PMDX-126 and Clearpath - Drive signals are too low voltage.

New build

Clearpath SDSK motors aren't responding to step and direction signals from the PMDX-126. I'm using Test mode of the PMDX-126.

I don't see the signals from the PMDX-126 in the clearpath software gui, not completely I should even be able to, But FWIW the drives work fine using only the MSP software. Auto Tune and jog work fine.

What I can see visually is that the LEDs on the PMDX board are blinking as I would expect, DIR coming on and off with each press of the test button (while in TEST mode) and STEP barely illuminating and getting more intense as the pulse train ramps up. It matches what I see on the scope.

I looked at these signals with a scope looking for pulse duration (which is fine) and just to better see what was actually happening and thats when I saw the low volt signal. I understand that excess current would drag the signal down, but the specs look like the board should be able to provide what the Clearpath requires and I know some other folks have this working.

I don't have a current probe for this scope so I can only read the DIR signal's current with a meter, I can do that tomorrow just to be sure its not dragging the voltage down. FWIW I did reverse them with the config dip switches and the results were the same. Its also of note that the PC +5v pin is fine the whole time.

The Clearpath motor is set for Step and Direction -doesn't appear to have any other options available, they are greyed out in the software. I never saw any way to tell it what signal types to look for but I was hoping there was such an option in there somewhere.

Set for 800 Pulses per rev, as delivered.

Again, the pulse train occurs as it should and direction toggles properly as well, its just that the voltage on the signals is too low. Anyone else seen this or did I connect or jumper something wrong? I saw in the book that in some cases a 1k resistor is placed across the signal and ground, but that would only cause more current loss.

Ideas please?


https://www.pmdx.com/PMDX-Forums/ind...tach=729;image

How do you have the servo's connected to the PMDX-126?

PMDX has the best support in the business.
Give them a call tomorrow.

The CP servos need more than 5v to work when wired up for a "differrential" signal. Sounds like the PMDX board differential signal is referenced to 5v. See the engineers note on the bottom of page 46 of the CP manual.

The way to get around this is not to use the differential wiring connections. On my UB1 board all the (+) inputs were connected to a single terminal. Then the individual (-) inputs were connected to their corresponding board outputs. I don't know about the PMDX board but I suspect either your positive or negative signal inputs will be connected together or, share a common terminal.

Sent from my Pixel XL using Tapatalk

Quote:

---

Originally Posted by 1Jumper10

How do you have the servo's connected to the PMDX-126?

---

Brown, Yellow Org, Red to Gnd (Neg side of all)
Black (InputB) to J4-2 (Step)
White (InputA) to J4-3 (Dir)
Blue (Enable) to PC+5V - motor LED responds to this signal
Green (HLFB) floating - gotta work out the rest first

Quote:

---

Originally Posted by ger21

PMDX has the best support in the business.
Give them a call tomorrow.

---

I was eager to make a motor move before then. I only get weekends to work on this.
But I did talk to them before I bought and other than making a good case for Mach4 vs Mach3 (which I already owned) They said all my choices (126 with S/S to C/P servos) were fine.
You are right, they are great to talk to. Really helpful. Just wish they had no life so I could get an answer 24x7 :)

Quote:

---

Originally Posted by 1Jumper10

The CP servos need more than 5v to work when wired up for a "differrential" signal. Sounds like the PMDX board differential signal is referenced to 5v. See the engineers note on the bottom of page 46 of the CP manual.

The way to get around this is not to use the differential wiring connections. On my UB1 board all the (+) inputs were connected to a single terminal. Then the individual (-) inputs were connected to their corresponding board outputs. I don't know about the PMDX board but I suspect either your positive or negative signal inputs will be connected together or, share a common terminal.

---

Saw that note too. I assumed the 126 was sourcing the signal with a transistor not driving it with an op amp. But I didn't trace the circuit. If that is the case I am going to need a buffer circuit to drive the motor, I would have thought PMDX would have said something about that. I wouldn't have gone down this road with that caveat.

If you reconsider your board, I've had good results with CP servo's and a UB1 BoB from cncroom.com. Regardless, it sounds like you have it wired up right and I hope you get it sorted.

