PDA

View Full Version : BUILD LOG: First Build 5 x 12 Steel Frame CNC router



Scott Damman
03-03-2016, 05:22 PM
First I would like to say the information here is amazing and has greatly helped my design and learning for this CNC router build.

Objectives,

5 foot by at least 10 ft of machining area
Z height high enough for paddleboard foam blanks ~ need roughly 12 inches under the tool
ability to machine wood, aluminum, foam
good accuracy/precision (still trying to figure out where I want/need to be)
Vacuum hold down


Parts I already have
Linear rails - INA 25mm 4 row linear rails - I have enough to do all axis's and enough bearings
Y axis ballscrew - 10mm/rev rolled ballscrew
1850 oz/in stepper motors - 2 for X (long axis) and 1 for Y
2 4hp regenerative vacuum blowers
Colombo 10hp spindle with ISO30 ATC

I will start out where I am now and then go into some questions I have.

The gantry legs mock-up 4 x6 x 1/4wall box tubing, 1/2 inch plate for the feet.
17792

17793

17794

Gantry partially welded up, Gantry is built from 4in x 6in x 1/4in wall box tubing and will get blocking in between the 2 tubing with some vertical bracing on the back. I machined the ends of the gantry before welding them up so I can put an e-stop on each end. The ballscrew is laying ontop of the gantry and will live between the 2 box tubing when complete.

17795

Here is the plan for the Z axis, made from 3in x 6in x 1/4wall box as a starting point. I plan to make an A axis and have mounted the motor up and off to the side. Although this could still get in the way. Should I put the motor for the A axis above the spindle motor? This will make the HTD belt 20 inches center to center. If I did that would I be looking for trouble as far as belt stretch is concerned?

17796
17797

Heres what I was thinking for A axis

https://youtu.be/Lq3Ei8oALXY

A few questions,
Gantry and Z axis is going to be ~550lbs (~250kg), am I shooting myself in the foot using steppers for this weight? I would like to get decent speed out of the thing without the risk of missing steps. When I purchased the steppers I was trying to balance cost but now im worried I will have trouble with it when im done.

What pitch should I shoot for on the Z axis screw? I was planning on using the stepper motor with a ~4:1 reduction. Total travel as it stands right now will be 16 inches. Weight of Z with Spindle will be around 100lbs should I add spring assist to this? or air springs etc?

Next steps for me,

I am working on mounting the rails on the Y Axis and getting prepped for pouring epoxy as soon as the temp rises a bit. Here is the plan for the pour and looking for any insight.

Lay gantry on its back
Level with starrett precision level as close as I can in both planes
pour epoxy for Y axis linear slides
at the same time pour the mount pads for the Y Axis ballscrew, in my mind this gets the ballscrew on the same plane (albeit at a different height) as the linear rails
Paint Gantry
Mount the lower linear rail for the Y Axis
Flip gantry upside down
Use precision level on the linear rail and set rail level and also use a precision square and the level to level in the other plane
Pour expoxy on gantry feet (my thinking here is that I can then get the gantry feet 90 deg to the Y axis linear rails and also on the same plane as the linear rails) I would assume there may be some slight shimming still but the epoxy will be close.


Thanks
Scott

Scott Damman
03-03-2016, 09:04 PM
Some updates on the gantry

All of the bracing clamped in place ready for spot welding. The ballscrew is in place to set the distance between mounts.
17802

17803

Here is one of ballscrew mounts
17804
17805
17806

this is the end of the gantry showing the space for the estop
17807

This is one of the longer braces that connect to the ballscrew mount block

17809
17808

Here is the backside with all the bracing welded up
17810

And the front side, notice I drilled some 1/2 in holes on the wrong side. My linear rails are threaded from the backside so I have to through drill from the back. The good news is I am welding on 3/8 plate (I drilled the holes to the correct spacing on the mill first and used the 3/8 plates as templates) on the front side where the linear rails are going to mount.
17811

sinnsvak
03-03-2016, 11:13 PM
Very impressed, with a lot of things. Very excited about your A-axis, sometimes one should just go to ahead and build it instead of pondering if it will work or not :D I have not attempted to build it yet, as I dont even have an idea of how I would use it. I really have no clue how one would incorporate 4 axis programming.

I would probably fit the widest possible belt there, if I were to use belt, you want the spindle to be rigid enough. What kind of bearings, if that is how you are pivoting the A-axis? Got any detailed view of it?

Nice work so far.

Scott Damman
04-03-2016, 03:07 AM
What kind of bearings, if that is how you are pivoting the A-axis? Got any detailed view of it?

Nice work so far.

Thanks for the comments

Heres a view of the bottom of the Z axis and a section cut into it.

17814

17815
**sorry the drawings are upside down**

17816
Here is the box tubing removed

The bearings are roller bearings that I can get from McMaster Carr easily. 1 1/4 diameter shaft. Bearings are designed to be held into aluminum housings bolted into the 3 x6 steel tubing. The shaft is a 6 inch threaded on both ends. Again something I can just order from McMaster. The bolts into the shaft allow me to put preload into the bearings. I was thinking of bolting the pulley hub (pulley is not pictured) to the spindle mount.

Open to thoughts on this design. I think it will work....

Scott Damman
04-03-2016, 06:51 PM
A couple more updates sorry for the bad pictures
Drilled backside holes out to 1/2 inch to allow bolt and ratchet to be put through from the back. Im thinking I will just order a bunch of plastic plugs to put into these holes once I am done.
17818
17819

I threw a couple bolts into one of the linear rails just to see how things were coming together.
17820
17821

So far I am pretty happy with how the build is coming together. Time will tell if I have some precision/accuracy when im done...

Here is a ~bad video of me test running the used colombo 10hp ATC spindle that I got. Running it at ~18,000rpm after running at ~8,000rpm for 10 mins to get the grease moving in the bearings.

https://youtu.be/yyiDn-8hFaE

Boyan Silyavski
05-03-2016, 08:55 AM
Very ambitious for a first build.

Some thoughts:

-the gantry legs dont seem adequate for the rest of the parts you will use on your machine

-the A may work for foam but i dont see it serious for anything more than wood

-the stepper motors , ok- the huge stepper motors, what drives will power them? 230VAC?

-that bracing may be was totally unnecessary. I dont see how that will strengthen anything. It could even make things worse, strengthening one point too much and not allowing the forces to spread uniformly and creating pressure points. So later all rings.


What will you be using this machine for? Remember- the weakest point is the criteria for how strong your machine would be. Again-the gantry legs.


PS:
That gantry legs will greatly benefit from bolting or welding on them 10mm steel plates in both directions, 2 per leg in total

JAZZCNC
05-03-2016, 11:05 AM
Hi Scott,

This is how I see couple of not so little things.

Steppers:
Really Cannot comment if they will be good enough because you haven't said anything about what your using for linear motion on the X axis. I assume Rack n pinion but what ratio etc.? . . . But let me put it this way 550lbs or put anohter way 1/4 Ton is lot of Mass to Accel and stop so better get this correct or you'll seriously regret using steppers.


A Axis: You need to think again on this I'm afraid NO WAY will Stepper with such small ratio hold that Big spindle at angle without moving or vibrating.
Remember the Z axis is the MOST important area on machine so weaken this and all that OTT work you have done on the frame is pointless.

Scott Damman
05-03-2016, 02:06 PM
Very ambitious for a first build.

Some thoughts:

-the gantry legs dont seem adequate for the rest of the parts you will use on your machine

-the A may work for foam but i dont see it serious for anything more than wood

-the stepper motors , ok- the huge stepper motors, what drives will power them? 230VAC?

-that bracing may be was totally unnecessary. I dont see how that will strengthen anything. It could even make things worse, strengthening one point too much and not allowing the forces to spread uniformly and creating pressure points. So later all rings.


What will you be using this machine for? Remember- the weakest point is the criteria for how strong your machine would be. Again-the gantry legs.


PS:
That gantry legs will greatly benefit from bolting or welding on them 10mm steel plates in both directions, 2 per leg in total


Hi Scott,

This is how I see couple of not so little things.

Steppers:
Really Cannot comment if they will be good enough because you haven't said anything about what your using for linear motion on the X axis. I assume Rack n pinion but what ratio etc.? . . . But let me put it this way 550lbs or put anohter way 1/4 Ton is lot of Mass to Accel and stop so better get this correct or you'll seriously regret using steppers.


A Axis: You need to think again on this I'm afraid NO WAY will Stepper with such small ratio hold that Big spindle at angle without moving or vibrating.
Remember the Z axis is the MOST important area on machine so weaken this and all that OTT work you have done on the frame is pointless.

I think you are talking something along these lines for the sides of the gantry legs? I was wanting to do FEA on these legs but never got around to it, I did however do calculations (free body diagram) to try to calculate total deflection based on 1000lb force at the bit and keeping the total deflection ~3 thou at full load.

I could go higher with the bracing as well this is just a quick drawing.
17835

The thinking behind the bracing was to connect the upper and lower gantry beams. I used box tubing between the beams. Now the rear bracing, the thought there was to reduce the twist in the gantry. Hopefully it doesnt cause issues...

So the A axis, this is exactly what I am worried about, holding it in place when at angle and also moving it. The A axis would really only be used when machining foam blanks for paddle boards. However since I want to be able to machine wood, and aluminum as well then I really need to make sure this thing is right.

Couple things I am thinking thoughts are greatly appreciated.
- Eliminate A Axis...
- greatly increase the belt ratio
- keep A Axis but have an air operated taper pin that holds it at 0 when machining wood/alum
- look into a harmonic drive rather than belt drive

X drive
I was planning on using rack and pinion with a pinion of 1 inch I was going for 4:1 ratio to the pinion.
so 3.14 inches per revolution of the pinion, 0.785 inches per motor revolution, works out to 0.0039 inches per step. I was thinking of micro stepping them x10 = 0.000395inch per micro step.

I am brand new to this and trying to learn what feed and rapid movements I need to shoot for to cut wood/aluminum properly. I know it depends on the number of flutes on the bit etc. Am I crazy to think I should shoot for 1000ipm?

The motors I have are chinese 1841 oz/in with MA860H drivers along with 60V drives. Based on the questions it sounds like these are not adequate?

Servos
So if scrap the steppers for X axis then im looking at servos. Something I am even more NEW to than steppers. I need to do more learning on how to size them please correct where I have errors!

Things to keep in mind when sizing?
Gear ratio - want to keep the servo motor spinning at higher speed to stay in its power range - unlike steppers which have the most power at 0 speed
Load to motor inertia, this one I have seen 10:1 ratio but I dont really understand how to use that

One thing I dont understand, when X is not moving what is holding if from moving? I had assumed that the steppers hold the load from moving at 0 speed (is this the case with steppers?). If that is the case what is doing the holding when using servos?

Can my router have both steppers and servo motors? I think the stepper would still be adequate for the Y Axis and could use one for the Z axis (as Im sure I can not send them back)... I plan to use ethernet smooth stepper with mach4.

Got the X axis steel drilled these needed to be through drilled as well since the linear rails are bolted from the back. 244 holes!!!!
17836
17837

Any help with servos for this size of gantry is greatly appreciated.

mekanik
05-03-2016, 04:54 PM
Hi Scott
I don't have any experience of the 4/5 axis machines but i would imagine that the "A" axis you are trying to emulate would be a wheel & pinion mechanism.
Regards
Mike

Boyan Silyavski
05-03-2016, 06:20 PM
Ok, lets start from the beginning. I will try to resume some things, you just need to further research the topics:

1. You should absolutely know what weight your gantry is with all things mounted. Knowing the mass of the gantry will be decisive on what motors, what size ball screw, rotating nut, Rack and pinion, etc..

