around 15-20mm square where it bolts on, there is around 75mm from the quill to the ball-screw so there is a massive twisting moment acting on the joint and the ball-nut.

There is already movement visible between the nut and the quill - you can see the connecting arm twisting when nudging the ball-screw pulley by hand so it would be pretty rough under drilling or plunge-milling forces. I can see a very short life for the ball-nut as they don't like twisting moments, axial or radial but not twisting.

It was assembled with loctite but with only the one bolt I doubt that would do much.

I have a feeling it would have been better to go with the drive idea in the pictures below, the quill rack is designed to take all the Bridgeport can do, there is not much backlash and what there is could be helped with a gas-spring maybe to keep it under positive pressure.

No idea what the pinion gearing ratio is but likely something like 3" per turn so would use a bigger reduction from motor.

Just looking ahead, but as i said, I can see issues here already and thats only after a few jobs....

19486
19487