Upgrade? Chinese CNC Machine

[Carlb1058]

Hi All,

Total noob to the world of CNC, but not thick* - I'm sorta leaning toward thickendum at the moment though. I bought a Chinese CNC machine (yes I know now) and the controller a YooCNC is naff and couldn't cut a straight line if its life depended on it and it does so it's now one for the bin.

'Er indoors is getting a tad peeved shall we say at the fact that a) I bought a naff machine (eBay Australia: Buy new & used fashion, electronics & home d (http://www.ebay.com.au/itm/300599305131?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649)) and b) I gotta spend more money to fix it so with that said how do I go about sorting the controller, remembering the budget is getting slimmer coz I so wanna buy a Korg Oasys to pad out my ever growing synth collection :yahoo:

I've been looking at various controllers but this is where I get confused with amps watts and volts. The pictures attached are of the current controller internals showing Primary 220v Secondary 0-18v and 0-36v.
Can I just buy for example one of these
CNC 4 Axis TB6560 Stepper Driver Board Controller +CD | eBay (http://www.ebay.com.au/itm/CNC-4-Axis-TB6560-Stepper-Driver-Board-Controller-CD-/181098084744?pt=LH_DefaultDomain_15&hash=item2a2a497988)
and mate it to my existing PSU/controller box or would I be better off buying something like this New CNC Standard 4Axis TB6560 Stepper Driver Full Kit & Motor/PSU/Keypad/Display | eBay (http://www.ebay.com.au/itm/New-CNC-Standard-4Axis-TB6560-Stepper-Driver-Full-Kit-Motor-PSU-Keypad-Display-/350750739256?pt=AU_B_I_Electrical_Test_Equipment&hash=item51aa5f8b38)
and just using my existing frame work which from what I can see is pretty damn solid.

Any help would be greatly appreciated.

8599
8600
Old naff controller box

Cheers
Carl

[Swarfing]

Yaki Da carl

Before you go wasting more money on controllers similar to what you already have. Tell us more about your machine. Straight lines could be down to that and not the electronics. You could probably do with some settings changing if your loosing steps etc. pics are good

[martin54]

I doubt if it's the electronics if it's cutting but not a straight line, that could be a number of things as has been said including the operator setting wrong speed & feed rates. Pictures would help as has been said. A video of it cutting would be even better.

[John S]

Don't bother with those boards, they are the same as what you already have

Two choices.
Either try to get it running better and from experience these boards do not like micro stepping, they loose steps. perhaps start here ?

Second choice is the box has a decent power supply so scrap the board and fit 3 decent drivers and a breakout board.
see
CNC Micro-Stepping Stepper Motor Driver 2M542 Bi-polar 2phase 4.2A Switch | eBay (http://www.ebay.co.uk/itm/170846930771) and
Upgraded 5 Axis CNC Breakout Board Stepper Motor Driver + 5V USB Cable + Mach3 | eBay (http://www.ebay.co.uk/itm/321094044282)

BTW does anyone recognise the power supply board in that box ? A quick look on ebay doesn't get me anything.

[Jonathan]

Might not even need to get a new PSU to use with the m542 drivers. The transformer says 36V on one of the secondaries, which is ideal since when rectified that will give just under 50V. It depends what rating that winding is on the transformer, but you might as well try it if you do decide to get the drivers that John S (and I) recommend. Do you have a part number / datasheet for the stepper motors this machine uses?

[John S]

Yes spotted that as well, even included it in the post that disappeared when I clicked on the picture again to check.
I reckon the tranny will kick enough out given that these motors only take a bit of power and volts on these things is far more important than amps.

When you have had your fingers in a switch on electric toaster you get a feel for these sort of things. :highly_amused:

[John S]

Still interested in that board that rectifies the current from the tranny.
I reckon they must be commercially available boards if they are in that box.

Hard bit is know what to search for on Ebay.

[Jonathan]

The motors look fairly small, so I tend to agree. However I've just noticed that the 36Vac output will be for the spindle, since that requires up to 50V and 230W according to the eBay listing. Unless that spindle is replaced another transformer would almost certainly be needed.


Still interested in that board that rectifies the current from the tranny.
I reckon they must be commercially available boards if they are in that box.

Hard bit is know what to search for on Ebay.

Er why, surely you don't want one? It looks like it does a bit more than just rectifying though - I think it does the PWM drive for the spindle.

[John S]

Well spotted and that also answers the question of the pot on the front so scrap that query.

The reason I wanted one is that is often quicker, faster, cheaper in the long run to buy off the shelf components than build.
You then also have the facility to just ship a component out in the event of a fry up.

However in this case it's not what I was looking for.

[Jonathan]

You want something like this then?

10000uF 50V X4 Rectifier filter power supply board kit s-8 | eBay (http://www.ebay.co.uk/itm/10000uF-50V-X4-Rectifier-filter-power-supply-board-kit-s-8-/271171158811?pt=US_Amplifier_Parts_Components&hash=item3f230f831b)

Just search on eBay for 'power supply rectifier' and you'll find plenty, although they are a bit expensive.

Edit: If you're happy to solder it then..
http://www.ebay.co.uk/itm/63V-2200UF-8-Rectifier-filter-power-supply-board-KIT-/321025445293?pt=LH_DefaultDomain_0&hash=item4abe9b8dad
http://www.ebay.co.uk/itm/63V-2200UF-8-Rectifier-filter-power-supply-board-finish-/220774675890?pt=LH_DefaultDomain_0&hash=item33673235b2

[Swarfing]

Straight lines

God knows where that came from?

[Carlb1058]

Yaki Da carl

Before you go wasting more money on controllers similar to what you already have. Tell us more about your machine. Straight lines could be down to that and not the electronics. You could probably do with some settings changing if your loosing steps etc. pics are good

Sorry, I was bein' a sarcy clever dick re the straight line reference :devilish: I've attached another pic as to what it does. I thought if it can't engrave something, surely it should be able to cut out a 52mm diam hole a 39mm diam hole and a rectangle around the pair so I could make a vacuum hose mount - boy was that a total failure. My son-in-law's da used to work in the tool industry and knows his way around g-code so I showed him the g-code to make sure I didn't have any errors and to the best of his knowledge it should do just what it's written for which is making me lean toward the controller not doing what it's bloody well told which is probably Karma comin back for me for bein' the same as a kid :friendly_wink:

The blue line in the image is where the rectangle is SUPPOSED to be, as for the holes well they're just plain outta shape. Oh and that was a scrap piece of MDF so try and ignore all the other lines at the bottom of the image and the upside down letter in the top right etc and you will see what happened to the rectangle :wink:

Cheers
Carl

8607

[Carlb1058]

Hi,

Unfortunately, all that came with the machine was a teeny disk with a manual on that when printed tells you nothing in regards to tech specs, essentially it's just an overbloated example of how to find the power button.