2. Be realistic about the feeds and speeds. And what you really need the machine for. All in one would be usually a disaster in some area.

3. Motor sizing. People drive 3nm nema 23 motors at 70VDC. And we are talking low inductance motors. Expected reliable performance from such motors coupled with xx10 ball screws is 10000mm/min. The motors you have are high inductance motors (http://www.cncrouterparts.com/why-should-i-buy-your-motors-company-xx-has-higher-torque-motor-p-187.html). I am not a specialist here but there are a lot of knowledgeable people here that will tell you at what voltage, but i am sure that it must be more than 110VDC. Do your research and find at what voltage these motors perform well. And if these drives will be ok.

4.Without knowing the total weight of the gantry... but i see you start ambitious, so yes, you will need servos to drive the gantry reliably fast if you aim at 25000mm/min. And you will need rotating ball nut or rack and pinion

5. There is a very well made calculator at the forum here, search for it, and i am sure that the gantry legs deflection will not be as you say...but may be i am wrong




I dont want to discourage you, but i am having the feeling that if you continue without thinking it too much, there will be disappointments on the way. WHY JUST DONT YOU SLOW A BIT AND READ MORE BUILD LOGS. All the info is there.

And by the way it seems the american thing is to desire rapids of 1000ipm. And make rack and pinion machines. We here in EU prefer precise ball screw machines and rotating ball nuts for big machines and heavy gantries.

routercnc
05-03-2016, 10:27 PM
Hi Scott,

Just to scale point #3 from Boyan - 3Nm Nema 23 at 70VDC could be used up to gantry weight of 50-70kg tops. As yours is 5 times this weight the stepper / driver combo needs careful thought.

I can sort of see the machine you are trying to make by piecing all the photo together - and it looks like a commercial wood router with raised gantry.

If you want to cut aluminium as well I would suggest you chop the legs off the gantry, and go for a raised X axis instead. Lift the X axis members up on supports away from the bed until they meet the bottom of the gantry. This makes a much stiffer machine for little effort. Your span is quite wide so you will need all the help you can get for aluminium.

Can you post up a picture of the overall machine design?

Scott Damman
05-03-2016, 11:12 PM
Ok, lets start from the beginning. I will try to resume some things, you just need to further research the topics:

1. You should absolutely know what weight your gantry is with all things mounted. Knowing the mass of the gantry will be decisive on what motors, what size ball screw, rotating nut, Rack and pinion, etc..

2. Be realistic about the feeds and speeds. And what you really need the machine for. All in one would be usually a disaster in some area.

3. Motor sizing. People drive 3nm nema 23 motors at 70VDC. And we are talking low inductance motors. Expected reliable performance from such motors coupled with xx10 ball screws is 10000mm/min. The motors you have are high inductance motors (http://www.cncrouterparts.com/why-should-i-buy-your-motors-company-xx-has-higher-torque-motor-p-187.html). I am not a specialist here but there are a lot of knowledgeable people here that will tell you at what voltage, but i am sure that it must be more than 110VDC. Do your research and find at what voltage these motors perform well. And if these drives will be ok.

4.Without knowing the total weight of the gantry... but i see you start ambitious, so yes, you will need servos to drive the gantry reliably fast if you aim at 25000mm/min. And you will need rotating ball nut or rack and pinion

5. There is a very well made calculator at the forum here, search for it, and i am sure that the gantry legs deflection will not be as you say...but may be i am wrong




I dont want to discourage you, but i am having the feeling that if you continue without thinking it too much, there will be disappointments on the way. WHY JUST DONT YOU SLOW A BIT AND READ MORE BUILD LOGS. All the info is there.

And by the way it seems the american thing is to desire rapids of 1000ipm. And make rack and pinion machines. We here in EU prefer precise ball screw machines and rotating ball nuts for big machines and heavy gantries.

The gantry will weight around 550lbs, this number comes from the 3d drawing which worked out to 494 (this included the Z axis and the spindle) and I added 50 lbs for cables and cable routing. This was my bassis for sizing the motors. Sorry I thought I had been clear on it earlier in my thread about the weight.

Feeds I am happy to be realistic and dont care much about super fast rapids but I want to be able to drive the machine at proper cutting speeds.

Im sorry I came across as I jumped into this without any research when the opposite is true, I have done lots and lots of research both on forums and by inspecting commercial machines. I would not have been able to get to the point I am without lots of learning from all sources. I find there a very different views on most subjects associated with the gantry style cnc router so its hard to nail down whats the best design. And thats where some of my questions came from. The stepper route was more of a cost thing and I should have done more research on servos before going down that road.

I will find the calculator you speak of and give it a try. I just started out by doing the math, the hard part was actually finding out how much force would be applied at the tool so I could back calculate the flex at the gantry.

**I want to say machining Aluminum is a bonus for this machine as it will mainly be used for wood**


Hi Scott,

Just to scale point #3 from Boyan - 3Nm Nema 23 at 70VDC could be used up to gantry weight of 50-70kg tops. As yours is 5 times this weight the stepper / driver combo needs careful thought.

I can sort of see the machine you are trying to make by piecing all the photo together - and it looks like a commercial wood router with raised gantry.

If you want to cut aluminium as well I would suggest you chop the legs off the gantry, and go for a raised X axis instead. Lift the X axis members up on supports away from the bed until they meet the bottom of the gantry. This makes a much stiffer machine for little effort. Your span is quite wide so you will need all the help you can get for aluminium.

Can you post up a picture of the overall machine design?

I looked carefully at doing a bridge style machine with high sides, the problem that kept running into this design was not easily being able to side load the machine. I do alot of work with large wood slabs which this machine will be used to flatten and I need to be able to side loads these as I dont have an overhead crane. (they weigh in excess of 1200 lbs)

Here is the Frame starting, I have blocked up the frame to level and used the gantry to set the exact distance.
17845

Initially checking square with a precision square, final squaring will need to be done once its all welded up and before pour epoxy.
17846
17847

Scott Damman
05-03-2016, 11:51 PM
Alright so I ran the numbers through the excel spreadsheet.

Heres what I got
17848

0.010mm deflection
0.00028 deg delfection
deflection at tool 0.002mm
17849

0.00094mm deflection

Maybe I am entering the numbers wrong I will run through the numbers again but these look better than what I got when I just did the free body diagram. Thoughts on these?

These seem acceptable but im not sure what I should be shoot for as far as max deflection?

Thanks for any help. Want to make sure this is right before I get too far that I cant make changes.

routercnc
06-03-2016, 09:40 AM
Hi Scott,

I wrote that spreadsheet some time ago to compare simple shapes. For the numbers you have entered it looks like it is giving the correct answers.

Things to note:
It assumes everything else (e.g. spindle, tool, bearings, bedplate, joints . . . .) is infinitely stiff - so it is best case
Have you entered just one of the two gantry beams? If so the real stiffness will be higher as you have 2
Or have you given the overall gantry dimensions as a single beam? If so the real stiffness will be lower
It makes no prediction for deflection due to cutting vibration and machine resonance

It terms of max deflection it is personal choice but I would say wood <0.05mm and alum <0.01mm as a starting point. Others opinions welcome!

I get the point about raised gantry vs raised X axis (in the US this can be Y axis) for material loading, so as long as you've thought about the options and made a choice that's fine. Didn't know what you knew.

In summary you have a very stiff machine that is stiff enough for wood, but that gantry is heavy (hence our concern over stepper/driver/voltage - if you know these are OK then that's good), and resonance is unknown (the sheet can't show this).

Ger21
06-03-2016, 03:20 PM
Is this what you have?
http://www.ebay.com/itm/3-Axis-CNC-Kit-1841oz-in-Nema-34-Stepper-Motor-MA860H-Driver-CNC-Plasma-Mill-/231424600113

If so, you're better off re-selling them and taking a loss.

A quality 300oz motor driver kit will typically cost more than that one does. What does that tell you? Do you really think that you're getting 6x the performance for the same price?

As Boyan mentioned, those are high inductance motors. Ideally, you want an inductance of around 2mH. Those motors really need about 100-120V to get the best performance from them. But even then, they won't spin nearly as fast as a much smaller motor.
You generally won't find large motors with low inductance, because they would require very high currents, and very expensive drives.

A few other things to consider. As steppers get larger, their maximum rpm's tend to decrease. I wouldn't be surprised if you can't get much more than 300rpm from those before they run out of torque. Which probably equals around 250ipm with your setup?

Also be aware that you often can't count on the microstep accuracy. Especially with large motors, where the microstep torque may not always be able to overcome the high detent torque those large motors have. When calculating resolution with steppers, 1/2 step is the best that you should hope to achieve.

Personally, I think a machine like that would really need something like 1Kw AC servos. Expect to spend at least $2000-$2500 for three.

As for cutting speeds. I run a Morbidelli at work, and we cut at 1000+ipm every day. Rapids much faster. But this is cutting sheet goods, plywood, MDF, particle board.
Hardwoods are not cut as fast. I typically will cut hardwoods at 300-500ipm with 1/4" and 3/8" bits.

If you're not cutting at these speeds, then you're not taking advantage of your machine and spindle.
One thing that most people don't consider, is that it can be very difficult to hold materials in place at high cutting speeds. My hardwood speeds are limited by my vacuum holding power. If the hardwoods are not held tightly enough, cut quality will suffer, and you'll need to cut slower to get acceptable quality. Slower = shorter tool life.

JAZZCNC
06-03-2016, 03:47 PM
Ok the bottom line is that large steppers need Mains voltage Drives to get any speed from them. Like Ger says 110v will only get you limited speed.

If using R&P then you 100% will need Servos because they are so much less efficient than ballscrews so with heavy gantry like this you will need really high torque motors and Large steppers just won't give the speed you'll need.

Servos from China are now so low that it's No brainer to me that you need to look east.! Approx $400 per set drive/mootr/cables.

However be aware that you'll also need match the rest of the electronics to match so you'll need an external motion control card with high pulse frequency to get the full speed/smoothness of the servos.

Scott Damman
07-03-2016, 01:38 PM
Gerry, those are the ones, I have come to that conclusion as well. I think I will re-list those and find some servos.

Now do I keep a stepper for the Z (a smaller stepper not these big honking ones)? I would lean toward 2 servos on the X and 1 on the Y. Then I am thinking a smooth stepper in between it looks like that is fast enough to run the servos...

I know that my table is still going to be vacuum limited even with 8hp of regenerative vacuum. I plan to have 4 zones and also thinking about ways to include T-slot for hold down of hardwood parts.

Jazz, Im coming up empty on sets that are $400 im finding ~$600 for 1kw AC servo with driver. If you have any links to what you are talking about that would be greatly appreciated.

Im really leaning toward running a powered ballnut/ballscrew setup for the X, like Boyan said most builds here in the states use R&P but the 0 backlash and less wear aspect of the ballscrew is really appealing. Learning as much as I can from this forum and how other people implemented it. Im sure I will have questions...

A few updates, I dont love the look of the legs they look out of proportion to the rest of the machine but its steel that I already had on hand from other projects and its plenty strong with bracing. It gets a 3rd leg just havent gotten that far yet. Lots of bracing. I am trying to keep the one end open so I can use some of the space under the machine for rough sawn lumber storage...
17870
17871
17872

Thanks for all the help so far! opened my eyes to some issues.

Just for fun I took a quick vid of me manually moving the gantry, its amazing with the linear guides allow...~move 500lbs with just fingers.

https://youtu.be/OTUE6K_w3Zk

Ger21
07-03-2016, 02:34 PM
I'm building a 4x8 machine and using AC servos, and I'm using steppers and Leadshine AM882 drives for the Z axis (two, it's a dual Z machine) to save money.

