Open Collector vs. DB25 for SERVO DRIVE

[PDS150]

Hi all - apologies up front if this was already covered and I missed it, but I couldn't find this specific topic

I have a CNC router from China that came with an XHC controller, and ran Mach3 which I didn't like (the controller or the Mach3)
The controller did have differential outputs, which was nice, and they were 5v signals

I converted to Centroid Acorn after a week or so of fighting issues with the stock controller and it's been mostly awesome since then, but there's one issue that has been in the back of my mind:
My motor drives are Leadshine H2-2206 (AKA HBS2206) and the closest thing I can find for a manual is the Leadshine ES/DH series. I know my motors were controlled with 5v based on the previous XHC controller, and this manual also recommends 5v control.

Per my Acorn manual, I am using the H2 and H3 screw terminals since it's a SERVO drive and not a STEPPER, and have it hooked up per the attached schematic. Most of the literature shows the open collectors using 24v not 5V, but my limited understanding of this circuit is that it is an open collector where really the voltage is whatever I decide to provide???

Upon startup, I couldn't get the motors to respond to move commands - some would move and sound terrible, while others didn't move at all. I researched the issue, and determined that the communication frequencies available in the wizard (400kHz/200kHz..) may be too fast for my drives, and that a larger selection was available under parameter 968 (THANK YOU FRANCO CNC!! I was able to achieve smooth motion at 40,000 Hz (but no faster) and put that issue behind me....
-EXCEPT - I don't know where exactly I may be capped [4000 steps/inch - 300in/min = 20,000 Hz per axis?]

Further reading on this site has led me to examples where servo drives are using the DB25 connection, not only stepper drives, and there are threads here that discuss the signals being cleaner from the DB25 outputs than from the H2/H3 terminals. Also, it seems like an uncommon practice here to use the screw terminals for 5v rather than 24v, but I don't see why that is exactly....AND the DB25 is already 5V like my servo drives recommend. I am very confused now as to why a step/direction drive input would be any different for a servo that it would be for a stepper.

If all looks good (the machine does seem to be running fine at 40khz) then great, I just wanted to tap into the experience on this site to see if there was a better way than what I chose or if I'm doing something to cause instability over 40kHz... Thanks in advance!

[Gary Campbell]

I would test one motor -drive combo on the db25 under conditions that were known to cause issues and see what the results are

[cnckeith]

i'd use the DB25 5 volt ttl logic step and direction signals for the drives you have, follow garys advice and hookup one drive only.

[ShawnM]

And post a current report for us to look at so we can see your current settings. Something sure seems off as even the crappy Leadshine clones run great at a 100,000 step rate.

[PDS150]

Thanks guys - that all makes sense. I’ll try to post a report tomorrow.
Pete

[PDS150]

Here's a report - not sure which files are useful here, so just posting the whole .zip file:

[rruizvial]

5V is open collector with resistances.

[PDS150]

After thinking about the issue some more I have a theory.. This is a CNC router with a cabinet the width of the machine (5ft), and the controller is 4ft away from the servo drivers (so 5 or 6 ft. cables). Also, they had the Y motors electrically paired (wires connected together and plugged into a single channel with the drives wire opposite each other for counter-rotation). The previous drive had differential outputs, which is the one thing I really liked about it - and which made it basically immune to noise at the distances we're talking about here.
In switching to the Acorn, I am sending single ended signals over the same distances and sharing one of those signals across two cables feeding two drives, which would explain why the Y was the worst behaving axis of the three. I will put an oscilloscope on those lines before changing anything, but I have decided to switch to the 5v TTL outputs on the DB25 connector, and to feed them immediately into a differential driver, then out to the drives.
I suspect I am currently suffering from signal noise/loss/attenuation through these lines, and that pull-up resistors could possibly sharpen them and compensate for sluggish rise times from the capacitance/impedance of the long, and doubled wires - but at the end of the day I feel much better about converting to differential signals and knowing it's as solid as I can easily accomplish.
It appears, from the schematics and manual, that it is less common to use these screw terminals with 5v. 24v may give a larger signal margin, but my drives are spec'd at 5v. I'm also guessing most builds have the drives mounted much closer to the Acorn board so those are possible reasons I am seeing these issues more so than many others.
I'm still unsure why the screw terminals are recommended for servo drives, and the DB25 for Stepper drives. I'm probably missing something here but it seems to me that the drive inputs would be very similar for the two applications and that it would be more important to match the required drive voltage - meaning if you need a 24v signal for your drive you need to use the screw terminals (again - I'm sure I'm missing something here and over simplifying)

[cnckeith]

Acorn "use screw terminals for servos" is a general recommendation and applies to ac servo drives from the big boys, Yaskawa, Delta etc..

many stepper motor companies have started to make AC servo drives and they often use their stepper tech in them.. not all but many of these china AC drives work better on the Acorn's 5 vdc TTL step and direction even if they claim the drive works with open collector.

so a a better description would now be:

Use Acorn's 5 vdc TTL step and direction output for stepper motor drives and AC drives designed by stepper motor drive companies.

until they improve.....

[PDS150]

Ha! I like that description better