My biggest problem is when it comes to electronics I have the IQ of a door knob, if the thing was powered by a Meriden twin I'd have it torn down, rebuilt and running as sweet as nut but it's not so it's back to being as bright as a blown bulb for me :wink: How many of those m542 drivers would I need, three - one for each axis or is my thickendum sowing again?

Cheers
Carl

Ignore the last question John S pointed out I'd need three *slaps forehead*

[John S]

Looked at the pic and the steps are out on the setup.
You probably don't know what that means which proves you are thick but no sweat for a Welshman living in OZ we all feel sorry for you.

OK we need a DECENT picture of the stepper board in the controller box from vertically down, hell take 3 or 4.
then we need a screen shot of the setup page in Mach.

That's Config > Motor Tuning

Next double check it does have ball screws and the pitch is 4mm

If you get stuck, shout up we need a good laugh.

[Carlb1058]

Looked at the pic and the steps are out on the setup.
You probably don't know what that means which proves you are thick but no sweat for a Welshman living in OZ we all feel sorry for you.

OK we need a DECENT picture of the stepper board in the controller box from vertically down, hell take 3 or 4.
then we need a screen shot of the setup page in Mach.

That's Config > Motor Tuning

Next double check it does have ball screws and the pitch is 4mm

If you get stuck, shout up we need a good laugh.

Last things first, short of tearing the machine down to measure the screw I got no idea if it's 4mm or not but going off appearance it's def a ball screw. The Config > Motor Tuning is set to machine manufacturer specs as shown in the manual that came with the machine.

As for images and screen caps here goes...

8620862186228623

8625862686278628862986308631

[John S]

Mach setup looks OK

Pics don't really show anything.

010.jpg is the best shot, I'm looking for dip switches that set microstepping. X and Y should be the same if the screws are the same and the motors are both direct drive.

On the picture of the circles, is the wide diameter X or Y ?

If it's X change the steps per on Y from 400 to 200
If it's Y then change X from 400 to 200 and run the file again.

Remember to click save settings, then OK and reboot.

Off to bed now, it's half past dark here, can you play quietly with yourself until morning ?

[Carlb1058]

Mach setup looks OK

Pics don't really show anything.

010.jpg is the best shot, I'm looking for dip switches that set microstepping. X and Y should be the same if the screws are the same and the motors are both direct drive.

On the picture of the circles, is the wide diameter X or Y ?

If it's X change the steps per on Y from 400 to 200
If it's Y then change X from 400 to 200 and run the file again.

Remember to click save settings, then OK and reboot.

Off to bed now, it's half past dark here, can you play quietly with yourself until morning ?

Def no dip switches anywhere, I'll have a muck about with changing the steps and see what happens and post results - and I never play quietly, the volume is always set to 11

[martin54]

If it were powered by a meridian twin it wouldn't have come from China would it lol,

[Ricardoco]

Hi there, try and slow things down a bit 200 steps per and velocity 200 accel 100 and see if you get different results, also what are you using to Generate your G-Code. Also have you run the driver test for mach to see if it is running as should be. Is the machine dedicated to CNC or are you running other stuff in the background..

Rick

[Carlb1058]

If it were powered by a meridian twin it wouldn't have come from China would it lol,

Good point :bi_polo: although the Hinkleys are made in Thailand :stupid:

[Carlb1058]

Hi there, try and slow things down a bit 200 steps per and velocity 200 accel 100 and see if you get different results, also what are you using to Generate your G-Code. Also have you run the driver test for mach to see if it is running as should be. Is the machine dedicated to CNC or are you running other stuff in the background..

Rick

Rick,

Been toying with the steps and velo etc - getting close but still get it ovalling (if that's a word) on the Y instead of a circle, that is my vernier is showing me right dimensions for a 52mm circle on the X but narrower on the Y - might knock it down some more and see what that does.
As for Driver Test, yep ran that and it's fine, the machine is a not so old PC that was fresh installed with XP SP2 and Mach3 - nothing else on there - not even a virus checker seeing as it's not net connected it's not needed.

I use Aspire 3 on a different machine to create what I want

Addon bit:


http://www.youtube.com/watch?v=QAvjainGpEM

As you can see, it's not a circle??? oh and I started videoin' as an afterthought

[martin54]

If you have been playing about with the steps does in now move the correct distance in each direction? Have you tried just moving an inch or two in a straight line in each direction doing each axis separately? Did you remember to hit save on the motor set up after you had changed the steps on each axis, simple thing but easy to forget to do when things aren't going as planned lol.

[Ricardoco]

Rick,

Been toying with the steps and velo etc - getting close but still get it ovalling (if that's a word) on the Y instead of a circle, that is my vernier is showing me right dimensions for a 52mm circle on the X but narrower on the Y - might knock it down some more and see what that does.
As for Driver Test, yep ran that and it's fine, the machine is a not so old PC that was fresh installed with XP SP2 and Mach3 - nothing else on there - not even a virus checker seeing as it's not net connected it's not needed.

I use Aspire 3 on a different machine to create what I want

Addon bit:


http://www.youtube.com/watch?v=QAvjainGpEM

As you can see, it's not a circle??? oh and I started videoin' as an afterthoughtOk thats better than you had before, so that means we are in the right track, i wonder if both the ballscrew are the same. could you do me a favour and cut a strait line 100mm long on the X and 100mm long on the Y and then measure them and let us know what you get.. Just Like Martin suggested.