Quote:

---

Originally Posted by 1Jumper10

If you reconsider your board, I've had good results with CP servo's and a UB1 BoB from cncroom.com. Regardless, it sounds like you have it wired up right and I hope you get it sorted.

---

From cncroom's page re: UB1

Quote:

---

• New! Differential line driver for motion signals, allows for longer wiring with more resistance to interference when compared to TTL open end. Also can be used as single end both source and sink connection.

---

Looks like I would face the same problem. Diff mode is specifically warned against in the CP docs. I just can't win.

No, you'd be fine. I used the UB1 and ClearPath's on mine and I've had no problem. You just can't wire them in differential mode. I have a wiring diagram for the UB1 and CP' s that I can send you if you're interested.

Sent from my Pixel XL using Tapatalk

Thanks for the offer - I would love to see a scope pattern of the signal levels you are seeing on your installation that does work.

I hope to fix the current investment, but I will look there next if PMDX can't deal with this.

I'm a little annoyed this wasn't brought up when I spoke to PMDX. Thats why I think its my doing, these are reputable folks with a good history, surely if there was a problem they would have mentioned it. .

Placeholder

I want to commend both Clearpath and PMDX for taking a serious interest in this. They both jumped on it with vigor, and as stated, the problem is abated. I'm completely satisfied in thier sincerity and integrity. And even though my measurements were and are correct, 3.8-4v signal with 6mA draw, the motor responds to them fine.
The source of the issue I was experiencing was apparently due to tuning the motor at 25% power. For some reason I never looked into, tuning it at that power left it unable to respond to the inputs. Too weak maybe. I autotuned it again at 100% power and it worked after that. This is further confirmed by the fact that the other 3 motors that I had not autotuned on this rig responded properly.

BUT, I do have a disagreement. PMDX (Steve) confirmed that the voltage I was seeing from the PMDX-126 board , 3.8-4V (not 5V) is completely normal and expected because of the drivers they use on the board. They are not TTL levels. I did mention this to Teknic as well. Tyler I think.

So, The motors work, but I am concerned that as stated in Teknics manual I will experience problems later on due to age and heat. Only time will tell. I have yet to find evidence that this occurs anywhere so at this time I'm not going to worry about it.

Hi,

I'm an applications engineer with Teknic who worked with PMDX to help troubleshoot this issue. I thought it might be helpful to readers of this thread if I briefly explained the ClearPath nominal (versus min/max) input voltage specification to clarify any misunderstanding of the specs.

ClearPath inputs will work properly with logic high signals ranging from 5-24VDC, nominal. The maximum logic high voltage is 28VDC, (to prevent electrical damage to the input). The minimum guaranteed logic high voltage is 4VDC (assuming that enough current is sourced to turn on the input's opto isolator LED). The reason for both the nominal and the max/min specs is that most 5V nominal systems experience some amount of voltage droop across wires and interconnects, so a signal that starts out as 5.0V at the supply is never actually 5.0V when it reaches the motor's input. As the signal voltage decreases, there will be a point where it is no longer high enough to trigger the servo drive’s input. Our specs are designed to keep the user comfortably away from that point, but I can tell you that the spec is conservative and I believe 3.8VDC will work just fine.

Regarding differential signals, as 1Jumper10 commented, differential signals work differently from single-ended signals. The reason differential signals are not recommended for use with ClearPath motors is primarily because 5VDC differential signals have a resultant (post-math) voltage swing of significantly less than the 4VDC as seen by the ClearPath input circuitry. Although not advertised, it is worth mentioning that a 10VDC (or higher) differential signal, fed into + and - inputs of a ClearPath motor, will work.

Based on conversations with PMDX, their 126 BoB (breakout board) outputs regulated 5VDC driven signal, and PMDX verified that their product is used with ClearPath motors in many applications. (As a side note, we found the folks at PDMX to be very helpful and professional.) I know you recorded lower voltage, but I can't really speak to that.

It sounds like the issue with the PDMX-126 BoB has been resolved. If this is not the case, or other issues arise, please feel free to call or email [email protected]. Thanks!

Best regards,

Jim W. - Teknic Servo Systems Engineer