Stepper missing steps tried PLC968 not fixed

[Toysrfun]

Attached is report. Quick breakdown of machine

Bench Mill with Linear Rails, 5 pitch screws Nema 34 800oz 1.8 deg 3.5 amp motors. Spindle is 3hp R8 7500 rpm belt driven and controlled manually since i dont believe the vfd can connect to centroid. NO limit switches connected at this time it has an estop only. Drivers are LEadshine DM860A. IVe also tried a set of Nem24 400oz, Nema 34 1600oz and nothing is working and they all miss steps. In fact they miss even worse doing a bench test but under load improve. Ive tried every dip switch setting and mirrored those settings in the wizard, When i lowered the KHZ in the PLC it was a noticeable improvement but not a fix. Ive also tried every option in PLC 968. This goes back to previous issues ive had with other builds where ive never been able to utilize open loop steppers. Running AC Servos they always work fine but i dont have any small enough here right now to bolt up. All axis are missing steps. Any insight would be helpful. Steppers are direct drive and even at a 25ipm miss significantly i will attach photo for further explanation.

[Toysrfun]

Also using a 70v Toroidal power supply. Shielded and stranded cable, misses steps even with spindle off or free air operation. expected since bench test misses as well.

[Toysrfun]

Here are some photos. You can see the circles keep moving over as it changes doc. I did a free air with a marker and it’s better so obviously torque seems to be very low even at low speeds causing missed steps. Bench test misses steps and and pressure on the table such as myself leaning against it. Nema 34 1600oz behaves identically. The 800oz are 8 wire motors and low inductance and the 1600oz are 4 wire higher inductance and the 400oz nema 24 were in between inductance.

[Toysrfun]

Here is a more complete photo of machine. Btw the axis, base, rails, etc were all machines on my atrump centroid bed mill. The rails are all within .0001 to .0003 far as accuracy. It’s very smooth and a fingertip can glide the table around.

[Gary Campbell]

Do you have a picture or schematic of how the drives are wired?

[cnckeith]

missing steps are caused by:

1.) bad wiring practices
2.) mechanical force (drag) exceeds stepper motor power at a given rpm, therefore it skips position (lost steps)
3.) running a stepper motor at high RPM speeds where it has the least torque, therefore it skips position (lost steps)
4.) running Acorn at a step frequency higher than the stepper drive can handle, therefore it skips position (lost steps)
5.) motor size and drive size ( or amp settings) not matched, therefore it skips position (lost steps)
6.) bumping the axis into a hard stop
7.) hitting Estop when axis is rapiding and then the axis coasts after the estop condition.

and there are other reasons.. put it this way if you are losing steps it is not Acorns fault, Acorn's step output is solid.

an easy solution which we highly recommend is use closed loop stepper motor/drives and/or AC brushless motors and drives.

that being said....open loop can be used with great results if the wiring is done right, the motors are sized correctly, the drives are on the right settings and the machine force to move doesn't exceed the motor power.

[ShawnM]

that being said....open loop can be used with great results if the wiring is done right, the motors are sized correctly, the drives are on the right settings and the machine force to move doesn't exceed the motor power.

Adding to that, if you drive any stepper outside it’s speed/torque curve it will lose steps. Bigger is not always better. Triple check your wiring, voltages and drive settings. This is definitely not an Acorn issue. Check your motor wiring, how are you wiring the 8 wire motor? Series or parallel? This changes the inductance and drive/motor performance. Are any of the motors getting hot in any configuration? There’s a lot going on here for all 3 axis to be missing steps.

I’m running open loop, NEMA 34, 495 oz/in steppers on a 48 volt power supply on my Acorn powered router moving a rather heavy gantry with a ridiculously heavy 4.5 hp water cooled spindle and it doesn’t miss steps. I cut aluminum with it almost exclusively. Take a step back and work on one axis at a time and triple check everything.