Rick

[Carlb1058]

100mm on the Y I get 52mm
100mm on the X I get 42mm

[Ricardoco]

100mm on the Y I get 52mm
100mm on the X I get 42mm now we are talking.. thats why it dosnt do a circle

What have you set your X to in motor tuning,
and
What have you set your Y to in motor tuning,

Rick

[Carlb1058]

now we are talking.. thats why it dosnt do a circle

What have you set your X to in motor tuning,
and
What have you set your Y to in motor tuning,

Rick

Slowed them down to as per your previous suggestion - one thing I did notice when it was doing it was the X was smooth in movement, the Y however seemed to jerk, as if going for example 10mm stages, a barely but still noticeable pause then continuing on?

[Ricardoco]

Slowed them down to as per your previous suggestion - one thing I did notice when it was doing it was the X was smooth in movement, the Y however seemed to jerk, as if going for example 10mm stages, a barely but still noticeable pause then continuing on?Well lets see if i can get the math sorted to get them cutting the correct amount first.

Rick

[John S]

Change X from 200 to 384.6
Change Y from 200 to 476.2

Steps per that is

Then try running 100 mm in X and Y

[Ricardoco]

Slowed them down to as per your previous suggestion - one thing I did notice when it was doing it was the X was smooth in movement, the Y however seemed to jerk, as if going for example 10mm stages, a barely but still noticeable pause then continuing on?LOL my brain has been doing loops here.. What size cutter are you using..

Rick

[Ricardoco]

Change X from 200 to 384.6
Change Y from 200 to 476.2

Steps per that is

Then try running 100 mm in X and YThats what i was comming up with john but dosnt the cutter size have a bearing on it...LOL (where's that human calculator when you need him LOL)

Rick

[Carlb1058]

LOL my brain has been doing loops here.. What size cutter are you using..

Rick

20 deg - 0.5F - 3.175 diameter coz that's what was in it when I wandered out to the shed this morning :)

Onwards and upwards to try these new figures - digits x'ed

[Carlb1058]

20 deg - 0.5F - 3.175 diameter coz that's what was in it when I wandered out to the shed this morning :)

Onwards and upwards to try these new figures - digits x'ed

97 on the X 86 on the Y or so my vernier says :numbness:

[Ricardoco]

97 on the X 86 on the Y or so my vernier says :numbness: so your cutter is a 3mm cutter?

[Ricardoco]

ok Try this

X 408.1
Y 512.8

with the same cutter of course

Rick

[Carlb1058]

ok Try this

X 408.1
Y 512.8

with the same cutter of course

Rick

100 on the Y 105 on the X - I know this'll all go way over my head but how do work it out?

[Swarfing]

so your cutter is a 3mm cutter?

Surely that is an 1/8" cutter?

[Ricardoco]

100 on the Y 105 on the X - I know this'll all go way over my head but how do work it out? Ok divide 200 ~ (the 'steps-per' that is set on the axis in motor tuning) by the cut measurement minus the cutter measurement(Dia 3mm as the software is running in metric mode 0r 1/8" if you require brain drain ), and then multiply by the measurement you told it to cut which was 100mm..

Rick

[Ricardoco]

Surely that is an 1/8" cutter? It may well be.

[Swarfing]

Are we actually sure these are metric screws? did they kick you a machine built for our American cousins and give you one with imperial screws?

[Ricardoco]

Are we actually sure these are metric screws? did they kick you a machine built for our American cousins and give you one with imperial screws? Im not sure what they are but something is not right but then when its cutting a circle we can get him to look a bit deeper into it.. I do however suspect you may be correct at least in some part..

Rick

[Carlb1058]

Are we actually sure these are metric screws? did they kick you a machine built for our American cousins and give you one with imperial screws?

No idea, it came with a 220v PSU not that that means diddly squat - so if I tear it down to the point I can reach the screws how do you do measurements, the same as measuring the pitch on a normal screw?

[Carlb1058]

Ok divide 200 ~ (the 'steps-per' that is set on the axis in motor tuning) by the cut measurement minus the cutter measurement(Dia 3mm as the software is running in metric mode 0r 1/8" if you require brain drain ), and then multiply by the measurement you told it to cut which was 100mm..

Rick

I get 1500.68 :hororr: methinks I did summut wrong

[Swarfing]

Carl measure the pitch of the leadscrews

[Ricardoco]

I get 1500.68 :hororr: methinks I did summut wrong Ok

200/ (52-3) x 100
or
200 / 49 x 100 = 408.16

where 200 is the original steps per and 49 is the output cut measurement minus the tool diameter, multiplied but the input cut measurement..


is that a Bit easier?

Its a brain doo'er i know

LOL

Rick

[Ricardoco]

Carl measure the pitch of the leadscrewswhy not just follow 1 path at a time then the chap wont get too confused..

Rick

[Swarfing]

Agreed......

[Carlb1058]

why not just follow 1 path at a time then the chap wont get too confused..

Rick

Just turned it over and measured the Y and I get 3 to a cm - whatever that works out to *shrug*

[Ricardoco]

Just turned it over and measured the Y and I get 3 to a cm - whatever that works out to *shrug*3 turns to a cm yes? What ever they are i suspect the pitch is not the same for x and y

Rick

[Carlb1058]

Ok

200/ (52-3) x 100
or
200 / 49 x 100 = 408.16

where 200 is the original steps per and 49 is the output cut measurement minus the tool diameter, multiplied but the input cut measurement..


is that a Bit easier?

Its a brain doo'er i know

LOL

Rick

Just a tad - cheers :) I ended up with 409.62 using 3.175 as the tool diameter

[Carlb1058]

3 turns to a cm yes? What ever they are i suspect the pitch is not the same for x and y

Rick

Yep 3 to a cm the biggest problem with trying to measure the X is it is fully enclosed and I'd pretty much have to tear it apart to get to it but I shall attempt to get a close as damn first :)

[Ricardoco]

Just a tad - cheers :) I ended up with 409.62 using 3.175 as the tool diameterOnce we have got both axis to move the same amount i can do the math again and give you the imperial equiv, then we can change the system to old school as like swarfing said it could be a yank model. but we will get it for you then of course that will rule out the mechanical and software stuff as the source of your trouble, leaving only the electrics but they may be fine and just need proper BOB and stepper drivers should you be so inclined..

Rick

[Ricardoco]

Yep 3 to a cm the biggest problem with trying to measure the X is it is fully enclosed and I'd pretty much have to tear it apart to get to it but I shall attempt to get a close as damn first :)

Dont bother it can be worked out with the software al be it a little long winded..