I would think that a 5mm pitch ballscrew and a 600oz stepper would be OK for your machine? That's just a guess, though.

As for $400 servos. I've seen 400W servos on AliExpress for around $400, but not larger ones. And at that price, you may not know exactly what you're getting. I'm paying around $575 for 400w Leadshine motors and drives for my machine. 750W and 1Kw are closer to $800 a set. The price difference between 750W and 1Kw is very small, so I'd definitely go with larger ones for a little more money. http://americanmotiontech.com/motor-drive-sets/ac-servo/high-voltage-ac-input/

As for vacuum, you can never have enough. We have two 10HP Becker's on our Morbidelli with a 5x12 table, and could always use a bit more.

Boyan Silyavski
07-03-2016, 02:43 PM
Our friend Fred from BST Automation has 1kw servo motor + drive (http://www.aliexpress.com/store/product/New-Servo-motor-and-driver-set-4N-M-1KW-2500RPM-90ST-AC-Servo-Motor-90ST-M04025/314742_761020438.html) for 332euro +shipping which will cost you in total~400$ for the combo when you buy 4.


I am not sure you need exactly 1kw if you go with rotating ball nut. But as prices are similar. I think you will need 0.75kw geared like mine 20t:30t and that will give you speed of 20000mm/min on the long axis/ball screw with rotating nut/.

Scott Damman
07-03-2016, 08:10 PM
I'm building a 4x8 machine and using AC servos, and I'm using steppers and Leadshine AM882 drives for the Z axis (two, it's a dual Z machine) to save money.

I would think that a 5mm pitch ballscrew and a 600oz stepper would be OK for your machine? That's just a guess, though.

As for $400 servos. I've seen 400W servos on AliExpress for around $400, but not larger ones. And at that price, you may not know exactly what you're getting. I'm paying around $575 for 400w Leadshine motors and drives for my machine. 750W and 1Kw are closer to $800 a set. The price difference between 750W and 1Kw is very small, so I'd definitely go with larger ones for a little more money. http://americanmotiontech.com/motor-drive-sets/ac-servo/high-voltage-ac-input/

As for vacuum, you can never have enough. We have two 10HP Becker's on our Morbidelli with a 5x12 table, and could always use a bit more.

Gerry do you have a forum for your build? I tried to do a search and didnt come up with anything.

The vacuum does worry me, I may have to add more or sell these I have an get a single larger unit... But we will see I think for now I will be able to work around the vacuum pumps that I have and see how things are.



Our friend Fred from BST Automation has 1kw servo motor + drive (http://www.aliexpress.com/store/product/New-Servo-motor-and-driver-set-4N-M-1KW-2500RPM-90ST-AC-Servo-Motor-90ST-M04025/314742_761020438.html) for 332euro +shipping which will cost you in total~400$ for the combo when you buy 4.


I am not sure you need exactly 1kw if you go with rotating ball nut. But as prices are similar. I think you will need 0.75kw geared like mine 20t:30t and that will give you speed of 20000mm/min on the long axis/ball screw with rotating nut/.


Ok I had came those on ali but like Gerry said I had no idea what I was looking at or what I would be getting. So having someone point them out is very helpful. For the cables, do you just make up longer cables or order specific lengths? 3m wont make it for the far side X motor. How have you found the shipping to be? I see it says 23 days... I wonder if thats pretty accurate or if they come faster than that.

The other motors/drives I came across was dynamic motor motion technology.

I need to dig deeper into your build Boyan and see how to built those ballnut drives. I found a few options online and on the forum here for how to build them.

Have your guys seen these
http://www.stober.com/motion_control/EZM_series/

I havent priced them and I know they cant be geared so that has to be taken into account but seems like a great way to run the ballnut.

I hope to have a plan here in the next couple days then I can run it by you all for advice before ordering everything up...

Boyan Silyavski
07-03-2016, 08:24 PM
BST automation is the usual choice for Chinese stuff and linear rails, so its a well known and recommended seller. Price wise, packing, support and warranty.

Do you have a cable chain? If not i highly recommend the Russian ARKO 22x80 R60 (http://www.cable-tracks.com/) which is cheaper and better than the similar Chinese

Ger21
07-03-2016, 10:19 PM
Leadshine has 10m cables available for their motors.
I would email BST and ask if he has them for his motors. Or, you can always make your own.

At those prices, I'll have to seriously consider going that route, if it saves $225/motor.

I have a "build log" at CNC Zone, that's about 7 years long, with almost no actual building, and another 2 year old thread when I started the design.
I've got a few thousand dollars in components, but life doesn't leave me any time to get to it.
I also spent the last 3 years of my spare time working on this: www.jointcam.com
(Actually 2 years, and a year of doing nothing)

JAZZCNC
08-03-2016, 12:25 AM
BST will provide cables upto 7mtr long. This is Video I made showing there 1.8Kw set.


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

mekanik
08-03-2016, 11:26 AM
Hi guys sorry to but in but could someone post a link for BST Automation.
Please
Regards
Mike

AndyGuid
08-03-2016, 11:33 AM
BST Automation = http://www.aliexpress.com/store/314742

Clive S
08-03-2016, 11:34 AM
Hi guys sorry to but in but could someone post a link for BST Automation.
Please
Regards
Mikehere you go http://www.aliexpress.com/store/314742 but you can email him direct bstlinear [[@ ]] gmail.com (bstlinear@gmail.com)

Scott Damman
08-03-2016, 03:17 PM
BST automation is the usual choice for Chinese stuff and linear rails, so its a well known and recommended seller. Price wise, packing, support and warranty.

Do you have a cable chain? If not i highly recommend the Russian ARKO 22x80 R60 (http://www.cable-tracks.com/) which is cheaper and better than the similar Chinese

Awesome, those are good prices, I need to see how much shipping is going to be but that was good timing cause I have not purchased chain yet.

Jazz, thanks for the video exactly what I was looking for in a review. Time to order some motors and drives... I emailed to see if I could get a price for motor/drive and 7m cable set.

So the resolver feedback into Mach is awesome, shows how new I am to this I didnt even know that was possible. Anyone know if the smooth stepper from warp9 can do this? It shows resolver connections on the board but does it send back to mach? Also looking at the CS lab boards as well since Jazz has a good review on them. Im thinking I might go for a servo for Z as well since that can give me position feedback into Mach rather cool feature.

Gerry, I had started a build log over at CNC Zone but after finding the great information here I decided to put my build log here instead. Also for me the cnc zone website is painfully slow!

Currently using the rotating ballscrew design to design one that I can build with the machines I have. Might be a little crude, also sending out designs to some local cnc metal shops for quotes on parts. One question, (of many im sure) do you guys find you need to grease the ballnut more since it is sometimes spinning rather fast?

Thanks for all the help, I will post some design ideas shortly for the rotating ballnut.

JAZZCNC
08-03-2016, 03:49 PM
Anyone know if the smooth stepper from warp9 can do this? It shows resolver connections on the board but does it send back to mach? Also looking at the CS lab boards as well since Jazz has a good review on them. Im thinking I might go for a servo for Z as well since that can give me position feedback into Mach rather cool feature.

Yes and No. The ESS can send the encoder signals back to Mach but only for display purposes and to different set of Encoder DRO's on separate screen. (or can add to Custom screen if you like) Mach will not know anything about positional error and the Main Work DRO's will not update if position moves.

To work like you see in my Video the Controller deals with the Encoders and informs Mach of position directly. The Closed loop between motor and drive is then extended to the controller which keeps track of position and updates Mach's Main DRO's. The ESS cannot do this.

Very Few Step and Dir Controllers do this. Even the CSlabs IP-S which uses STEP & DIR cannot do this. Only the Analog IP-A Version does this.

Ger21
08-03-2016, 05:43 PM
One question, (of many im sure) do you guys find you need to grease the ballnut more since it is sometimes spinning rather fast?

On our Morbidelli, it's recommended to grease the spinning nut every 40 hours of run time. linear bearings and non spinning nuts get greased every 3 hours of run time with an auto lube system.

But a lot more grease goes into the spinning nut, as it's much larger.

Scott Damman
09-03-2016, 03:30 PM
So Im looking into and learning more about long ballscrews and how to run them. Obviously spinning the ball nut is a way to go but each unit gets to be pretty costly by the time you are said and done.

Another thing I found was supports along the ballscrew which effectively push the critical speed to a higher rpm. Has anyone ever looked into this type of design?
http://www.linearmotiontips.com/achieve-long-travel-high-speed-ball-screws/

Thermwood does this on the long axis of some of their CNC machines. It looks like the support flips as the ballnut passes and then supports the ballscrew again on the other side of the ballnut.
17899

I assume these are just nylon but cant find any further details on this type of support. I wonder if it could be considerably less expensive to incorporate this, provided the same performance is achieved in the end. Thoughts?

So still learning about the CSLabs IP-A and trying to decide if the extra money is worth it. With all ballscrews, I dont know that there will ever be a time that the system could move while the drives are disabled? I.e. I would have to manually turn the ballscrew or the motor to put it out of position. However, if Mach knows absolute position does that mean that the system needs to be "homed" less often? Between cuts etc. Are there other benefits to having Mach know the absolute position from the resolver? I hope to make a decision here soon on a motion control card... If anyone has any further thoughts I greatly appreciate it.

Thanks to Fred over at BST for working with me, I have ordered motors, 2 for X, 1 for Y, and 1 for Z. I decided to go with servo on Z as it seemed silly for the price difference to have that axis be a stepper.

Next up is HTD timing gears/belts just need to figure out speeds and gear ratio. Thanks Boyan for posting the 20t:30t Starting point and after putting it on paper sure makes sense to me.

So some math just to make sure I have things right
Lets say 20000mm/min travel
Ballnut is 5mm per revolution

(20000mm/min ) / (5mm/rev) = Ballscrew rpm => 4000rpm

Gear ratio 20t:30t = 0.666

4000rpm * gear ratio => Motor Speed => 2666rpm.

The motors I purchased from Fred are 3000rpm so this looks reasonable to me.

Ger21
09-03-2016, 03:45 PM
Figure out what you want your max velocity to be, and gear it so it reaches that at 80% of the servos rated speed.

Boyan Silyavski
09-03-2016, 03:56 PM
ball nut is 10mm per revolution. Only z is 5mm

Scott Damman
09-03-2016, 04:30 PM
ball nut is 10mm per revolution. Only z is 5mm

Figure out what you want your max velocity to be, and gear it so it reaches that at 80% of the servos rated speed.
Gerry thats just the problem, dont know what I want or need my max velocity to be...

Boyan, so you are taking the gear ratio the other way then? i.e. screw speed (or ballnut speed) * 1.5 to give motor speed. If I take the same equation from before your 20,000 mm/min with a 10mm/rev screw gives a motor speed of 3000rpm?

What would be the disadvantage to doing a 5mm screw? I know the torque issue with gearing the motors the other way. However I ended up ordering 1.8kw 6Nm 3000rpm motors since they ended up being cheaper than the 1kw motors so I think I can sacrifice some torque to a gear ratio. Are there other issues with 5mm? I havent ordered the long screws so I can go either way just asking the question.

I am thinking about implementing brake resistors, I do this on my VFD's (to run 3 phase woodworking equipment) for quick motor braking. Im just trying to decide if the inertia of the gantry being so high warrants braking resistors or not. Has anyone needed to add them to their setup?

