Need help wiring the "Clearpath killer" for Clough42 ELS - iSV57T-180

[GunsOfNavarone]

I hear ya, and that's why half of me is...."if it's not broke, don't fix it" but I would like some more torque...

 

[xr650rRider]

Well I got everything wired up to test, and it seems that I must have blown another of those TI boards. I had it working with the display the other day (without the encoder or motor connected) and was super careful about static control (anti static mat/wristband, etc). I tested and retested every connection from the boards, through the connectors on my enclosure, to each device and all sounded out correctly. Long story short, I can't offer any insights into configuring the motor as I never got it to turn. It powered up fine, but the TI board seems dead - again. I swear, I have programmed and installed over a dozen arduinos and a few raspberry pi devices and have worked on countless PCs and never had this problem.

Does it display the Clough42 and version number when you power it up on display? Or nothing? Those spring loaded connectors can be a pain to get a wire under them (better to use the crimp pins) but it won't fry the board.

 

[xr650rRider]

Funny, I put that in the comments of the Clearpath killer video...would he be doing a follow up video for all of us Clough ELS people that you KNOW are going to do this. This wasn't a walk in the park for me the 1st time. He something to the effect of, "it's a direct drop in/plug and play" Why did I feel like I'd be about 3 hours into it when I decided that wasn't the case.
So at the end of your install, what all did you have to do/change/alter? Did you have to get back into the driver to change anything? Did it come with it's own driver and does that use the same EDS software?

The difference between the integrated servo and the stepper he was using is the driver. In the first version you had a driver in your cabinet that had 2 cables that ran to servo. In the new one the driver is on the servo but you still have 2 wires for power from your 48 volt power supply and then from the Clough42 daughter board you have basically the same wires as before. If you buy the same version that Clough used, he shows what software to download but you have to make sure you have a good Prolific chipset cable and driver so that you can communicate to the driver (on the motor), he set the stepper resolution to 500 vs 1000 he had first time, since he was double the gear ratio from 3:1 to 6:1, that way he didn't even have to reflash the TI board. I used a JMC 400 watt integrated Servo and it has double the torque of the one he tested. I used JMC software and also changed resolution to 500. I put this on a PM-1340GT and set my gearbox for a 6:1 reduction which is D-6 and my spindle to encoder is 1:1 and servo to gearbox is 1:1 using 40 teeth gears all around. It has plenty of torque and I've made several .1" DOC at .8"/rev at 1100 rpm, it throws some serious chips.

I made some code changes
Code changes
#define LEADSCREW_TPI 8
#define STEPPER_MICROSTEPS 6
#define STEPPER_RESOLUTION 500
#define STEPPER_MICROSTEPS_FEED STEPPER_MICROSTEPS * 4.2405

Now the bottom line is kind of a waste as I'm pretty sure that gets placed in an Integer register and is rounded off BUT it only effects the feed rate and it's still many time more accurate on actual feed rate than the labels on the front of the gearbox. They are totally incorrect on the PM-1236GT and PM-1340GT.

 

[paradox_pete]

Did you have to get back into the driver to change anything? Did it come with it's own driver and does that use the same EDS software?

I'm not sure I understand the question, but the "Clearpath killer" servo has an integrated driver, a separate stepper driver is not needed or usable with this setup.

I did need to download the configuration utility from the stepper online web site, and I'm pretty sure the only that specific configuration utility will work with the servo. Getting the servo configured was the biggest hassle for me, as I'm a linux guy and haven't run windows on a personal computer in ages (10+ years I'd guess), and the utility is windows only. I ended up using a Virtualbox windows machine to get the steps per rev set to 500.

Personally I wouldn't change a working setup to the servo unless you actually have problems with missed steps or something. It won't add any functionality.

 

[GunsOfNavarone]

Agreed...it's just the Tim Allen syndrome..."more power!" At some point I want to make acme threads....in at least, steel. If I'm gonna have issues, probably then. I'll have to watch the video again. I saw it the day he posted and it's much worse than a bit fuzzy. Now that you mention it, he did have trouble getting the servo working (even found issues with the manufacturer instructions or something PIA like that) I'm just not up for that right now. I'll let you guys work the bugs out!
@xr650rRider sounds like you have that buttoned up already. Subscribed hoping @paradox_pete gets this checked off his list.