Rick

[Carlb1058]

Yep 3 to a cm the biggest problem with trying to measure the X is it is fully enclosed and I'd pretty much have to tear it apart to get to it but I shall attempt to get a close as damn first :)

well that was easier than expected, managed to get a 1cm wide flat file between the frame work and as near as damn the X is 3 turns as is the Z, so all the screws are the same size

[Swarfing]

confused but could be possible?

7.26TPI 0.13774" 3.499mm pitch is about as close as you could get to that. Stick with tuning it out for now like Rick said before. Imperial or not you would still like it working with what you have. Sorry i took this away from that.

[Carlb1058]

confused but could be possible?

7.26TPI 0.13774" 3.499mm pitch is about as close as you could get to that. Stick with tuning it out for now like Rick said before. Imperial or not you would still like it working with what you have. Sorry i took this away from that.

No drama, it's always good to learn new stuff :)

[Swarfing]

Got to hit the sack, hope get somewhere soon

Paul

[Ricardoco]

Got to hit the sack, hope get somewhere soon

PaulLightweight...LOL

Rick

NN mate

[Ricardoco]

Lightweight...LOL

Rick

NN mateLooks like im a lightweight too Im off to bed as well, speak tomorrow Carl

Rick

[Carlb1058]

Looks like im a lightweight too Im off to bed as well, speak tomorrow Carl

Rick

no problems - all the help so far has been great

[phill05]

Why not get Mach to do the work for you ? go into "Settings tab", just above the reset button "set steps per unit" start with X enter 100 measure how far it moved then tell it how far it moved and will set steps up for you.
Do the same for all axis do a test cut if not correct go though it again.

Phill

[Ricardoco]

Why not get Mach to do the work for you ? go into "Settings tab", just above the reset button "set steps per unit" start with X enter 100 measure how far it moved then tell it how far it moved and will set steps up for you.
Do the same for all axis do a test cut if not correct go though it again.

PhillYep that would be better, ide forgot about that.

Rick

[Jonathan]

I dislike just scaling the step/mm values to make it work as it can mask more fundamental issues, and generally isn't as accurate. You should be able to measure the ballscrew pitch and work it out. The comment earlier saying the 'the Y however seemed to jerk' implies it could be stalling as the motion should be very smooth. If that is the case you need to find what's causing that, else the distance readings for that axis are meaningless.

Instead of just measuring a small distance along the screw, measure say 20 turns and divide the result by 20. Assuming you can access the motors, you shouldn't have to dismantle anything. Just measure the distance between the axis and a convenient point, turn it 'n' times by hand and measure the distance again. Divide the difference between those distances by 'n' and you have the pitch. If you've got a digital caliper then measure it over about 150mm to get the most accurate reading.

Doing a test cut is not the best way either, since spindle runout and tool deflection can introduce small errors and it's not possible to measure in wood that accurately anyway. Instead measure how far the axis moves. Measuring the distance between two parallel lines would be easier, and a bit more accurate, than measuring the length of the line, since you're measuring between straight edges not arcs.

Assuming the '3 turns per cm', i.e. 3.3mm pitch is reasonably accurate, the step/mm values we've ended up with don't make much sense. The motors are run from TB6560 ICs, which only support microstepping settings which are powers of 2 up to 2^4 (16). Assuming (as there's a slight chance it may not be) the motors full step resolution is 200 step/rev, that means your motors can be set to 200*2^n where 0<n<4. We need to find steps per mm, so divide by the pitch-> 200*2^n/3.3 = 60.6*2^n, where 0<n<4. That means the correct step/mm value could be one of the following (rounded to 2DP); 60.61, 121.21, 242.42, 484.85, 969.70. None of those seem anywhere near the 408.1 and 512.8 you tried... so I reckon there's something else causing problems here.

You could measure the voltages at the M1 and M2 pins of the TB6560 ICs on your driver board to determine what the microstep setting actually is. The datasheet here (http://www.toshiba.com/taec/components2/Datasheet_Sync/382/27885.pdf) shows which pins to probe.

[Ricardoco]

I dislike just scaling the step/mm values to make it work as it can mask more fundamental issues, and generally isn't as accurate. You should be able to measure the ballscrew pitch and work it out. The comment earlier saying the 'the Y however seemed to jerk' implies it could be stalling as the motion should be very smooth. If that is the case you need to find what's causing that, else the distance readings for that axis are meaningless.

Instead of just measuring a small distance along the screw, measure say 20 turns and divide the result by 20. Assuming you can access the motors, you shouldn't have to dismantle anything. Just measure the distance between the axis and a convenient point, turn it 'n' times by hand and measure the distance again. Divide the difference between those distances by 'n' and you have the pitch. If you've got a digital caliper then measure it over about 150mm to get the most accurate reading.

Doing a test cut is not the best way either, since spindle runout and tool deflection can introduce small errors and it's not possible to measure in wood that accurately anyway. Instead measure how far the axis moves. Measuring the distance between two parallel lines would be easier, and a bit more accurate, than measuring the length of the line, since you're measuring between straight edges not arcs.

Assuming the '3 turns per cm', i.e. 3.3mm pitch is reasonably accurate, the step/mm values we've ended up with don't make much sense. The motors are run from TB6560 ICs, which only support microstepping settings which are powers of 2 up to 2^4 (16). Assuming (as there's a slight chance it may not be) the motors full step resolution is 200 step/rev, that means your motors can be set to 200*2^n where 0<n<4. We need to find steps per mm, so divide by the pitch-> 200*2^n/3.3 = 60.6*2^n, where 0<n<4. That means the correct step/mm value could be one of the following (rounded to 2DP); 60.61, 121.21, 242.42, 484.85, 969.70. None of those seem anywhere near the 408.1 and 512.8 you tried... so I reckon there's something else causing problems here.

You could measure the voltages at the M1 and M2 pins of the TB6560 ICs on your driver board to determine what the microstep setting actually is. The datasheet here (http://www.toshiba.com/taec/components2/Datasheet_Sync/382/27885.pdf) shows which pins to probe.

Hi Jonathan, this was initially to find out if it was an electrical or a mechanical problem, or indeed both, as i understand it there is no way to change the steps on the drivers that could be discovered. So Im glad you are here to cast your eye over it, we also dont know if the screws are metric or imperial although the chap has said they all measure the same..