My motor order was as follows (in case anyone is interested)
Qty: 3 - 6Nm 1.8kW 3000 rpm servo motors - these will be for 2 for X axis and 1 for Y axis
http://www.aliexpress.com/store/product/New-Servo-system-kit-6N-M-1-8KW-3000RPM-110ST-AC-Servo-Motor-110ST-M06030-Matched/314742_761095604.html

Qty:1 - 2.4Nm 0.75kW 3000 rpm servo motor - this one is for the Z axis
http://www.aliexpress.com/store/product/New-Servo-motor-and-driver-set-2-4N-M-0-75KW-3000RPM-90ST-AC-Servo-Motor/314742_760999511.html

Boyan Silyavski
09-03-2016, 05:07 PM
i was trying to say / but as i was busy at workshop/ people typically use for woodworking machines xx10 screws on all axis and xx05 on Z, as this more or less gives compromise between speed, precision and step count/pulse train needed to generate movement


So say you have a servo motor that will spin to 3000rpm. From look of the servo it says 2500PPR/ pulses per revolution/ if i am not wrong. And ball screw moves 10mm per revolution.
Then ball screw needs 1/10 rev to move 1mm. If no geared then 3000/max rpm/x10=30 000mm/min max speed , like Gerry says thats your max velocity in Mach3. So from above it seems then you will need 250pulses per mm , so 1mm/250pulses=0.004mm resolution.

There is something important that Gerry told you but you missed his point. The servos should not be greatly OTT as you may think, they must be closed mutch to what effort they will do, otherwise youwill run in other problems. read literature about servo sizing. Its not a good thing to be much bigger than job they are meant to do.

Then read carefully page 14 from my build thread (http://www.mycncuk.com/threads/6619-Quite-an-Unusual-one/page14) where Johnatan explained me very well how to calculate details and especially about snappiness.

And by the way i was suggesting 20t at the motor and 30t at the rotating ball nut, not vice versa. So you finish with slightly geared up machine, 0.7-1kw servos, and very fast acceleration. But just do your self the calc. My gantry is moved by 2x 400w servos, not 2x1.5kw ha ha

Clive S
09-03-2016, 05:26 PM
Scott DammanDid you remember you needed longer cables !! that listing say 3mtrs

Ger21
09-03-2016, 05:38 PM
Most industrial servo driven routers use 25-32mm pitch screws, which virtually eliminates whipping. With steppers, you don't want to go that high because you lose resolution, but with 10,000 steps/rev with servos, resolution isn't an issue.

Imo, on a machine that large, you want at least 1000ipm, preferable closer to 1500ipm. With a 10HP spindle, you should have no trouble cutting sheet goods at 1000ipm if you want.

Scott Damman
09-03-2016, 07:49 PM
i was trying to say / but as i was busy at workshop/ people typically use for woodworking machines xx10 screws on all axis and xx05 on Z, as this more or less gives compromise between speed, precision and step count/pulse train needed to generate movement


So say you have a servo motor that will spin to 3000rpm. From look of the servo it says 2500PPR/ pulses per revolution/ if i am not wrong. And ball screw moves 10mm per revolution.
Then ball screw needs 1/10 rev to move 1mm. If no geared then 3000/max rpm/x10=30 000mm/min max speed , like Gerry says thats your max velocity in Mach3. So from above it seems then you will need 250pulses per mm , so 1mm/250pulses=0.004mm resolution.

There is something important that Gerry told you but you missed his point. The servos should not be greatly OTT as you may think, they must be closed mutch to what effort they will do, otherwise youwill run in other problems. read literature about servo sizing. Its not a good thing to be much bigger than job they are meant to do.

Then read carefully page 14 from my build thread (http://www.mycncuk.com/threads/6619-Quite-an-Unusual-one/page14) where Johnatan explained me very well how to calculate details and especially about snappiness.

And by the way i was suggesting 20t at the motor and 30t at the rotating ball nut, not vice versa. So you finish with slightly geared up machine, 0.7-1kw servos, and very fast acceleration. But just do your self the calc. My gantry is moved by 2x 400w servos, not 2x1.5kw ha ha

Yeah I think I did miss understand what Ger meant by OTT. Anyway Fred is holding the order for me so I can make some changes if I want.

Alright doing the math for the inertias of the system I did it based on a rotating ballscrew at the moment.

Some assumptions:
with 2 motors each motor on the X will handle 1/2 of the gantry inertia
Steel pulleys
2/3 ratio on the pulleys

Ballscrew inertia = 7.4E-4 kgm2
Gantry equivalent inertia (1/2 of total) = 3.44E-4
Pulley inertia Driven = 9.86E-5
Pulley inertia Drive = 1.48E-4
Total Inertia = 1.33E-3

Motor inertia for 1.8kw from BST page = 7.6E-4kgm2

So Inertia ratio with the 1.8kw motor = 1.75

I also ran Jonathans Matlab Script and got
Feedrate in m/min: [x y z]=[20.0 10.0 7.50]
Inertia in g-m^2: [x y z]=[1.56 0.96 0.80]
Torque in Nm: [x y z]=[4.29 2.63 2.27]
Inertia ratios: [x y z]=[2.06 1.27 1.06]

It looks like my X axis needs 4.29 Nm, how much above this number generally do I go?
These numbers show a slightly different inertia ratio for X but pretty close.

I know the general rule of thumb here is the keep the ratio under 10 but is there a lower limit as well? Ive searched to see if there is any information about that particular motor for inertia ratio but I havent found anything. Boyan I see your motors show an allowable of 12 but I dont see that for the motors BST sells (searched the model number online as well) so not sure how high I can go with them.

I guess based on these numbers the motors are probably oversized however by how much? These motors show 6Nm continuous, how much is that number trusted for the chinese motors? If I need 4.29Nm per the torque calculation?
Should I be closer to an intertia ratio of 5,6,7,etc?

Dont know if I know more now than I did or less...

Thanks for the heads up

Scott Damman
09-03-2016, 07:50 PM
Did you remember you needed longer cables !! that listing say 3mtrs

Email to Fred before I ordered and he confimed he can make the longer cables for me.

Scott Damman
09-03-2016, 09:20 PM
OK, maybe I am missing something here, but the more I read about proper servo motor design the more I see that the load inertia to motor inertia should be as close to 1:1 as possible. So although the rule of thumb is to be under 10 what I am seeing is for the ratio to be closer to 1. From what I am reading (and it is very possible I am mis-understanding this) the system should be better tuned and more responsive the closer to 1:1 ratio that can be achieved. This can be achieved with gearing or with sizing the motor to the load. I have included a 2:3 ratio in my calculations.

http://www.motion-designs.com/images/DTrends_Dec_2008.pdf

http://www.diequa.com/download/articles/inertia.pdf

Am I mis understanding this? To me this would tell me that my inertia ratio of 1.75 --> 2.06 that was calculated for the system would be pretty good.

So as long as the inertia ratio doesnt go the other way...i.e. the motor inertia is larger than the equivalent load inertia... it is better to have the load to motor inertia as close to 1:1 as possible.

Of course from there one has to consider continuous torque and peak torque for the motor etc etc...

JAZZCNC
09-03-2016, 10:37 PM
Then ball screw needs 1/10 rev to move 1mm. If no geared then 3000/max rpm/x10=30 000mm/min max speed , like Gerry says thats your max velocity in Mach3. So from above it seems then you will need 250pulses per mm , so 1mm/250pulses=0.004mm resolution.

Boyan Your forgetting the Encoders are Quadrature so that's 2500 x 4 =10,000ppr.!


Scott welcome to the reality's of BIG heavy Machines. Need to Slow down with the Ordering and rushing in. The fact you even considered 5mm pitch screws and Steppers on such large machine shows you need to do much more research.

You enter another world which becomes expensive very quickly if you go rushing in without doing home work.!!

Boyan Silyavski
10-03-2016, 09:52 AM
Boyan Your forgetting the Encoders are Quadrature so that's 2500 x 4 =10,000ppr.!




I am always confused about that info. Why are they never more clear. When i bought my Samsung servos, it stated 2500ppr, then i thought like you say- that they are quadrature, but it went they really meant 2500 pulses per revolution. So i actually for my machine have them programmed in Mach3 at 2500/10=250x(30t:20t)=375steps

Thats why i said he must check in servo manual and every time i think about servos ppr i check in manual.



Scott,
they are talking about controlling the resonance frequency, so if you are near it will be ok.

The bst motor inertia could not be different than typical servo motor inertia. Just make sure you know the correct length of motor and shaft diameter and you could use data from comparable motor

Scott Damman
10-03-2016, 02:52 PM
The fact you even considered 5mm pitch screws and Steppers on such large machine shows you need to do much more research.


Sometimes I forget about the speed of the screw also being an important factor. I put together a little table for resolution based on ballscrew, based on my calculations I need to go with a higher pitch or up the gear ratio to keep the speed ~80% max. Upping the ratio of course hurts resolution so Im looking for a 20 pitch screw. Higher pitch also helps keep the ballscrew speed down, but of course hurts resolution per mm.

17903

Thoughts on this?

Scott Damman
10-03-2016, 05:13 PM
From my table above I am looking at a 2525 screw for the long axis. If I do a 2:3 ratio I get 37.5 mm travel per one motor revolution. 37.5/10,000 pulses = 0.00375 mm per pulse. This would be the best resolution. Or 0.0001476 inch resolution. This seems plenty good to me but thought I would ask

This reduces the speed of the screw considerably to reach the same machine speed and actually gives me some headroom on the motor max speed as well.

At the moment I am planning on rotating the screw and putting in multiple ballscrew supports. Something like...

https://youtu.be/NWB6FAJCPhA

If I choose to do rotating ballnut I will need to recalculate the inertia and just make sure everything still matches up. I am leaning the other way as my parts are coming back around $300 usd per side....

Does this sound like a reasonable solution?

Scott Damman
10-03-2016, 05:19 PM
From my table above I am looking at a 2525 screw for the long axis. If I do a 2:3 ratio I get 37.5 mm travel per one motor revolution. 37.5/10,000 pulses = 0.00375 mm per pulse. This would be the best resolution. Or 0.0001476 inch resolution. This seems plenty good to me but thought I would ask

This reduces the speed of the screw considerably to reach the same machine speed and actually gives me some headroom on the motor max speed as well.

At the moment I am planning on rotating the screw and putting in multiple ballscrew supports. Something like...

https://youtu.be/NWB6FAJCPhA

If I choose to do rotating ballnut I will need to recalculate the inertia and just make sure everything still matches up. I am leaning the other way as my parts are coming back around $300 usd per side....

Does this sound like a reasonable solution?

Got my gear ratio backward. I need to redo my calculations...

Scott Damman
10-03-2016, 08:55 PM
Alright new table now that I got the gearing correct the ballscrew and motor speeds were correct but I took the ballnut travel per motor rev the wrong way.

17904

So based on these numbers still looking at 25mm lead screw. With a ratio of 2:3

Ballnut travel per motor rev 25*0.66 = 16.5mm
Precision mm = 16.5/10,000 = 0.00165 (is this real? i.e. with a quadrature resolver will I truly get 10,000 counts per motor rev?)
1600 rpm ballscrew speed to get to 40,000mm/min (~1500ipm)
Motor speed of 2424rpm (67% of max motor speed)


So am I looking good here? The precision seems great and the speeds would be very acceptable with some headroom on the motor speed.

Thanks

AVF
10-03-2016, 09:46 PM
BST will provide cables upto 7mtr long. This is Video I made showing there 1.8Kw set.


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

Hi Jazz,
I saw your video which made me curious: would it be possible to use this 1.8kW servo and controller for driving a spindle or could I hit trouble doing so?
BTW: Nice thread - and sorry for my little hijack!

