[fuelsguy]

I'm a beginner to CNC and so far I have progressed to having my router up and running in a fashion.
It's a home built wood frame machine with skate board bearings on aluminum tracks for each axis, and using a Stepperworld board and -23 stepper motors, with Mach3 software to drive.
I've had it running for a day so far, but I'm in trouble with the repeatability of the travel. I'm trying the Mach3 feature of asking for a test motion in an axis (I used 5 inches as a test case), then measuring the travel, telling Mach3 the answer and readjusting the steps per inch value for the axis. I then repeat this to check the motion is correct. The problem is I get a wide range of results, for instance I input 5 inch travel, measure 5.14" travel, Mach 3 updates the steps per inch and I try again and get 4.85" or 5.3". from my view there could be two issues, One; loosing steps in the process, or Two; slippage in the drive between the motor and the lead screw. I currently have a short length of hose clamped to the motor shaft and lead screw end. I 'm planning to switch to a real flex coupling as soon as it arrives. I don't know what to do about loosing steps, do I slow the speed down, (I'm at 6 inches per minute right now), or what?
As a separate question, the motor seems to get quite hot after doing these five inch jogs and back with pauses while I measure. After a half hour of this the motor is hot enough I don't want to touch it. Is this normal?

Thanks for any help in advance.

Ivor

[ger21]

First thing you need to do is make sure you're not getting any mechanical slippage. You need to rule out one thing at a time.

When you change the steps/unit, make sure you close and restart Mach3 before you test again.

What voltage are the motors rated at, and what voltage are you running them at? Are you using proper resistors if using a higher voltage?

[fuelsguy]

Gerry,
I certainly agree on the ruling out one thing at a time.
Thanks, yes I thought about slippage, I have changed the hose/clamp connectors to a softer rubber and there doesn't seem to be any slippage. I have better couplings on order to get rid of any concern there. I am planning to do some slippage tests by putting a pointer on the motor and one on the lead-screw and seeing if they track.
I haven't thought about shutting down Mach3 each time as the new re-computed steps per inch shows up in the appropriate spot after the test /re-compute has finished. I'll try that right now.

As far a motors etc, I'm using a Stepperworld FET4 system, with 23HS8303 motors rated at 6.75v and 1.5amp current. Per Stepperworld inputs I'm using 12v as the driver and 1.5 Ohm resistors as supplied by them. I have Aluminum heat sinks attached to the resistors but they do get hot.
as I said in my first post, the motors seem to get hot in a hurry which worries me especially for doing long run times later.
I'm also looking at the friction in the carriage, I don't have a good idea on what is OK, other than lower friction is goodness.
I'll go do some more checking.

Thanks for the inputs.
Ivor

[fuelsguy]

More data:
Checked for slippage using two pointers, one on the motor end shaft and one on the lead screw, I couldn't see any slippage.
I tried varying the drive speed of the motor using Mach3 settings. with a steps per inch of 3000, I could run at 15 inches per minute but above 20 or so the motor would just sit there and stutter. At 15 inches per minute the motor would run OK but there seemed to be a lot of hesitating as it turned, wasn't a nice smooth rotation. I tried it with the motor disconnected from the lead screw and got a similar result. I'll try the same experiment with the test software from Stepperworld. I would have thought with no load I wold get a very smooth rotation, but have no experience with this kind of stepper motor.

More experiments tomorrow.

Ivor

[ger21]

Does the Stepperworld do half stepping, or full stepping only? What are you using for leadscrews?
I have a feeling that you're reaching the limit of your setup.

[fuelsguy]

Gerry,
Half or Full stepping? I don't know, I'll have to dig into the Stepperworld documents. Can I change Mach3 from one to another to test?

Lead screws are ACME 5/8" X 10 TPI single thread, Delrin nuts, ball bearing supports at each end.

Tomorrow I'll test run the motors using the Stepperworld software to see what speed I get with Mach3 out of the loop.
On your comment about limiting out my set-up, any ideas on what to do?

Ivor

[ger21]

Where did you get 3000 steps/inch from? 10 turns/inch x 200 steps/rev = 2000 steps/inch using full step. It may be full step only, as the website didn't mention anything about half step. Mach3 doesn't know if it's half or full step, it just needs to know steps/inch.

If you want to go faster, you have a few options.
1. Change to screws with fewer turns/inch. 1/2-10 2 start should double your speed.

2. Increase power supply voltage. You'll also need bigger resistors. The resistors will give off a lot of heat. The Stepperworld drive is not very efficient, and will waste a lot of power as heat through the resistors. Not sure how much voltage it can handle, but your speed will be roughly proportional to the voltage. Double the voltage, double the speed. Just about every homebuilt machine you'll see here is using at least 24V.