Rick

[Shinobiwan]

This won't fix your axis scaling issues but in general config are you set to exact stop or constant velocity? I'd recommend constant velocity for smoother movement and less chance of lost steps.

[martin54]

dislike just scaling the step/mm values to make it work as it can mask more fundamental issues, and generally isn't as accurate. You should be able to measure the ballscrew pitch and work it out. The comment earlier saying the 'the Y however seemed to jerk' implies it could be stalling as the motion should be very smooth. If that is the case you need to find what's causing that, else the distance readings for that axis are meaningless.

This is pretty much what I was going to do with the gerber when the time came so how would I go about setting that up? It's belt driven via gearing rather than being a ballscrew.

[Carlb1058]

I dislike just scaling the step/mm values to make it work as it can mask more fundamental issues, and generally isn't as accurate. You should be able to measure the ballscrew pitch and work it out. The comment earlier saying the 'the Y however seemed to jerk' implies it could be stalling as the motion should be very smooth. If that is the case you need to find what's causing that, else the distance readings for that axis are meaningless.

Instead of just measuring a small distance along the screw, measure say 20 turns and divide the result by 20. Assuming you can access the motors, you shouldn't have to dismantle anything. Just measure the distance between the axis and a convenient point, turn it 'n' times by hand and measure the distance again. Divide the difference between those distances by 'n' and you have the pitch. If you've got a digital caliper then measure it over about 150mm to get the most accurate reading.

Doing a test cut is not the best way either, since spindle runout and tool deflection can introduce small errors and it's not possible to measure in wood that accurately anyway. Instead measure how far the axis moves. Measuring the distance between two parallel lines would be easier, and a bit more accurate, than measuring the length of the line, since you're measuring between straight edges not arcs.

Assuming the '3 turns per cm', i.e. 3.3mm pitch is reasonably accurate, the step/mm values we've ended up with don't make much sense. The motors are run from TB6560 ICs, which only support microstepping settings which are powers of 2 up to 2^4 (16). Assuming (as there's a slight chance it may not be) the motors full step resolution is 200 step/rev, that means your motors can be set to 200*2^n where 0<n<4. We need to find steps per mm, so divide by the pitch-> 200*2^n/3.3 = 60.6*2^n, where 0<n<4. That means the correct step/mm value could be one of the following (rounded to 2DP); 60.61, 121.21, 242.42, 484.85, 969.70. None of those seem anywhere near the 408.1 and 512.8 you tried... so I reckon there's something else causing problems here.

You could measure the voltages at the M1 and M2 pins of the TB6560 ICs on your driver board to determine what the microstep setting actually is. The datasheet here (http://www.toshiba.com/taec/components2/Datasheet_Sync/382/27885.pdf) shows which pins to probe.

Hi Jonathan,

Clarification on the following please:
1) measure the distance between the axis and a convenient point
Tipping the machine over gives me clear access to the Y screw, both X & Z have a lot of body work around them which seriously limits any access - so with that said, please define axis and convenient point, don't forget you're talking to a total newbie here, the only machines we had in school were a bench drill and lathe - damn, calculators didn't even surface until years after I left!!
2) measure the voltages at the M1 and M2 pins
All I have in regards to voltage measurement gear is a simple multimeter that I use for checking the wiring on the Meriden and my FLH will this suffice?

[Jonathan]

1) measure the distance between the axis and a convenient point

Measure how far the axis moves. So for example on X move the gantry to one end, then turn 20 turns (or whatever) and measure how far it moved. You don't need access to the screw.

The meter you have will be fine. Just connect the negative probe to negative, (i.e. ground) and the other one to the correct pins. You may have to unscrew the board to access one of them.

[Carlb1058]

Measure how far the axis moves. So for example on X move the gantry to one end, then turn 20 turns (or whatever) and measure how far it moved. You don't need access to the screw.

The meter you have will be fine. Just connect the negative probe to negative, (i.e. ground) and the other one to the correct pins. You may have to unscrew the board to access one of them.

Access to the chips wasn't too bad - off all three I got a reading of between 2.2 and 2.3 and occassionally -2.2 & -2.3 but I'd say that's my multimeter playing silly buggers.

As for turns per 20 on the X gave me 80mm, 10 on the Z gave me 40mm (couldn't turn it 20) and 20 on the Y gave me 62mm

[John S]

Recheck Y in a couple of places and in both directions
Reason for this is 62 mm doesn't work out to a normal pitch and you did say y was jerky.

I'm wondering if you have a loose coupling or similar ?

[Ricardoco]

Recheck Y in a couple of places and in both directions
Reason for this is 62 mm doesn't work out to a normal pitch and you did say y was jerky.

I'm wondering if you have a loose coupling or similar ?I concur

Rick

[Carlb1058]

Recheck Y in a couple of places and in both directions
Reason for this is 62 mm doesn't work out to a normal pitch and you did say y was jerky.

I'm wondering if you have a loose coupling or similar ?

Just measured it at centre of table and the other end to original and still got 62??

Addon bit: Just re-checked and made sure all connections were tight and they were, this is where it gets weird again - using a spanner on the opposite end of the Y screw I turned it 20 got 80mm, turn the knob on the end of the motor and get 62 but if I try and tighten the grub screws up any more all I succeed in doing is bending the allen key????

[John S]

Is the knob on the end of the motor slipping ?

You need to sort this bit out first, try welding the stepper motor to the ball screw.









Just joking.

[Carlb1058]

Is the knob on the end of the motor slipping?

You need to sort this bit out first, try welding the stepper motor to the ball screw.

Just joking.