JAZZCNC
10-03-2016, 11:50 PM
Hi Jazz,
I saw your video which made me curious: would it be possible to use this 1.8kW servo and controller for driving a spindle or could I hit trouble doing so?
BTW: Nice thread - and sorry for my little hijack!

Don't want to let this run on so please start another thread if you want to know more.!!. . . But yes large Servo's are often used for Lower RPM spindles. Better controllers like the Cslabs IP-A let you control the spindle using Servo. This means you can use the Encoder for spindle orientation for things like ATC or Ridged tapping etc and do things like this.!


https://www.youtube.com/watch?v=3xJuIeYgkr4

Scott Damman
11-03-2016, 04:04 AM
Got a little more done on the frame, havent had as much time as I would like in the last few days. My bed is 16 inch on center and I decided to brace in between each of the bed cross members only the first ones in so far.
https://lh3.googleusercontent.com/qXXcZd7UYa9XT5xYQ1tJxdxDYJvCg6xIy-nAcGQC8OyMLvpBVCqy49IlhKqlKV_EDgQy1rN1sy1ESWLC1pwz 8kUhu7HOHx-g7Sae-sxyaQTZEMb-qf8wpI0LZS7DP9o6dgDRPA6Cy-auWmYCUgi_3PFJPCBLfZthjpRFXZ68X3WK-PvuHa48NcbTHMJaP1GEV6vOFjMyZyorQKk3msxjJOBF0dSscwN ddYyDZet1_Geafgh1rEJ7q7TCAzsCvLunuJ5cws-08lixjD7uO2RYv3HoAglL0qeK6oO4YTZlFutVS9eMQvkc4o_Ag LKDGru49rdRnoZC88T9kXj-maj51_4yTfxTd5b2emeB-vRDDwcOe9pWYQaoTqd3xPN-sJgxpmXB6ZoZc7KbeCiZlMpi2Ind50aDSR9xxrM2Lhj_zOon_0 9U9EgPU6iQigVbA2GXqCiqbdydW-L5c5KmCJWqXM5N6ZdxaGjxIn3c3bdskH4OIzZ4VZyQ7ou3cjg4 T-75eJ-fNrxbo4MCVJpXGc0CgCTPeJ4hOSpAz_FcFYtopMKyr1JYeKIP4 SlX0_LoD1LKzoYjsQ=w547-h971-no.
https://lh3.googleusercontent.com/vcvsn13Lt0dqKIO7YUjjceSQbp62DZenpVWP6ggnDtcE8ZdIKr wtOGZCowcrlGCyMT6Q_n63uFZGR62m3ak5oQEn9pZIvCX8SNPc _sLZd0bw8UvOrIyrLx0KTXdEGrP9DwpdCrwbEFX3GzLdQhejPb vysGvIOPMpbtJdVZWumx2dJplC4P6xu-q5SYFhtpLQkPzmBr6A2K3okeS1Nb7tB6AgiSSWn4LJ-6yiSFf4FPmc9PeMncS__cmVtq79bWDLumsuSfn78K_FxEgfsQx mb_HPRwXgM1cRHKuaKm8Q_bt5ZES9z5y7SZQUDed6vk4e7f2xr 1vP1_4F7AF_UErJU6c2aYtsujVdKgeaLq4oX0IsKKQQ8XW3GSz QhZEqn62eWquAGdbUR-yWPJP2Q-PFWxECa8p7CHQJVhD0wLm5JtbWwAM2HCcl0K2OHM8LdpIQL6Mo kVo4iRQoNEBfzp9hHeFzxEd1FZg7fptUBh2fBd-7WOYrpw05lpqy-bDs3mTtxvWhN5uGei17yO6W629h446vHs30qhqjcz9fRLeLejn Iq6NTAKI3j-ReZOBDCOEybQYq2A=w1727-h971-no

Ballscrew
I need to finalize my plans for the X axis ballscrews so I can make mounts for them. Then motors mounts.
So a couple questions -
Do the numbers for the 2525 screw look good and make sense? On this long screw should I go with fixed:fixed mounts, I have read that this can stiffen things up a bit for a long screw.
Like some of the big guys do, can I get away with intermediate supports along the ballscrew and drive the screw? Or go the rotating ballnut route?
Do I even need the belt ratio on the X axis ballscrew? Would the motor speed be to slow without the gearing (~1600rpm at 1500ipm)? If I spin the screw, I could go direct drive with a 0 backlash coupling. Not that belting is all that difficult to implement it just takes some complexity out if I go direct coupled.

Motors
Ive run my inertia numbers multiple times with a few different numbers to try to account for some unknowns.
What I ended up with is, 2 of the 1.8kw motors for the x, the inertia ratios in my calculations worked out between 1.75 ---> 3.4 depending on the numbers. And the torque worked out between 4.25 -->5 Nm per motor, these motors are 6Nm so I feel like this is a pretty good choice.

Then for the Y I went with the 0.75kW motor 2.4Nm motor, inertia ratio calculations worked out between 2.25 --> 3.5 and the required torque was around 1.5Nm so a little headroom on torque.

I also picked that same 0.75kW motor for the Z, my inertia ratios get closer to 1.5 here and if I find issues I could change the gear ratio

If anyone has thoughts on these options for the motors Im all ears.

One thing, that I cant yet confirm, but it looks like the drives from Fred show a limit on the inertia ratio of 5. This was based on data I found online from who I think is the supplier of the drives.

Resonance
I think I will cross this bridge if/when I need to. In my day job I am a noise and vibe engineer (although automotive and nothing to do with machine design) so I may be able to figure out the correct countermeasures. I will say, as the frame sits it rings when hit, so I may borrow some accelerometers and do some impact testing (frequency response...) on the machine and just see where things are if I get ambitious.

Thanks for all the help this far!!!

Scott Damman
18-03-2016, 07:38 PM
179601796117962

I think this will be my final Z axis design. 20 inches of travel with the ballscrew inside a 3inch x 5inch x1/4 wall box tubing. I have added supports to the internal of the box tubing to add strength and support the tubing when I cut the large slot on the backside for the ballnut.

I think this will keep it pretty compact and also plenty strong. As always comments are appreciated.

Thanks
Scott

Scott Damman
21-03-2016, 01:29 PM
I got a bit done on the Z axis over the weekend. When I was laying everything out for the Z I decided that 4.5 inches between linear rails was not wide enough. So I welded on some 3/8inch plate in order to widen it to 8 inches center to center on the rails.

Starting the milled slot for the ballnut to travel
17975

Slot cut through and now cleaning up the edges
17976
17977

Internal view showing 2 internal supports. I figured these were needed to help keep movement to a minimum when I cut the large slot in the tube. They had to be driven in with a sledge as they were a little tight but I got them in. They are welded in place by drilling 3/8inch holes in the box tubing and then spot welding through those holes.
17978

Begining the face milling process. I milled a linear reference for one of the rails. The curves are just the tool paths I had to take to make it all work as my mills travel isnt enough to do the whole thing in 1 pass.
17979 17980

Here you can see the milled linear reference for the rail
17981

And a blurry picture of one of the rails test fitted
17982

sinnsvak
21-03-2016, 01:50 PM
Is that a manual mill or CNC controlled? Either way, good craftmanship!

Scott Damman
21-03-2016, 02:26 PM
Is that a manual mill or CNC controlled? Either way, good craftmanship!

It is a manual mill but it has a rotary head. Pretty cool machine, Kearney Trecker 2D rotary head die mill. I had never seen one before and just stumbled across this one when looking for a mill. It can cut circles up to 8 inches. The one I have was built in 1946 and was a little rough, very helpfuly with this cnc router build. Although the table only travels 18 inches X and 12 inches Y so for the Z axis its a bit limiting... but I find ways around that so not a big deal.

Thanks for the comment.

Scott Damman
23-03-2016, 04:35 PM
Quick update

Z axis machining is done. Now just need to start putting this together waiting on ballscrews to arrive. I hope to do the epoxy pours this coming week as the weather is starting to get warmer...

1800718008

Boyan Silyavski
24-03-2016, 08:42 AM
Quick update

Z axis machining is done. Now just need to start putting this together waiting on ballscrews to arrive. I hope to do the epoxy pours this coming week as the weather is starting to get warmer...




Looks very nice1 You are a man of action Scott :applause:

Scott Damman
05-04-2016, 03:18 AM
Finally a little bit of progress, got the Z axis final machining done, painted the metal work, and installed the linear rails. I also made up the Z to Y plate and did initial checks for squareness.

18098

18099
18100
18101

I am waiting on my ballscrews to arrive so I can finish the build out of the major mechanical parts. Crazy how they can go from China to US and clear customs in 1 day and then take 4 days sitting 3 hrs from me... Still no guarantee they are coming tomorrow.

Once ballscrews are mounted and some initial adjustments are made I will do the epoxy pour.


So a question, I received my CSMIO - IP/A in the mail and did some initial hook up to it. Followed the instructions in the manual for setup and download of mach3 plugin. I however can not connect to it. I can ping its IP address but both the firmware uploader and Mach3 can not connect to it.

Any thoughts on what I am missing?
I made a new rule in my firewall settings to include/allow the IP address
Intially I can ping it and get flashing lights on the IP/A
after about 3 - 5 minutes the lights go solid and I can no longer ping it...

Thanks

JAZZCNC
05-04-2016, 04:18 AM
So a question, I received my CSMIO - IP/A in the mail and did some initial hook up to it. Followed the instructions in the manual for setup and download of mach3 plugin. I however can not connect to it. I can ping its IP address but both the firmware uploader and Mach3 can not connect to it.

Any thoughts on what I am missing?
I made a new rule in my firewall settings to include/allow the IP address
Intially I can ping it and get flashing lights on the IP/A
after about 3 - 5 minutes the lights go solid and I can no longer ping it...

Thanks

Strange because Usually very straight forward provided you set Network adapter manually it works.

So have you set internet protocol version4 (TCP/IPv4) address manually.?

Scott Damman
05-04-2016, 02:26 PM
Strange because Usually very straight forward provided you set Network adapter manually it works.

So have you set internet protocol version4 (TCP/IPv4) address manually.?


Yeah 10.1.1.1 was entered manually into my computer. I pinged 10.1.1.2 when reaching the IP/A.

Couple other things I tried, I made 2 new rules for the CSMIO firmware software and Mach3 still no luck.

What I am assuming at the moment, I am using my work laptop as that is the only computer I currently have, I bet it is something with my IT policy on the machine that is blocking it (other than windows firewall). So time to pull my old desktop out of the rafters and get it back up and running...

Still on the search for a good enclosure for the electrical components. Anyone ever put the spindle motor vfd on the gantry rather than run long VFD motor cable to the spindle? I was thinking about do this to keep the 3 phase cabling (higher current to the spindle at least) as short as possible. I think some of the big CNC players do it this way where they have a control cabinet on the backside of the gantry. Thoughts on this?

Thanks
Scott

Boyan Silyavski
05-04-2016, 10:24 PM
Anyone ever put the spindle motor vfd on the gantry rather than run long VFD motor cable to the spindle? I was thinking about do this to keep the 3 phase cabling (higher current to the spindle at least) as short as possible. I think some of the big CNC players do it this way where they have a control cabinet on the backside of the gantry. Thoughts on this?

Thanks
Scott

I was contemplating this many times or to integrate it in the machine somehow. First the dust will be a problem, but you could make a cover for it. Biggest issue for me is if cable disconnects and touches a metal while i am touching that metal too. though the machine must be grounded. I don't know. there must be a reason not to do it. It has some vibration requirements also.

JAZZCNC
06-04-2016, 09:36 AM
Do-able but involved and costly.

