Hello Guest it is May 22, 2024, 03:46:53 AM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - thosj

I think it CAN be run from a non-administrator account, but I'll defer to the experts! I was just suggesting to TRY that in Dave's case, as it's often what Bryanna suggests when requesting support for such issues.

Bryanna will say, "run Mach4 as administrator." Not just from an "administrator" account, but right click, Properties, Compatability, Run this program as administrator.

I don't know whether Teckincs have provided a similar facility for their Clearpath servos.

ClearPath DOES have exactly this. Hook a USB cable to the drive, start their software, MANY, MANY selectable options for the "scope." I've discovered that this runs fine while Mach is running and controlling the servo. The problem here will be the dummy (me!) watching the software!!! I was smart enough to figure out Auto Tune, and how to set the error mode, and now, to figure out the following error setting, but the rest seems pretty complicated to me.

I got the ClearPath servos to get away from this PID tuning. The Gecko's have pots and I could never get the Z motor to not buzz all the time, over 8 years of reading posts on how to do it. I actually replaced the encoder, drive, and motor and couldn't get it. I tried CNCDrive servo drives with software PID tuning, but manual, and couldn't get the PID set AT ALL, finally returned all 4 and gave up. Bought ONE ClearPath for the Z, a PS and for future expansion, a Power4-HUB, and it worked flawlessly after Auto Tune. Thought about it for approximately 3 seconds and bought 3 more, done with buzzing servos. My system is WAY better than it's ever been, Mach acceleration higher, top speed higher, NO buzzing/screeching. I could perhaps squeeze more out of it, but I'm happy with it the way it is and I'll tell you, I'm inclined to stop chasing it and just enjoy.

OH, and the mechanics of my system did NOT change, same ball screws, same belt ratio's, everything the same since the Geckos were first installed in 2009 when I converted the machine.


Hey Tom,

Glad to here you got it working.

Unrelated question. Assuming your fourth motor is not for a fourth axis, how did you go about tuning the slave motor? I've been considering upgrading to a dual drive for my X but was not sure how the tuning portion worked.


Sorry I can't offer up any help, my 4th axis is a 4th axis. I have a knee mill, Z is the quill and A is the knee. I simply, (well, not TOO simply), use the knee for tool length offsetting.

That said, I have seen, somewhere, something about tuning a slave axis. I can't recall where, but read it somewhere. You might try support@teknic.com, they're responsive but due to their large array of available servos they have a difficult time understanding YOUR problem!!


Thanks to Craig, I got the HLFB's working on my PMDX-108 Input, Port 2 Pins 6-9. Code like Craig's but adjusted for the Inputs I used, 4 sections of code, one for each axis so I can easily tell which axis errored.

That said, soon as I moved two of the 4 axes I got the error. I hooked the USB cable to the ClearPath's and started their software. I have the ClearPaths set to 6400 counts per rev. which in my system equates to 96000 counts per unit (Inch for me) on X/Y, 153,000 counts per inch on A, and 78222.22222 on Z. The count error was set to 10 in the ClearPath software, a default, I guess. That's ~ .0001" on X/Y/Z, .000065 on A, so I set them to 100 and everything seems OK.

That made me wonder if there's some magic following error percentage one should use, like 1.5%, which is what I have, approximately, arrived at by guess and by gosh? Or something else?



You might look into a PMDX-108 Input. You're not using the ESS port 2 now, right, even though it's connected? The 108-Input goes between the PMDX-126 and the ESS and adds inputs from port 2 pins 2-9. A bit more than a C10, but very nice and keeps it in the PMDX family!!!


People FAR more knowledgeable then I have weighed in, I'll be going the separate input route when I to this, seeing as I have them available!!

Dustin, do you have something one the SECOND port of the ESS/PMDX-126? I have a PMDX-108 Input on there with only 2 or 3 used, so enough left to hook the HLFB's up and code them separately. You could use any small BOB, like a CNC4PC C10 for using the ESS 3rd port.


Following with interest!

Isn't the serial wiring of the HLFB signals what Brett was alluding to when he said "daisy chain"?

And, is it not simpler to have, in my case FOUR, drives wired to one input, one set of code, or is there some reason to use 4 inputs and 4X the code just because you have these inputs available? What is to be gained with the multiple input method?

Dusty, where IS your code in the screen set?


Let me just say this, the stuff PMDX has mastered, they have mastered well!

So now, get on with a ClearPath to PMDX-126 adapter board:)

Thanks for all the effort you've put into this, Craig. I'll ponder it all but I'm leaning toward and simpler input solution. BTW, I have 4 axes, not three, not that that matters!!

My physical Estop button has two contacts, one to the PMDX-126 and one to the ClearPath power supply. I also have a latching relay on the power supply input, I press a momentary button to activate the power supply. I can start the PS either before or after Mach4 is running, doesn't matter. But if, like you say, the ClearPath's are "de-asserted" and Mach4 is started first, I can understand there might be an issue.

I won't have time in the very near future to get at this, but definitely will get to it. Reading your advice, I'm leaning toward the inputs/programming method and not trying to use the Gecko ERR/RST input/outputs.

Thanks again,