Welding - cool, now ya talking :encouragement:

The knob itself is on the motor that tight I'd break it if I tried to remove it so the only thing left is the coupling I might go tear the motor off and suss that lot out

[Carlb1058]

Y now equals 80mm on 20 turns and I've got one bent allen key :witless:

[Ricardoco]

Y now equals 80mm on 20 turns and I've got one bent allen key :witless: Now your where you should be... thats cool it was the hardware(maybe LOL) not the electrics (maybe) All the clever ones are called john in one form or another LOL

Rick

[Carlb1058]

Now your where you should be... thats cool it was the hardware(maybe LOL) not the electrics (maybe) All the clever ones are called john in one form or another LOL

Rick

Rick,

Yep gettin' there, just uninstalled, rebooted, re-installed, rebooted, re-setup Mach3, rebooted, loaded the g-code for the 100mm X line and the 100mm Y line and get 102mm on both. . .

used stock settings of 400 steps, 2000 velo, 200 accel for both X&Y, Z is 400s, 700v, 100a

[Ricardoco]

Rick,

Yep gettin' there, just uninstalled, rebooted, re-installed, rebooted, re-setup Mach3, rebooted, loaded the g-code for the 100mm X line and the 100mm Y line and get 102mm on both. . .

used stock settings of 400 steps, 2000 velo, 200 accel for both X&Y, Z is 400s, 700v, 100aits very odd because the measurements indicate that you have 4mm pitch screws.
Well at least they are both the same now.. LOL You can now try the math we talked about last night.. 400/ (cut length - tool dia) x input cut length= new steps per..

Rick

[Carlb1058]

its very odd because the measurements indicate that you have 4mm pitch screws.
Well at least they are both the same now.. LOL You can now try the math we talked about last night.. 400/ (cut length - tool dia) x input cut length= new steps per..

Rick

I'll try the new number (413.11 to the nearest dp) shortly, it's bucket'in down and the part time work I've got to keep me in toys keeps askin' me when I'm gunna do some so I best do some aye LOL

I'll let ya know how it goes. . .

[Jonathan]

its very odd because the measurements indicate that you have 4mm pitch screws.

The eBay listing indicates he has 4mm pitch screws and 20 turns moved it 80mm, which also indicates 4mm pitch screws as 80/20=4mm. So nothing to worry about.

The voltage readings are strange - should have got 5V or 0V, but nevermind as we now know they're set to 1/4 microstepping.

[Carlb1058]

(bit snipped for space)
The voltage readings are strange - should have got 5V or 0V, but nevermind as we now know they're set to 1/4 microstepping.

I didn't get anywhere near 5v, the readings I got were after atleast a dozen tests the first few were around the 2.7 - 3.2 range and they slowly dropped until they stayed pretty static at 2.2 - all three chips did exactly the same thing.

I wonder if it's worth just throwing an extra 90 bux at it for a board with dip switches etc?

[John S]

NO.

If you are going to throw money at it get the separate drives.

More reliable and if one goes down you can replace it, if one of those Toshiba boards goes down the whole lot is toast.

[Ricardoco]

Hey John I agree those boards are rubbish, what would you suggest he gets remembering his budget?

Rick

[Lee Roberts]

(0r 1/8" if you require brain drain )

lmao.

.Me

[Carlb1058]

Getting there, thanks to one and all who chimed in with all the info it's greatly appreciated.

Attached is a short run video - I really should get the camera out before I start aye :whistle: anyhoo I'm pretty sure you can pick who it is that's being engraved, all I need to do now is get the probe working i.e. retracting on touch not spinning the motor and twisting the probe cables in to a ball :crushed: The finally motor tuning ended up at 400s 2500a 200v - my initial test of this setup resulted in a 100mm straight line on X and on the Y so with that, WOOHOO although I still need to sort the Z out, time to sit and mind melt over the mach manual

BTW the line art used for this came from my tattoo line art collection a job I did for a few decades before I got bored. . .

So if anyone is looking for anything in regards to line art feel free to let me know, odds are I've probably got something that can be worked on if not exactly what you want.


http://www.youtube.com/watch?v=oENqLPxGsmc

8645

[Carlb1058]

NO.

If you are going to throw money at it get the separate drives.

More reliable and if one goes down you can replace it, if one of those Toshiba boards goes down the whole lot is toast.


Like this? 3x M542 CNC Stepper Driver For Nema17 Nema23 Nema34 Motor +5 Axis Breakout Board | eBay (http://cgi.ebay.com.au/ws/eBayISAPI.dll?ViewItem&isRelated=1&item=350736424891) if so that'll be a doddle, her who must be obeyed should only hit me two or three times at that money :yahoo: plus that means more money for longer screws etc as I want to build a bigger one now!!

[Swarfing]

That would certainly do the job and you can also run the steppers at a higher voltage (50v) with those.

[Ricardoco]

Getting there, thanks to one and all who chimed in with all the info it's greatly appreciated.

Attached is a short run video - I really should get the camera out before I start aye :whistle: anyhoo I'm pretty sure you can pick who it is that's being engraved, all I need to do now is get the probe working i.e. retracting on touch not spinning the motor and twisting the probe cables in to a ball :crushed: The finally motor tuning ended up at 400s 2500a 200v - my initial test of this setup resulted in a 100mm straight line on X and on the Y so with that, WOOHOO although I still need to sort the Z out, time to sit and mind melt over the mach manual

BTW the line art used for this came from my tattoo line art collection a job I did for a few decades before I got bored. . .

So if anyone is looking for anything in regards to line art feel free to let me know, odds are I've probably got something that can be worked on if not exactly what you want.


http://www.youtube.com/watch?v=oENqLPxGsmc

8645

Its always nice when you can actually make something, its like light at the end of the tunnel..Im glad you are now getting some good results.

Rick

[GEOFFREY]

Well done, I knew you would get there - with a little help from your friends! G

[martin54]

If your thinking about building a bigger machine then make sure anything you buy to get this one running can be used again with the new machine, will help keep the cost down plus collect parts for the new one quicker.

[Carlb1058]

Well done, I knew you would get there - with a little help from your friends! G

Understatement there G - methinks I've come across a damn good site fulla damn good people!!

[Carlb1058]

Its always nice when you can actually make something, its like light at the end of the tunnel..Im glad you are now getting some good results.

Rick

Cheers Rick, and thanks for the help - atleast now I can do some of the required math should I need it again :smug:

[John S]

Wait till you get the bill !!

[Ricardoco]

if you are going to build another machine you can use that one to make the parts for it!!

If you need any help Just give a hollar and im sure there will be plenty of advice forthcomming.
Down load Google sketchup and start designing.. Its very easy and if you get stuck give me a shout..

Rick

[Swarfing]

If he does that then he will be rocking and rolling

[Carlb1058]

Wait till you get the bill !!

Payable in Hobnobs? :bi_polo:

[Carlb1058]

if you are going to build another machine you can use that one to make the parts for it!!