[Toysrfun]

Sure here are a couple photos. I’m not new to this and have a year college course of low voltage electronics/dc wiring and 20 years wiring automotive computers everyday then programming so the premise is much the same and these are far less complex to wire. I’ve built 4 different projects so far and rebuilt my centeoid Atrump and it’s all fairly easy with a some issues that can arrive when mechanical problems exist like the atrump had so I’m not new to figuring out all the problems down to the ways when applicable and have the indicating equipment here to check all of it. So as you can see it’s a Simple step/dir with no enable connected. I route all my 5v power wires back to the source as well the ground so it appears 4 separate wires but obviously not. Far as Keith’s comment he obviously didn’t read my post or just tosses out generic hoopla rather than being helpful and more so insulting because we know centeoid couldn’t be the issue until a windows update decides it’s incompatible and I lose my folder with my license keys all in a week of building that pc. Still haven’t received a call back despite many attempts to get that help. Sorry Keith but I did leave detailed info covering most of what you stated so please understand repeating the obvious is frustrating to me. I’ve been very patient with centeoid controls and overall I really like them. So not all steppers are instant go with centroid and the fact you don’t see them all over YouTube probably backs it up. Now that’s not supposed to sound hurtful like ignoring my comments that the table is precisely fixed and moves around with a finger tip so no nothing binds at all. Same issues here I had with a commercial router I had bought with steppers that originated with Mach and never worked with centeoid until i installed some AC servos. If you recall you verbally reprimanded because you didn’t know have a schematic for me to connect them or any helpful information because I didn’t buy clearpath servos as though $165 each for my servos was cheap. That servo issue by the way was the fact they use 24v supply and not 5v which I did figure out despite lack of manufacturer literature and the router works fantastic with exception of probing which I’ve given up on beyond tool set and zero set. Anyways I did connect a couple AC servos to this machine that I have here to verify my centeoid incompatible stepper history and sure enough they move just fine however they’re for another project and require I make permanent nema 40 frame mounts. Not happening for this little machine unless I buy smaller frames motors. Only thing I can suspect is it goes back to the software settings and probably specific to PLC settings if more exist for stepper tuning. It would be great to have a list of what they all do ive searched online and found nothing. Tomorrow I will go back and test with mach3 on this since I keep a backup pc here when issues like this occur so I can always prove Mach to work but in fairness they do have the ability to tune servos for optimal use. I did try the standard leadshine DM canned Wizard and it only made issues worse. The more I microstepped the torque dropped off. The lower I set it the torque increased. When the frequency was at 100,000 it was too high and the problems went away as I lowered it to 20,000khz. Again the steppers as mentioned earlier Keith will miss steps doing a bench test not connected. Furthermore taking the machine out of the equation. Wires are paired to poles obviously or steppers wouldn’t move at all. I’ve also tried swapping them around to opposing A/B and no difference. These 800oz 8 wire steppers were also on a former Zay build I had using standard lead screws, dovetail ways, and Mach 3 and despite all the backlash comp it worked ok for a slap together machine and could cut at 100ipm all day long. The thing that threw me off is the fact when they were installed they didn’t miss as many steps. All my grounds go to a fixed din style grounding point and no ground loops exist. So you can see I’ve kept the install simple I don’t care to fully connect limit switches or several input outputs until I see the mill is functional. This should limit the possible issues diagnosing and so should the fact a disconnected bench test causes missed steps.

[Toysrfun]

To further clarify the black and red go back to the meanwell 5v supply. The green and white goto the board for step direct. My power and ground for the drives come from the 70v toroidal supply

[Gary Campbell]

I always seem to get a bit suspicious when someone feels the need to list their qualifications. Im not sure how you get anything to run with that spiderweb of connections.

Try this:
1) Set drive to 1/8 (1600) steps per rev Match in config setup
2) Connect one motor to the DB25 header (as recommended here, time and time again) using shielded wire
3) Set step pulse to 100,000
4) Invert the signal to step and direction

I have DOZENS of systems WORKING using the recommended connection with those exact and 4 other models of LeadShine drives