3. Get new drives. A hobbycnc drive running 36V might give you 100ipm. A Gecko G540 and 48V even more. Change drives and screws and much more speed is possible.


Keep in mind that there are a lot of factors involved when determining speed, and these are just rough estimates.

[fuelsguy]

Gerry,
First, thanks for the help,it's really appreciated, just having someone else to bounce ideas around is great.
Some progress in isolating the issue:
I disconnected the motors from the leadscrews, now thereisn't any torque issues.
I tried the test software from Stepperworld, and each axis works correctly. I input 200 tsteps and I get 360 deg of revolution. Input 2000 steps and I get 10 revolutions. I went to max speed and it came out at around 100 rpm.
Tried the same with Mach3, got a variable rotation about 60% of waht was commanded. This is why Mach3 wanted me to use 3000 steps per inch instead of 2000.
Next I tried a short wire harness to see if there was some interference that Mach3 was confused over. Got the same results as before, so I don't think it's in the wiring or board. I suspect there's something in Mach3 that I haven't set correctly so it's confused about something.Maybe I should reload Mach3 completely and try again. I have fired off a post to the Mach3 forum asking for advice.

All for now, go enjoy the weekend!

Ivor

[ger21]

Follow Hood's advice from the Mach forum and increase the pulse width. If 5 doesn't help, try 10 or 20.

[BobF]

I am surprised at the 100 rpm max speed, but I looked up the stepperworld stuff on google. The power supply puts out only 5 and 12 volts so that is probably a large part of the problem. At 100 rpm, you will only acheve 20 ipm with the screws you listed. Sounds like Gerry is right (he usually is) you have hit the limit of this driver/power supply. Get another driver/power supply and I bet you can get a lot more use out of your motors. The motors on the website look like 6 wire motors, but its hard to tell. They look like they should work with the hobbyCNC boards and maybe the Gecko G540. If so, you could get at least 800rpm with one of those setups.

[fuelsguy]

Continuing my learning saga, I tried the pulse width increase and it works. Thanks for the suggestion. I ended up with a 10 microsecond pulse width and that seems to have cured the problem.
Following up on Gerry's comments about upgrading is there a recommended place in this forum where I can get more information?
The speed I get from the current configuration is slow and I can always use the Stepperworld board in a foam cutter for model airplanes.
As a side question, the motors and the resistor seem to get pretty hot even when not actually moving the unit around, is that normal in these kinds of set-up?

Again, thanks for the advice,
Ivor

[fuelsguy]

Gerry, and anyone who wants to join in...
You proposed an upgrade to the system using a Geckodrive 540. My router is set up to have travels as follows: x=46: y=22" and Z = 15". I have checked torque relatively crudely, by attaching a lock wrench to the outside motor shaft and spinning it by hand. I can spin all three axes with one finger so friction doesn't seem to be a major issue. In doing a little reading of the forum a recommended motor would be the Vexta Stepper MotorPK266-02A. I have a 36v power supply that I use for cutting foam wings, so a 540, three of the Vextra motors would seem to be a good set-up. Any thoughts on what else I would need to make a useful set-up? I plan to add a fourth axis as a rotary spindle not as a lathe but more for making detailed shapes like model fuselages or complex chair legs.

Any thoughts would be appreciated.

Ivor

[ger21]

Are you going to keep the 10 tpi screws?

[fuelsguy]

Gerry,
Re. lead-screws, at this point I would say yes unless there is a really compelling reason to switch.

ivor

[ger21]

Is 2-3x more speed compelling enough?

[fuelsguy]

Well, if I can get 500-600 IPM out of my current lead-screws with the right motors I would say no.
Ivor

[ger21]

You'll be lucky to get 1500 rpm, which would be 150ipm with 10 tpi screws. Most consider 1000 rpm very good for a stepper driven machine.

[fuelsguy]

Ah well, back to the drawing board to see what I can do.

thanks for all the help,
I'll let you know how I get on.

Ivor

[ger21]

Also, consider that the screw may not be able to rotate that fast without whipping a lot. A lot of people find that screw whip is the limiting factor in the speed they can attain.

[fuelsguy]

Gerry,
I think I'll stick with the 10TPI for now, just to get some experience with the machine and then I can upgrade to what makes sense at that time.
I did think about using a 2:1 gear set-up but I take your comment about whip. I have built an RC airplane using a long shaft drive to the propeller and shaft whip was a big issue in the design.
At this point I want to get a good reliable machine making sawdust and move up the learning curve a bit.

More to follow when I start cutting wood!

Ivor