Dust and Vibrations are the biggest issues. Cover the VFD and heat becomes an issue so Good Filterd cooling is required. Vabration over time can cause issues so it will need mounting on rubber standoffs.

Davek0974
06-04-2016, 09:02 PM
I would not do it - if this is a single phase input/three phase output VFD then the supply cable will be carrying more current than the motor cable. As mentioned, vibration and muck are serious enemies of electronics.

Scott Damman
14-04-2016, 01:52 PM
Thanks for the thoughts on mounting the VFD to the gantry. I had kinda thought about those items as well, the main driver for me was actually 3 phase switching noise caused by the VFD. I use double shielded VFD cable but at these high frequencies (300 - 400Hz AC at 2.5khz switching) there can be alot of noise that causes interference elsewhere in the home. Anyway I have scrapped the idea in favor of wall mounted.

Finally some updates on the machine.

18199
18200
18201

X axis motor mount plates (laser cut) and servo motor mocked up.
18202
18205
18206

Long ballscrew mocked up for bracket design
18204


Also made some good headway on the electrics. Still cant get CSMIO to connect so will have to contact them here soon but Im a bit of time away from needing it.
18203

Tom J
29-07-2016, 05:35 PM
I fully agree, legs has to be revised.
I make another desktop router with fixed gantry this time and see how I made it:
18946189471894818949
all made from rectangular profile 50x25x3
I weld everything with Tig this time will start new topic about my build
Good luck
Tom

Scott Damman
07-12-2016, 03:52 PM
I have been MIA for a long period, normal life and other projects got in the way over summer. Finally back to working on the CNC.

19859
Leveling the Y to Z plate for gantry foot epoxy pour
198601986119862
Pouring the gantry feet
19863198641986519866
Ballscrew to Z axis plate, made with steel I had on hand
198671986819869
Installing limit switches for Z
19870198711987219873
Z home Prox Switch
198741987519876198771987819879
Finished underside of Z axis with dust shielding in place
198801988119882198831988419885
Z axis installed onto Y axis/gantry
198861988719888198891989019891
Lots more Z axis mockup
198921989319894198951989619897198981989919900
Building brackets for dust collection holding and pneumatic counter balance system
199011990219903199041990519906
Machining and installing guides for dust shroud, shroud/brush will be on pneumatic cylinders so it can be pulled up when needed and also when I add the tool changer.
199071990819909
As it sits today, my plan is to work backward from the Z axis finishing each axis as I go. Sorry for the dump with little text but I wanted to update this with where things are. I can give any specific information as needed

Scott Damman
12-12-2016, 01:49 PM
Made some progress forward... Z is moving although I cant get the video to download.

Also got Y going for initial tuning, here is a video jogging at ~1600 inch/min then I have a question for all the servo tuning experts out there.

https://youtu.be/SOFWMtNni1g

Ok, so trying to get that Y axis tuned through the CSMIO config it seems I can get either no overshoot and no oscillation but I never actually get to the number I requested. Or by changing the I value just 1 I get slight overshoot and oscillation. I am tuning at 2000 inch per min max rate and an acceleration set to 500inch/sec/sec.

Right now my numbers are, with these numbers there is no overshoot and the movement is very crisp with no oscillation at 0 speed. But I am always 0.003 of an inch from the requested number. (i.e. if I ask it to go to 8 I get 7.997). Error in this setup ends up around 10 to 12 counts
P - 800
I - 4
D - 10

If I change the I to 5 leaving the P and D the same, I get slight overshoot, but also get oscillation around the number, around 100 counts + and - . Also at 0 speed I am getting oscillation around 30 counts + and -. The CSMIO config dialog wont let me put a decimal into the value otherwise I think 4.5 could be the magic number.

Any thoughts? I think not getting to the requested number wont work in real machining? Also overshoot or oscillation could have some pretty negative impact. Kinda got to the point of pulling my hair out yesterday after around 3 hours of messing with it so I decided to move on for now. Oh and a BTW, the autotuning feature did not work at all. The numbers it left me with gave oscillations around 1 full inch of movement. I am happy with the crispness of the response on how I have it tuned just very odd I cant get to the number I asked for.

Thanks

JAZZCNC
12-12-2016, 04:18 PM
Now your ino the realms of Black Magic.!!

First thing first are you sure the Drive is tuned correctly to the motor.? If not then wasting you time tuning the controller.

However 500in/s/s is very high(12700mm/s/s) and could be the cause of your tiny overshoot. This is very abrupt stop and from high speed. You can see in the video how abrupty it stops. Try lowering the Accelration and see if the error disappears or reduces.

Also I notice you using kD value which often if required is much higher valule than 10. 100's even 1000's isn't unusual.
The kD value will fine tune the oscillation or any humming noice from motors.

You also don't mention the Kvff value. With such high acceleration then this value will come into play slightly as it trying to predict the next action and can help reduce errors if matched well to PID.

To be honest no one can truely help you with this unless in front of machine. Even then it's mostly trial and error. Also I wouldn't waste much time fine tuning untill the axis is fully finished and working as it will be when cutting. Just adding little more Mass or lowering resonance can change the Tuning when comes to fine detail.

Good luck and keep cool it will be worth the effort.

Scott Damman
12-12-2016, 06:59 PM
Yeah I figured the cart was ahead of the horse so to speak but it was fun to get something moving. I was planning on just doing a quick tune so that I could jogg the axis... 3 hrs later I was so frustrated I had to say stop.

I dont know that the motor is tuned correctly, these are the units from BST and I just made sure to set the drive to the correct number for the motor. I will need to do a little more learning on servo drive tuning before jumping into that can of worms... Should I need to dig into the servo driver setup? I was hopeful that just setting the number for the motor would get me where I needed to be.

The acceleration was set just trying to push the limits of the servos a bit, I ran all the way down to 50 inch/s/s and it is always 0.003 inch off the mark with the PID set the way I have it. Then if I changed the PID, the overshoot and hunting was always essentially the same, so with that I came to the conclusion that the acceleration may not be in play? But obviously more tuning needs to be done.

I do need to play more with the KD and Kvff, but I think I will take your advice and wait until the whole thing is together before performing the final tuning.

Have a Mach4 question as well, DRO readout with the CSMIO-IP/a, in Mach3 if I would turn a servo motor by hand the DRO would update. In Mach4 I am not getting that, hoping that its just a setting somewhere that I overlooked? Anyone have any thoughts? I was using the demo of Mach3 and decided to buy Mach4 so I dont need/want to upgrade as quickly.

So onto finishing the mechanicals. Thanks for the help

Here is the Z axis jogging, I didnt do much of any tuning here just got it running making it easier to work on.

https://youtu.be/NzzJ5Tek6u4

Boyan Silyavski
12-12-2016, 08:17 PM
Are you sure about that acceleration? My machine is tuned at 3000mms2 and is much snappier at 10000mm min feed rate. And from servos is still on 50-60% of their ability. Frankly when working at around 16000mm/min it looks really dangerous and leaves you no time for any reaction.

Davek0974
12-12-2016, 08:23 PM
I got my servos from BST too so likely the same, I think the thing with the drive number setting the drive up is that it has no idea what load the motor has - its just a ballpark setting. The downside is the manual is written Chinglish and unless you know exactly what parameter does what, then its all guesswork.

I spent hours and hours tuning, its made harder as you also need to test each tune at multiple speeds, I test at 3, 13,23,33,43 and so on as % of full speed - its easy to tune so that it runs great at say 60-100% but oscillates badly below 50%

In the end i tweaked in the order P - D - I until the following error was as low as possible, then added some kVff if needed.

I'm still not happy with it though but at least the mill is running now. No doubt an expert could tune it in minutes but I don't know any mobile servo experts ;)

Scott Damman
12-12-2016, 08:27 PM
Are you sure about that acceleration? My machine is tuned at 3000mms2 and is much snappier at 10000mm min feed rate. And from servos is still on 50-60% of their ability. Frankly when working at around 16000mm/min it looks really dangerous and leaves you no time for any reaction.

No I am not sure, its very possible your machine is snappier I will have to take a look at your video, pretty hard to compare on video I suppose. Anyway, I was just going based on the settings in Mach that I had in at the time, quiet difficult to confirm at what speed its running. Also I was just jogging at a % of max (I am not sure if that % reduces both the feed rate and the acceleration) If both then what the video shows would be slower...

Boyan Silyavski
13-12-2016, 09:55 AM
No I am not sure, its very possible your machine is snappier I will have to take a look at your video, pretty hard to compare on video I suppose. Anyway, I was just going based on the settings in Mach that I had in at the time, quiet difficult to confirm at what speed its running. Also I was just jogging at a % of max (I am not sure if that % reduces both the feed rate and the acceleration) If both then what the video shows would be slower...

I agree, its difficult to guess from video. All I was saying that you have to check that acceleration in Mach3, the plugin you are using and the drives. If either one of the 3 is set differently then its the lowest value active in reality. Though i dont know CSMIO and if you have to set the acceleration in the plugin separately, so yo better check manual.

My motors are only 400w but i assure you when i raise the acceleration snappiness from the servos themselves, its like shooting a bullet. Thats the sound how machine starts and stops. In fact my problems when tuning were related to Undershooting, cause i was not brave enough at first. So problem was it goes to position fast and then you could see how the motors for a millisecond were further positioning it in place. So i thought i am Overshooting. Yes but No. When i raised the servo response this disappeared. So auto tuning did not work i my case. Also it would be better if there is a total gain adjustment which auto adjusts all details, instead of separately adjusting each parameter. Which by the way is well described on many places around the web, but i still fail to comprehend it entirely .

Scott Damman
13-12-2016, 01:53 PM
Made a tiny bit of progress last night. Got one of the Y axis limits and also the homing prox installed.

19992199931999419995



Chinglish and unless you know exactly what parameter does what, then its all guesswork.

Dave, This is exactly what I was worried about when digging into the servo itself. I think if I get a bit more understanding of servo driver tuning I will have a better chance of understanding the Chinglish... I hope.

Boyan, that is a very good point, I never checked the max acc and decc in the servo drive. Likely that is the limiting factor. I didnt think about it until now but as I increased the acc I didnt notice an increase physically so probably the servo drive is limiting. I will have to take a look tonight.

Which by the way is well described on many places around the web, but i still fail to comprehend it entirely . I totally agree, seems like each spot that describes how to tune a servo does it slightly different, making it more confusing for those of us just getting into it... Like Jazz said Black Magic

Thanks for the help!

Chaz
13-12-2016, 03:19 PM
Let us know how you manage with the PID tuning.

I have done 3 machines with CS Labs and it's been hit and miss with all of them.

The first was my Denford. If the settings are 'weak' the machine is slow and cumbersome. If too aggressive I get PID errors although I have found a happy medium.

In all cases I have tried auto PID tuning (the controller, after tuning the drives themselves) but limited success.

Even the manual indicates its better to do manually in most cases.

For another machine I retrofitted, we managed to get a nice speed / rate on the machine but its small and doesnt carry a lot of weight.

With my Concrete machine (Thor), it will be interesting. The design is fixed gantry, Y moves a table of around 100kg and I am looking at speeds of >10m/min using a Panasonic 1.5 KW Servo and very beefy ballscrews.

The drives have an auto tune feature which tunes all the time. How this works in reality, I dont know. I will have the Y running again this weekend, just need to do some alignment and will see what I can get from the machine.