 

[Tomzo]

Does it display the Clough42 and version number when you power it up on display? Or nothing? Those spring loaded connectors can be a pain to get a wire under them (better to use the crimp pins) but it won't fry the board.

It displayed the Clough42 and version and functioned properly without the encoder or motor attached - and while outside the enclosure. I did use crimp pins all the way around except for the 8 pin "aviation" connectors for the connections through the enclosure. I had checked everything pretty carefully to ensure the connections were right - but I guess not carefully enough.

I had to step away for the evening as I was pretty steamed so I will get on it tomorrow and see if I can figure it out.


Sent from my iPhone using Tapatalk

 

[jeffkash]

It displayed the Clough42 and version and functioned properly without the encoder or motor attached - and while outside the enclosure. I did use crimp pins all the way around except for the 8 pin "aviation" connectors for the connections through the enclosure. I had checked everything pretty carefully to ensure the connections were right - but I guess not carefully enough.

I had to step away for the evening as I was pretty steamed so I will get on it tomorrow and see if I can figure it out.


Sent from my iPhone using Tapatalk

I'm sitting here staring at a dead TI board. They seem to be fragile. I programmed one with the Clough software, no problem. The second board was dead. A third board works well.

TI does not stand behind the product. After a week of going round and round with them, they agreed that the board was dead. Then said tough sh*t.

I hope that the working ones stay working.

 

[Tomzo]

I'm sitting here staring at a dead TI board. They seem to be fragile. I programmed one with the Clough software, no problem. The second board was dead. A third board works well.

TI does not stand behind the product. After a week of going round and round with them, they agreed that the board was dead. Then said tough sh*t.

I hope that the working ones stay working.

I don't feel so bad now. I have been buying them direct from TI - they have a rather odd ordering system that asks a LOT of odd questions but they do ship pretty quickly. My problem is that what should have cost about $35 is now going to be over $100 - and a bunch of lost time. They seem focused on whether I will use it in a military application or not - I sure hope the DOD is not buying these things because they seem pretty unreliable to me. I love James' system but if this keeps up I may look into one of the Arduino based control systems - those things are bullet proof ( at least comparatively).

 

[xr650rRider]

I'm not convinced that they are that fragile. James seemed to have his hanging off the side of his machine for quite a while and I'm sure a lot of overs have as well. I had mine laying on bench and not exactly treating it with kid gloves. Yes it is easy to make a wiring mistake. If you put 5 volts to that TI board, then yes you'll fry it. If you forget to remove the jumpers before powering it with the daughter card, then yes, you'll fry it. By the time you mount in panel and keeping all the pins correct from male to female, there exists a strong possibility that you'll have a wiring problem. KISS approach, is do one thing at a time. Get boards properly soldered, go thru all the continuity tests that James took the time to put together. When you add his board to the TI board, it could be easily feasible to get a bent pin, connector only contacting half the pins. You have to make sure it is oriented the correct way. When that is good, with nothing connected, you should be able to power the TI board via James board using a 5V DC power source but you have to have the jumpers on the USB port removed. If you see your 5V and 3V lights and any lights on TI board, you should be able to flash the TI board. The next part assumes you've properly built the display and have a good cable connected to the daughter card. If your successful flashing TI board with the release software and click the green arrow to start the code, you'll see the "Clough42" and version number on the display. When that's all good, you can connect encoder, power up and you should see same as before and now you should see RPM indicated on display when you spin encoder. When all that works, then move on to connecting servo.

If that TI board had Honeywell or Allen-Bradley silk screened on it, it would cost $2000 with that capability.

 

[Tomzo]

You are probably right and I followed all of those steps - I had it working without the motor or encoder connected. I connected both when I installed it in my cabinet so something must have gone wrong there. I am going to get my multimeter out and sound out every connection again in the cabinet to see where I went wrong.

I am not complaining - just a bit frustrated. I realize that this is a hobby project of James' and this sort of thing requires an amount of diligence and skill on my part that I clearly did not apply correctly. As with many projects of this type, the journey to getting it working is almost the best part as I learn new skills along the way.