If you need any help Just give a hollar and im sure there will be plenty of advice forthcomming.
Down load Google sketchup and start designing.. Its very easy and if you get stuck give me a shout..

Rick

Hey Rick,

That's the plan, although I'm looking at nothing larger than a 4' x 4' useable area table so I can knock out the occassional guitar. I did read your post on your build log - I don't need anything that large as I still need somewhere to park the FLH and the Hinckley, but it sure as gave me plenty to think about, thankfully I can weld so I was going to construct the framing out of 2" square tube, finding the screws to suit that size. There are some on eBay, but they're too short so I'll keep looking.

[John S]

I have always had a soft spot for Hinkleys, - Romney Marsh.

Have you given any though about using timing belts as screws ? If not it's often cheaper in the long run to buy off Chia in China and at least you know what you are getting.

[Sobczyk32]

Hi Carlb1058
can you help me?
I need to know,what colors are on the resistor.
because mine is burned and can not see anything :(
Thank you

[Carlb1058]

Hi Carlb1058
can you help me?
I need to know,what colors are on the resistor.
because mine is burned and can not see anything :(
Thank you

I've just looked at the original huge image but because of the fact colours change a tad on PC monitors they look like the could be a pale blue with dark blue bands or perhaps a blue/grey with dark blue bands. I'll wander out to the shed a little later as we've just gone off daylight saving and it's stoopid o'clock here at the moment and physically check it and let you know.

[Carlb1058]

I have always had a soft spot for Hinkleys, - Romney Marsh.

Have you given any though about using timing belts as screws ? If not it's often cheaper in the long run to buy off Chia in China and at least you know what you are getting.

Timing Belts? Interesting concept, might have to wander the boards and look for a build log to see how that's achieved, although I would have thought the tensions would be against the use of belts because of the stretch factor?

[tumutbound]

Hi Carl,
Not sure if it's of any interest but I still have the three Yoocnc controller boards I removed from my 6040 when I replaced them with a Gecko G540.
You're welcome to them if you want them. One axis was a little flaky but I never tracked it down due to the controller update.
Drop me a PM if you want them and I'll get them in the mail.

[I was in Sydney but have moved to Tasmania]

[John S]

Hi Carl,


[I was in Sydney but have moved to Tasmania[/I]]

God heavens things must be bad if the Ozzies are deporting people now

[tumutbound]

God heavens things must be bad if the Ozzies are deporting people now

Tassie was one of the worst penal settlements but they allow people to leave these days - not that anyone would

[Carlb1058]

Tassie was one of the worst penal settlements but they allow people to leave these days - not that anyone would

Now this statement I must agree with - my younger sister bought a small farm just west of Launceston a few years back and loves the place. Sydney on the other hand of late is nowt more than a script for an LA gang movie and a subset for SOA :disturbed:

Appreciate the offer of the Yoo boards, mine has been removed and is currently holding down pieces of paper while I wait on a complete motor bob setup to come in the post, so I'll pass but thanks again anyway..

[Carlb1058]

Well that didn't take long, got my 3 x 542's and new bob ready to get set up but needed to knock out a couple of plaques before I tore it down and spent the next few weeks trying to work the wiring out. On the 2nd plaque it started to double line again so I thought the coupling had come loose, but no I have to be different and snap the bugger on the X axis completely *groan* gets wots ya pays for, now to find new DECENT couplings.

Info and piccies posted here http://www.mycncuk.com/forums/trading-experience-feedback/6031-best-place-buy-couplings.html hopefully it's the right board. . .

[John S]

Replied in that forum but it is the wrong place, only the original poster can post there. Everyone one else doesn't get a post reply button.

Anyway to keep it all in one place here's my reply.

This coupling question often gets asked.

There is only one design of coupling that can handle misalignment in two axis and still work. That is an Oldham coupling.

http://www.mekanizmalar.com/oldham.html

If you were in the UK I'd say Farnell, CPC, RS or Arc

The 2, 3, 4, jaw Lovejoy type couplings whilst being cheap are not zero backlash or able to handle misalignment in both axis.
The type you have I hate with a vengeance and is about as much use as a stick of celery and two hose clips.

REPLACE ALL THE COUPLINGS.

[Carlb1058]

Ta for the info - I'll see what I can find here but not holding me breath!! Oh and seeing how cheap n cheesy this one is I was going to replace all and double up too just in case.

Oh well, while I wait it'll give me time to read the wiring docs for this new setup *shudder*

[Carlb1058]

bonus a local Buy Oldham Style Coupling Hubs Clamp oldham coupler,8mm ID 25mm hub Huco 452H25.28 online from RS for next day delivery. (http://australia.rs-online.com/web/p/oldham-style-coupling-hubs/6930786/) roughly 120 quid for X,Y,Z

[tumutbound]

Also available at Farnell (http://au.element14.com/jsp/search/browse.jsp;jsessionid=21RU4T2ZZOWS2CWNBBSRFWI?N=0&Ntk=gensearch&Ntt=oldham&Ntx=mode+matchallpartial&exposeLevel2Refinement=true&suggestions=false&ref=globalsearch&_requestid=3954)(Element 14)

[John S]

bonus a local Buy Oldham Style Coupling Hubs Clamp oldham coupler,8mm ID 25mm hub Huco 452H25.28 online from RS for next day delivery. (http://australia.rs-online.com/web/p/oldham-style-coupling-hubs/6930786/) roughly 120 quid for X,Y,Z


?????????????????????????

Fück me when we shipped all the crooks out to OZ there was no need to keep all the bastards.

[Swarfing]

Christ? i paid £3.50 each off fish4parts, good while ago i know but can't believe they go up that much?

[Philly]

I use these cheap jobbies: 8x8mm CNC Motor Jaw Shaft Coupler 8mm To 8mm Flexible Coupling OD 20x26mm # 02 | eBay (http://www.ebay.co.uk/itm/8x8mm-CNC-Motor-Jaw-Shaft-Coupler-8mm-To-8mm-Flexible-Coupling-OD-20x26mm-02-/300697660000?pt=UK_Sound_Vision_Other&hash=item4602fa2a60)

even bought a spare set which I haven't had to use :)

[Carlb1058]

?????????????????????????

Fück me when we shipped all the crooks out to OZ there was no need to keep all the bastards.