I have a video of the machine doing slow movements of I think 4000mm/min. I have had one servo run away (my own fault, made a mistake in config) and I can confirm that the table can move a metric fuckton faster.

https://www.youtube.com/watch?v=5sMb4XmzuXc

Davek0974
13-12-2016, 07:06 PM
I have a video of the machine doing slow movements of I think 4000mm/min. I have had one servo run away (my own fault, made a mistake in config) and I can confirm that the table can move a metric fuckton faster.

https://www.youtube.com/watch?v=5sMb4XmzuXc

LOL ;) nice unit of measure :)

Chaz
13-12-2016, 08:35 PM
LOL ;) nice unit of measure :)

Ye, fairly standard UOM.

Chaz
14-12-2016, 11:47 AM
LOL ;) nice unit of measure :)

Sorry, I know see what you were referring to (not the mm/min). Yep, very fast indeed.

Scott Damman
19-12-2016, 03:08 PM
Made a bit more progress over the weekend,

Got one of the X axis screws mocked up
20043
Here I am machining the spring loaded brackets that will act as X ballscrew mid supports. These will support the screw and when the gantry goes by will move out of the way on the pivot. I need to get some shoulder bolts still for these and do a bit more machining for clearance of the ballnut as it passes (to ensure that it cant get hung up on a change of direction).
20039
20040
20041
20042

I also got a air valve block installed on the gantry for all the air that the Z axis needs, also put a terminal block on the gantry in order to reduce the number of wires that needs to make its way back to the control. I ordered some 20 conductor shielded communication cable to go from the control to the gantry terminal block. Gives me a few extra wires incase I need any future expansions or additions.

Im getting excited as its closer and closer to cutting. I am waiting on new toothed pulleys for the Y and X as I made a mistake when ordering, once those are in it will be time for truing and squaring then cutting.

More to come....

Scott Damman
20-12-2016, 03:04 PM
Some more progress made, I got the ballnut mount made and installed and also finished one of the ballscrew mid supports. I made a video of one of the ballscrew supports in action, I am just pushing the gantry back and forth. I pushed the gantry from one end to the other as fast as I could and this one support seemed to damp the ballscrew frequency. Im hoping to finish up this side tonight and then do it all over again on the other side.


https://youtu.be/0gP4IUO0QKg

20052
20053
20054
20055
20056
20057

I need to do some learning on tuning 2 X axis servos at the same time, if anyone has any suggestions im all ears.

Scott Damman
07-01-2017, 10:03 PM
Made some good progress on the machine, I am to the point of wanting to run a gcode program, although there is alot to do before I can actually cut.

I loaded a sample Gcode program just to see all 3 axes move together

I moved the machine 12 inches away from X and Y and put my Z in the middle of travel and zeroed the DRO's for work start. When I hit cycle start (with my hand firmly on the e-stop...) I get a message "Move off limits before cycle start". I then checked the I/O window in CSMIO config and no limit switches are tripped. I tried a few other things within mach with no luck. I then turned off all limits (except e-stop) and I get the same message. Did some searching/reading on forums and didnt come across anything.

Anyone have any thoughts on what this could be? I hope its something simple that I just havent turned on/off?

I plan to a write up on the progress, just this issue has me frustrated so thought I would ask.

Thanks

Chaz
07-01-2017, 10:55 PM
Made some good progress on the machine, I am to the point of wanting to run a gcode program, although there is alot to do before I can actually cut.

I loaded a sample Gcode program just to see all 3 axes move together

I moved the machine 12 inches away from X and Y and put my Z in the middle of travel and zeroed the DRO's for work start. When I hit cycle start (with my hand firmly on the e-stop...) I get a message "Move off limits before cycle start". I then checked the I/O window in CSMIO config and no limit switches are tripped. I tried a few other things within mach with no luck. I then turned off all limits (except e-stop) and I get the same message. Did some searching/reading on forums and didnt come across anything.

Anyone have any thoughts on what this could be? I hope its something simple that I just havent turned on/off?

I plan to a write up on the progress, just this issue has me frustrated so thought I would ask.

Thanks

Are you certain they are wired correctly (you should manually trigger them and check that they match the input on the Mach 3 screen).

Also, do you have any soft limits enabled?

JAZZCNC
07-01-2017, 11:06 PM
Does it Jog Ok.? . . . . If so Post the G-code.

Try doing some MDI Commands. Like G0 x6 y6 (assuming inch units) or G1 x6 y6 f30.
If these work ok then I'd be suspect of the G-code.

Scott Damman
07-01-2017, 11:11 PM
Are you certain they are wired correctly (you should manually trigger them and check that they match the input on the Mach 3 screen).

Also, do you have any soft limits enabled?

Sorry should have stated that, all limits were checked manually and all seem to correspond to the correct input. Soft limits were turned off during the troubleshooting process.

Jazz, system jogs great, homes correctly on all three axises. I tried MDI commands and get the message "Move off Limit before MDI command". Whats puzzling to me is that Mach says this even with all limits turned off.

thanks for the quick replies

JAZZCNC
07-01-2017, 11:50 PM
Actually thinking about it I've had this before but can't remember what caused it.! . . . Got feeling it's something silly and obvious like a setting but can't for the life of me remember what.? . . . . . Nearly bed time so I''l sleep on it and get back to you.

Scott Damman
08-01-2017, 12:11 AM
Actually thinking about it I've had this before but can't remember what caused it.! . . . Got feeling it's something silly and obvious like a setting but can't for the life of me remember what.? . . . . . Nearly bed time so I''l sleep on it and get back to you.

I took a little break from the machine and came back to it. Ended up being a debounce issue, it looks like my cable grounding wasnt as good as I thought. I doubled the debounce to 8ms and the MDI and Gcode both ran....

Pretty darn cool to see it run on its own, I am so close to cutting...

Here is a pic of the CSMIO-IPA Config window for setting debounce for anyone in a similar situation.

20304

JAZZCNC
08-01-2017, 02:00 AM
I took a little break from the machine and came back to it. Ended up being a debounce issue, it looks like my cable grounding wasnt as good as I thought. I doubled the debounce to 8ms and the MDI and Gcode both ran....

Pretty darn cool to see it run on its own, I am so close to cutting...

Here is a pic of the CSMIO-IPA Config window for setting debounce for anyone in a similar situation.

20304

Not good idea to use de-bounce because it affects all the inputs and slows down re-action time. Find the problem and fix asap, esp if using servos because want any inputs to re-act fast.

Edit: Actually I see you only have set to 2 which isn't so bad thought you'd set it to 8 which would have been 32ms.

Scott Damman
08-01-2017, 02:13 AM
Not good idea to use de-bounce because it affects all the inputs and slows down re-action time. Find the problem and fix asap, esp if using servos because want any inputs to re-act fast.

Edit: Actually I see you only have set to 2 which isn't so bad thought you'd set it to 8 which would have been 32ms.

Jazz, good point either way, I think I will dig into the shielding a bit more on all of the control wiring. Would like to get it back to 4ms. Like you said fix with hardware, totally agree.

thanks again, more to come

JAZZCNC
08-01-2017, 01:08 PM
Jazz, good point either way, I think I will dig into the shielding a bit more on all of the control wiring. Would like to get it back to 4ms. Like you said fix with hardware, totally agree.

thanks again, more to come

Have you Earthed the Frame.? Common cause of troubles like this.

Scott Damman
10-01-2017, 02:35 PM
Have you Earthed the Frame.? Common cause of troubles like this.

No I hadnt, I added a separate ground for the frame and grounded each part of the machine, i.e. Gantry and Z. Thanks for the suggestion.

I learned some hard lessons last night while continuing to tune the servos.

1st, make 400% sure that the ballscrew mid-supports can not get hung up on a sudden change of direction.
2031720318

This bent back pretty easily but I will need to make some modifications to make sure it cant happen again. Basically the machine stopped before the support flipped and on change of direction it caught on the ballnut mount. No other damage observed which is good.

2nd, and this is the big one, when butting linear rail together remove the lead in on each rail. I had known this was the case but being that I got the rails for cheap and not having a way to accurately grind the rails I chose to try it.... bad choice.

2031920320203212032220323

What the pictures show (or what I assume happened), as the block went past the joint a ball dropped into the joint, caused alot of extra force and blew the end of the block off. At lower speeds it seems to be fine, but I was running past it at ~1000 ipm. I think the block counldnt handle it I was able to stop it before the other 2 blocks hit the joint. I have 4 extra blocks that came with the used rails so I can put it back together for now but I need to make a plan for future.

I think my options are
1 - remove rails and have them ground and put back together - this is a no go as the cost would be more than just buying Hiwin rails
2 - buy Hiwin rails and blocks - definitely the best option but costly
3 - try to fill the groove/lead in
- Im thinking of trying this method, couple options, epoxy, babbitt, other options?
- I think I just need to keep the balls from dropping in there, I think there shouldnt be too much stress on the fill otherwise.
4 - just use the 8ft of full length rail for now???

Any thoughts are appreciated...

JAZZCNC
10-01-2017, 06:44 PM
Ouch.!! . . . Bad luck. That's difference between Servo and steppers. They take no prisoners! . . . Stepper would have stalled before bending that bracket or possibly ripping the bearing end cap apart.!

1st Choice is no brainer new Hi-win rails.

2nd Choice new Hi-win Rails

3nd Short term Choice fill with Metal Epoxy. Think it's called JB Weld in states.

No other choices IMO.!

Scott Damman
12-01-2017, 02:22 AM
Ouch.!! . . . Bad luck. That's difference between Servo and steppers. They take no prisoners! . . . Stepper would have stalled before bending that bracket or possibly ripping the bearing end cap apart.!

1st Choice is no brainer new Hi-win rails.

2nd Choice new Hi-win Rails

3nd Short term Choice fill with Metal Epoxy. Think it's called JB Weld in states.

No other choices IMO.!

Short term I have moved the entire machine to the other side of the joint and moved my X home switch. Gives me 6.5 feet in the x and 5.5 feet in on the Y and gets me going with making parts. I have ordered machinable epoxy to do a pour to try to fill the joint. Long term after a few paying jobs, probably going to order some new full length rails. This will unfortunately require some modifications to the machine as the rail height will be different.

Im still waiting on my spindle cooling fan adapter so I cant cut anything yet. So I put a sharpie in the spindle and made a drawing. Pretty geeked that it actually made something, even just a drawing.
20335

https://youtu.be/J3Nyo5k5jGg

Scott Damman
13-01-2017, 04:09 AM
While still waiting on the fan adapter so I can cut with the machine, I decided to draw up a dust collection adapter mold to lay up fiberglass in. Video shows the machine running this code.

20339


https://youtu.be/KvlEG_NkDNg

I put the X and Y to 1000ipm as 2000ipm scared me and while I am learning about this machine I wanted to be a little safer.

I also have a question about what a ballscrew/ballnut should sound like. I have never been up close to a ballscrew before so I wonder if these sounds are normal. The sound I am referring to sound like bearings circulating through the ballnut but just doesnt sound real smooth... Sorry pretty hard to tell on the video I guess but if anyone has insight please let me know.


https://youtu.be/64T4WCVUdHk

Boyan Silyavski
13-01-2017, 11:41 PM
I have seen dust collection but your blasted me from my chair :encouragement: If thats not OTT, i dont know what is. May be what once i have seen the whole spindle in the dust tract.

Chaz
14-01-2017, 07:47 AM
While still waiting on the fan adapter so I can cut with the machine, I decided to draw up a dust collection adapter mold to lay up fiberglass in. Video shows the machine running this code.

20339


https://youtu.be/KvlEG_NkDNg

I put the X and Y to 1000ipm as 2000ipm scared me and while I am learning about this machine I wanted to be a little safer.

