@mrehorstdmd

There is a Kinimatic bed mount system available for the Railcore produced by Mandela Rose Works:
Kinematic Mount

@dc42

here is a more up to date config, that is actually printing right now (with shifting) config (1).g

and here is a the console output after running a few M122 commands for the 1XD boards and the main Duet console.txt

This was while printing the standard Benchy model. no hiccups at all. But plenty of shifting, there are no skipping/slipping belts.

I use Notepad++ works flawlessly.

@dc42

4.99v on 40
4.97v on 41 Likely due to the R18 and R20 tolerances.

Did the same test with the servo encoders with the following results:

RAS 6.2	RAS 6.2		
3600 pulse/rev	3600 pulse/rev		
			
Servo Encoder at X0 Y0		Driver board position at X0 Y0	
A	B	                B41	B40
-12564	23112	               -49695	13499
-12289	23124	               -49695	13499
-12031	22945	               -49695	13499
			
no change with or without RAS			

Unless you have any other ideas I feel like the Clearpaths are not compatible with the 1XD board for some reason, Ill contact Teknic to see what their thoughts are.

@dc42

After more research I have found multiple instances where Clearpath Servos are very picky about the signal voltage. Ive designed a small level shifter board and sent it off for a prototype run, I hope to have it in about a week or so. I will be shifting the signals to 12v.

Once the boards and components arrive Ill test the boards, then test them with the servos and ill update.

@dop said in Clearpath Servos with 1XD Expansion:

For others who find this thread: I just installed beta 3.3 and it's working correctly now, looks like root cause was in can communications.

If it was an issue in CAN, id assume that issue has been in all of their products that utilize CAN. its unfortunate that they left bugs like these in their code for so long unnoticed.

@dop I fixed it by abandoning it and going with a different hardware solution. My assumption after lots of trouble shooting is that the clearpaths en/stp/dir signals needed to sink more current than the 1xd could provide, or that it was a CAN issue. I ended up developing my own board that worked flawlessly.

@dc42

After more research I have found multiple instances where Clearpath Servos are very picky about the signal voltage. Ive designed a small level shifter board and sent it off for a prototype run, I hope to have it in about a week or so. I will be shifting the signals to 12v.

Once the boards and components arrive Ill test the boards, then test them with the servos and ill update.

I use Notepad++ works flawlessly.

@dc42

4.99v on 40
4.97v on 41 Likely due to the R18 and R20 tolerances.

Did the same test with the servo encoders with the following results:

RAS 6.2	RAS 6.2		
3600 pulse/rev	3600 pulse/rev		
			
Servo Encoder at X0 Y0		Driver board position at X0 Y0	
A	B	                B41	B40
-12564	23112	               -49695	13499
-12289	23124	               -49695	13499
-12031	22945	               -49695	13499
			
no change with or without RAS			

Unless you have any other ideas I feel like the Clearpaths are not compatible with the 1XD board for some reason, Ill contact Teknic to see what their thoughts are.

@dc42 ill check them when I get off work today, and I'll probably shoot teknic an email since they state 5-24v for signals and get their input.

I assume the 5v regulator also provides power to other ICs on the board?

Would it be possible to bodge a simple board that sends 12-24v to the servos and use a resistor divider back to the 1xd?

Is the 1xd schematic available yet, or will it be?

@dc42 correct I have all positive pins connected to the 5v on the header with the dir, step, en pins. And those pins are directly attached to the servos.

The clear path manual does mention 5v minimum for the inputs and also states that less than that may work initially but will eventually not. I'm thinking the less than 5 volts is the issue here.

I mentioned the currents because the 1xd page mentions that they can only sink 10ma, and I assume that is through a 500ish ohm resistor. If that resistor is a 5% tolerance that would put the max current bellow 8ma.

@dc42
Absolutely:
I removed Homing from the gcode file, but I dont have a start,stop, or cancel.g

here is the console text, there was one instance where I didnt do a G1 X0 Y0, but you will probably see that. It looks to me that the 1XD boards are receiving and sending commands correctly. please verify my observation. console.txt

Here are the current ratings of the clearpaths:
5 VDC - Min 8 mA
12 VDC - Min 9 mA
24 VDC- Min 12 m

All of the pins are rated at 10ma/Pin, what is the max of the 5v common?

If that sink current is limited by a resistor its very possible that we could be under 8ma@5v if using a 5% @ tolerance resistor @ 10ma Max per pin.

@Teknic_Servo any info you can add to this?

@dc42 copy that

@dc42

#2 is interesting.

Would it help if I hook up some external 2209 or 5160 drivers to the 1xd boards and swap in some steppers to see if the results change?