ROFL how true - it's easy to get stiffed big time for plenty of different things here and the politicians wonder why people buy stuff O/S

[Carlb1058]

Christ? i paid £3.50 each off fish4parts, good while ago i know but can't believe they go up that much?

Checked a couple of local sites/companies and the prices are pretty similar so either they import container loads (doubt that) or the UK manufacturer sticks it to 'em on wholesale or they're all just twonks in it for the money with a captured market - either way I need something that's not gunna break at the sight of pine :apathy: and I would like to have a go at light aluminium one day too :bi_polo:

[Carlb1058]

I use these cheap jobbies: 8x8mm CNC Motor Jaw Shaft Coupler 8mm To 8mm Flexible Coupling OD 20x26mm # 02 | eBay (http://www.ebay.co.uk/itm/8x8mm-CNC-Motor-Jaw-Shaft-Coupler-8mm-To-8mm-Flexible-Coupling-OD-20x26mm-02-/300697660000?pt=UK_Sound_Vision_Other&hash=item4602fa2a60)

even bought a spare set which I haven't had to use :)

I was gunna buy arf a dozen of these for emergency's 8mm X 8mm CNC Flexible Jaw Spider Plum Coupling Shaft Coupler D 25mm L30mm | eBay (http://www.ebay.com.au/itm/280836905281?_trksid=p5197.c0.m619)

[Jonathan]

You could get timing belts and pulleys instead...

[Carlb1058]

You could get timing belts and pulleys instead...


Thanks for the suggestion but I have enough problems getting my 'ead around the machine as is :wink:

[Patr1ck]

Hi all, this is my first post here, I just registered to say I have the same machine with 4 axis I bought from Ebay.

First of, I want to say the value of the resistor is 100 OHMS 1/4 watt to the person that asked, as his was burned out.9049905090519052

Second. I have the same driver board that drives X Y Z (YOOCNC-NT65-3X) It's a piece of crap. There isn't even any DIP settings for switching micro stepping etc. My fourth axis is actually a YOOCNC also, but is a red PCB and it has the same driver chip "TB6560AHQ" but it has the DIP settings. I had major problems with my X axis with a 0.9mm difference when returned to X=0 Y =0 after milling a doubled sided PCB with small traces for SMT and drill holes etc. So you see 0.9mm loosing steps on the X axis was a critical problem. I'll upload picture of the 4th axis driver board!

The initial problem was the settings when I got the machine first. The manual states "400 / 2000 / 200 / 3 - 1 " all of this is perfect, but the Velocity speed which states 2000. I have mine tuned down to 900 and I am running the Kernel @ 25Khz instead of the Manuals stated 35Khz or on the main BOB the motors stall, loose steps, vibrate etc.

Now I switched the X axis stepper over to the extra board (RED PCB) I was able to set the steps with the DIP settings on the board, and I don't loose steps, the stepper is running really smooth with little to no noise from it at all compared to the other steppers on the main BOB. Y and Z have high pitched sounds coming from them when in use, but now with the new DIP settings on the x axis I can run it @ 2000 in velocity @ 35khz kernel with out loosing steps, vibrations, and the motor just hums nicely.

So to some it up, I am getting rid of this YOOCNC breakout board and replacing them as soon as I have the dosh.

The machine it's self is nice and well worth the money, the controllers are shite !

Here is the one I bought in March I think.

220V CNC 3040Z DQ Upgrade From 3040Z 3040 Router With 230W 11000rpm Spindle | eBay (http://www.ebay.ie/itm/220V-CNC-3040Z-DQ-upgrade-3040Z-3040-Router-230W-11000RPM-spindle-/221236714378?pt=US_Video_Gaming_Replacement_Parts_ Tools&hash=item3382bc5b8a)



The CNC gantry system is good. The controllers are bad, and hell, the controllers could even be a clone of the already crappy controllers from YOOCNC.

Questions welcome, and anything else. I just thought I would share my thoughts and frustration and my how I fixed my issues here.

On another note, I am replacing all the boards except the transformer and other simple stuff to a Pokeys 56U + Postep25 motor drivers for Mach3 when I got the dosh.

Polabs (http://www.poscope.com/)

My software for initial use:

Diptrace
Vcarve Pro
Mach3
and shitty YOOCNC controllers, plus all cables leading from controller to steppers are unshielded !

Also I replaced the main CAPS on the BOB to Nichicon LOW ESR just incase. The IC chips (74HC***) should be really 74LS*** so that they can properly use the TTL logic from the parallel ports that don't support 4.5vcc. Some ports on computers actually output 3.3 I think or something in the that line.

Using the LC version means that they will support all types of parallel ports.


regards

[Carlb1058]

I replaced the yoo board with 542H's, actually to save time the only bits used from the original controller box is the E-stop, spindle speed knob and the on n off switch - the machine itself however apart from getting three Oldham couplings, ta v much to whoever it was that suggested them :encouragement: - sorry I can't remember who though *eep* is pretty much as it was out of the box oh and I modded the job clamps.

It's damn near paid for itself already so I'm well pleased with it. I Was gunna go buy three new nema 23's but they have from memory bolt holes that are 41mm <--> the bolt holes on this machine are 50mm <--> so the only way to have spares is to get the nema 34's which in the ad says have the same 50mm bolt holes? but are more expensive!!

Back to confusion again :witless:

[JAZZCNC]

I Was gunna go buy three new nema 23's but they have from memory bolt holes that are 41mm <--> the bolt holes on this machine are 50mm <--> so the only way to have spares is to get the nema 34's which in the ad says have the same 50mm bolt holes? but are more expensive!!

Back to confusion again :witless:

If they are 50mm centres then It's got Nema 24 motors on it which are 49.8mm bolt centres where has Nema 23 are 47.14. Nema 24 motors will measure roughly 60mm outer dimensions and Nema23 56.4mm.
Nema 34 are 70mm centre holes. They would also be a big mistake to use with M542 drives and give much lower speeds.

[Nikos]

Hello,

I'm new on this forum.

I have the same CNC, 3020Z-D52.
I bought the Gecko G540 to replace the stepper controller because is losing steps and I would like to inform me if I could use the Spindle controller of the D52 to connect it via PWM in VFD output of the G540.

Thank you in advance
Nikos