I also have a question about what a ballscrew/ballnut should sound like. I have never been up close to a ballscrew before so I wonder if these sounds are normal. The sound I am referring to sound like bearings circulating through the ballnut but just doesnt sound real smooth... Sorry pretty hard to tell on the video I guess but if anyone has insight please let me know.


https://youtu.be/64T4WCVUdHk

I helped someone with a kneemill yesterday. Not a new machine but still in good condition. I could also hear balls recirculating when we were moving the Y Axis. I *think* its normal but wouldnt mind the guidance from those more experienced on the forum.

Davek0974
14-01-2017, 09:21 AM
My 1605 screws make a Shhhhhh noise when doing rapids, they get a bit quieter after a grease.

JAZZCNC
14-01-2017, 11:05 AM
Looking good Scott coming along nicely.

The bearing noise is quite normal for that class of bearing at those high speeds. It won't be just the ballnut your hearing either, the linear bearings will make noise also.

Scott Damman
16-01-2017, 03:34 AM
I have seen dust collection but your blasted me from my chair :encouragement: If thats not OTT, i dont know what is. May be what once i have seen the whole spindle in the dust tract.

Boyan, thanks I think it might be a little much and also Im not real sure how it will work but I run 6inch to all my machines and have a 10hp cyclone dust collection system so I thought it would be better to have 2 4 inch tubes down to the bit rather than just one 6 inch. I will make an update on how it works once its all hooked up and the dust shoe is made/installed.

Thanks for the comments on the ballscrew/nut the sound just didnt sound great but like I said I havent ever been close up to a machine. Of course once it is cutting those sounds cant be heard.

So I decided to make a temporary spindle cooling fan adapter and get this thing cutting. So the first thing I wanted/needed to make is the dust collection adapter mold.
20377

Heres the machine cutting for the first time.

https://youtu.be/H3z2QBAujfE

https://youtu.be/pZp9UeCNjcE

Turned out pretty good for the first cuts, here is a final picture and the first 1/2 of the dust collection adapter laid up in fiberglass.
20378
20379

I learned a few things on my machine, first one was not to difficult, I needed a better tune on the Z axis. I think you can see in the movie that the Z actually moved up and down slightly through its pass, for this part it didnt really matter so I let it run and played with the tuning a bunch more and I think I get it pretty good.

The other thing is I was making ellipses instead of circles. Again this part didnt matter that much so I let it run to look into it after. So after confirming gantry to X squareness I started looking into what else it could be.

Heres what I found and looking for a little help on this, I also want to say that I really appreciate all the help the experts on this forum give and I hope this build helps others with theirs..

I put the machine on a 10 inch X movement back and forth at a moderate speed, while watching the gantry I could see a noticeable racking of the gantry. Upon stopping, the gantry comes back to square. It looks like to me that X axis slave motor/ballscrew is noticeably slower at starting. I tried to use the motor delay setting in Mach, I moved the milliseconds up (delaying the master motor since the slave seemed slower) but I could not see a noticeable change. So for all of the experts out there, is the motor delay something I should be adjusting? Or is there something else I should look into? I have confirmed all the settings in Mach on both motors are the same. I also confirmed the settings in the servo drives are the same. Any thoughts are appreciated.

Thanks
Scott

EDIT
I just looked closer at the delay, I had assumed that it was a signal delay, but as the manual says (which I should have looked at first) the motor delay is for delaying the enable signal for that output. So now I really dont know what to do.

The reason I say the slave axis seems delayed is the gantry racks as if the master motor is the only one driving it. I thought maybe the slave motor was/is not running but the fact that the gantry gets to square when it stops I think tells me the slave servo is running (resolver is bringing the system back to correct position). The servo drives output for speed match when running as well. Hope there is a good way to adjust for this delay?

Boyan Silyavski
16-01-2017, 11:45 AM
Make sure both servos have exactly the same settings in the servo drives themselves. I do that by copy-paste-upload from one servo to the other in the servo tuning program. So i am 100% sure they work the same. Do so in mach 3 in motor tuning tab and make sure to hit "apply". Close. go out of mach 3, then do check again the result. Then make sure the direction of motors is set right before you rip your gantry.

Then test.Now all must work normal





About the dust, i see. You are right that dust extraction better not be choked down. I assume you will fit a kind of shoe there. Or you will extract only the fine dust not to float around in the air? I use sometimes my dust shoe only like that, without the brush attachment. Chips stay on table but no fine dust in the air.

JAZZCNC
16-01-2017, 11:08 PM
May sound obvious but if you are using pulleys are they the same size.?

If this was Step/Dir setup I'd say the Step pulse was on wrong edge but with Analog IP-A then can't happen.

Does it look like it's running out over time while cutting or is it off straight away.? . . . . It could easily be electrical noise inducing voltage on the signal lines.
This would have the same affect has being told to move fractions of inch more than should. The encoders will always bring it back to zero when homing so very difficult to see.

Does it do it without Spindle Running.?
Do you have any Line filters on the Drives or VFD.? . . . If not fit some because Servo's, esp Analog don't tolerate noise in power line very well.!

Scott Damman
17-01-2017, 02:41 PM
Make sure both servos have exactly the same settings in the servo drives themselves. I do that by copy-paste-upload from one servo to the other in the servo tuning program. So i am 100% sure they work the same. Do so in mach 3 in motor tuning tab and make sure to hit "apply". Close. go out of mach 3, then do check again the result. Then make sure the direction of motors is set right before you rip your gantry.

Boyan, I went through all the parameters manually as the drives from BST Automation dont have a computer hookup. But I was able to confirm that all parameters are the same in the drives.


JAZZCNC
Re: First Build 5 x 12 Steel Frame CNC router

May sound obvious but if you are using pulleys are they the same size.?

If this was Step/Dir setup I'd say the Step pulse was on wrong edge but with Analog IP-A then can't happen.

Does it look like it's running out over time while cutting or is it off straight away.? . . . . It could easily be electrical noise inducing voltage on the signal lines.
This would have the same affect has being told to move fractions of inch more than should. The encoders will always bring it back to zero when homing so very difficult to see.

Does it do it without Spindle Running.?
Do you have any Line filters on the Drives or VFD.? . . . If not fit some because Servo's, esp Analog don't tolerate noise in power line very well.!

Thank you for the thoughts
Both sets of pulleys and belts are the same and came from the same supplier.

It is more of a delay to start, i.e. the master motor starts moving and the slave starts up just slightly after (which racks the gantry). When it comes to a stop the master stop slightly before the slave bringing it back into square.

Happens whether the spindle is on or off.

One thing to note the cables are 1meter longer to that motor than the master motor. I wouldnt think that should cause this but maybe?

I think to your point it is likely noise, I found yesterday while cutting some other parts that my Y is undershooting its commanded position. I was cutting a part that was supposed to be 8 inches in Y and it is 7.85inch. When I run the Y without the spindle on and a dial indication its within 0.001inch... So I am going to have to dig into the noise.

When you say line filters on the VFD's you are talking input line filters correct? I will see if I can come up with some input line filters. Also want to dig into the grounding more and confirm that everything is hooked up correctly. I have everything grounded back to a single point in the cabinet and then that run to our homes earth ground. I am thinking of driving a separate ground rod for the machine to give it a shorter path to ground, good idea? The spindle is run with triple shielded continuous flex VFD cable, the cable shield is connected to the chassis of the VFD using a VFD cable connector, i.e. the shield is connected 360 deg to the chassis of the VFD. The VFD is then grounded back to the single point ground in the cabinet. Open to help on this!


Made some progress on the dust collection adapter first picture is the first half out of the mold and the second is the second half laid up in the mold.
20388
20389


Also worked on a support arm for dust collection tube. It bit overkill but it was fun to make...
20386
20387

JAZZCNC
17-01-2017, 06:32 PM
It is more of a delay to start, i.e. the master motor starts moving and the slave starts up just slightly after (which racks the gantry). When it comes to a stop the master stop slightly before the slave bringing it back into square.

Again another obvious suggestion but got to be said to eliminate. Not got loose Pulley.?


When you say line filters on the VFD's you are talking input line filters correct?

Yes Input line Filters.

Did you Earth the Frame.?

Re: Separate ground rod then I'd say not because want to use the same ground that mains supply does so don't introduce ground loops. However I'm not expert on this and don't know the differences between American and UK electrics other than frequency.

Scott Damman
17-01-2017, 08:18 PM
Again another obvious suggestion but got to be said to eliminate. Not got loose Pulley.?



Yes Input line Filters.

Did you Earth the Frame.?

Re: Separate ground rod then I'd say not because want to use the same ground that mains supply does so don't introduce ground loops. However I'm not expert on this and don't know the differences between American and UK electrics other than frequency.

I appreciate all of the "obvious" questions as it is easy to overlook items when troubleshooting... Yes confirmed no loose pulleys, I also have been watching the actual belts on the servos on X (hard to watch both of course) but it looks like the master is moving before the slave. I plan to swap cables to those motors and then swap drives and find where the issue follows. I can then hopefully pinpoint, cables, drive, motor being the culprit. (or still noise)

Frame grounded and I have a ground going to the gantry and to the Z axis to make sure each part is grounded in the frame.

So the separate ground would need to be bonded back to the ground ref in our house (as I believe this is code). My thinking here would be that this ground rod would give a much shorter path to earth for the Hf noise. Not sure it would make a difference or not just something I was thinking of trying.

One question I have is about running cables in the same drag chain, I had assumed that having good shielded cables that I could run all of the cables together in the same chain. Is this a No/No?

I have been looking into line filters to try to get some ordered for the servo drives and the VFD for the spindle.

JAZZCNC
17-01-2017, 08:51 PM
Frame grounded and I have a ground going to the gantry and to the Z axis to make sure each part is grounded in the frame.

Not good idea unless you know they are isolated. It's little things like this that can easily introduce ground loops. I would remove the Gantry and Z axis earths and check again.

Also don't Earth the PC chassis to the Star point. This can cause all sorts of strange happenings. The PC PSU will have it's own path back to earth so adding another affectively creates G-loop.!


One question I have is about running cables in the same drag chain, I had assumed that having good shielded cables that I could run all of the cables together in the same chain. Is this a No/No?

If all is correct regards Grounding etc and with good wiring practices then shouldn't be problem. I do this all the time and don't have any issues.

However it's always good idea to keep some seperation of high power/frequency cables like VFD or Mains voltage in respect Signal wires. Esp if the wire runs are long which guess may be on your machine.

Davek0974
17-01-2017, 10:07 PM
Also don't Earth the PC chassis to the Star point. This can cause all sorts of strange happenings. The PC PSU will have it's own path back to earth so adding another affectively creates G-loop.!


This is a major one - spent weeks debugging a plasma table build just because i grounded the PC case to the table, disconnect it and instant fix!

ba99297
17-01-2017, 11:06 PM
Yes confirmed no loose pulleys, I also have been watching the actual belts on the servos on X (hard to watch both of course) but it looks like the master is moving before the slave.
Have you ever check your belts(tooth size, tension)?
Also are your motors and ballscrews bases tight enough?

Boyan Silyavski
18-01-2017, 06:30 PM
Gantry have to be grounded to frame on Plasma otherwise you can weld the Hiwin balls together :hysterical: I believe that is correct to connect gantry to machine body whatever the machine, to avoid that bearing fault, it had its own name, but i forgot it...

JAZZCNC
18-01-2017, 06:36 PM
Gantry have to be grounded to frame on Plasma otherwise you can weld the Hiwin balls together :hysterical: I believe that is correct to connect gantry to machine body whatever the machine, to avoid that bearing fault, it had its own name, but i forgot it...

Yes to the machine frame but not back to the Star point. Only one Earth cable should go back to this point.! Scott is runnig separate cables back to the Star point which Can/does create Earth loop.!