Hello,
I am currently on my second LAUNCHXL-F280049C (P/N 296-53092-ND) and possibly going on my third. The project I am using it for is the Clough42 electronic leadscrew for my lathe. With the first one, I loaded all of the software, and it was functioning properly on a dry run (nothing hooked up to the lathe) and the following morning I went to plug it in and got nothing. Plugged it into the computer, got a red LED on the board, but computer couldn’t communicate with it. After getting another, I uploaded the software again, and it was functioning properly. I went to plug it in today and now the second one is not working. The only thing I can think of is the High Frequency from a TIG welder (about 15 feet away,) but that was only used around when I had the second board. The encoder I am using also appears to have failed if that is any help. I am trying to see if I need to abandon this project if the source of the failure is HF interference, or if there is a work around. Thanks.
Hello Matt,
The Clough42 electronic leadscrew is an impressive project. The DIY aspect is attractive. The best part is the documentation and large user base for the product.
Here is a copy of the wiring diagram so that we can have a common conversation.
You may be onto something with regards to damage from electrostatic or electromagnetic induction. There are multiple sources of entry including static electricity from the user and the high-frequency plus high-power welder you mentioned.
One thing that immediately catches my attention is shielding. Where is the shield connection for the encoder? Likewise, is there a shield wire from the display cable and motor drive? The documentation has a few suggestions:
If you use shielded cable, connect the shield to a chassis earth ground at one end only. Don’t connect this to the logic ground or the motor power ground. Keep it separate and connect it at only one end.
The encoder usually has five wires, plus a shield. Check the markings on the LaunchPad board to determine the correct pins for 5V power, ground, A, B and I (index) wires. Connect the shield to a chassis ground, if practical.
Know that the I/O pin on the Launchpad’s TMS320F280049C are delicate in the same way as any other microcontroller. One way around this problem is to use optical isolation. An example is the drive interface on your motor driver. This optical isolation provides an electrical barrier that allow the signal to pass but protects the drive’s sensitive electronics from kilovolt voltage spikes. No such protection exists for the microcontroller.
Recommendations:
-
Provide a few pictures so that we can see if there are any obvious issues.
-
Seek out other people or even clough42 to see if other have had a similar experience.
Best Wishes,
APDahlen
Hello APDahlen,
Thank you for the response, I will provide some pictures tomorrow. Shielded cable was used from the servo driver (driver is built onto the servo motor itself) to the PCB, from the display to the PCB, and the encoder as mentioned in the documentation comes with shielded cable. All are grounded only on the enclosure side. Is there any way to test if it is in fact a failed I/O pin on the TMS320F280049C?
Thanks again,
Matt
Hello Matt,
We will be looking for your pictures.
With regards to testing. Did Clough leave any visual indicators on the Launchpad, interface, or the display boards. Specifically, are there any LEDs the respond to an encoder output line? Also, I wonder if there are any serial outputs used for debugging.
My apologies as I do not have the time to read and interpret the code.
At any rate, yes, it is possible to test the microcontroller and development board. However, if the two previous conditions are not met, you will need to program the micro. Normally I would recommend Energia as it provides a simple to use Arduino like programming experience for TI’s Launchpad product family. Unfortunately, the LAUNCHXL-F280049C does not appear to be supported. The solution therefor involves writing a program and the more difficult task of configuring the various I/O pins to act as inputs.
Perhaps others can provide better solutions.
Best Wishes,
APDahlen
APDahlen,
As far as visual indicators, when the launchpad is powered correctly through the PCB (getting 3v3) there is a blue and red led that illuminate. This is the same if you have the jumpers in the position to power it off of the USB, but that does not function after the failure either.
Another thing I have noticed is that when the PCB is still attached to the launchpad after the failure, the 5v led on the PCB barely flickers. When I pull the PCB off, the 5v and 3v3 LED’s illuminate as they normally should, and the display has power. This is shown in the pictures. I have tested the 3v3 regulator on the PCB, and it is still supplying the correct power to the launchpad.
Also, I don’t think I got the other Shielded cable grounds in the picture, but they are in fact grounded at the top of the enclosure.
Thanks,
Matt
Hello Matt,
That sounds like a rail to ground short in the Launchpad itself. If I understand correctly:
-
the normally lit Launchpad LEDs are abnormal when powered via the ELS board or when powered directly from the launchpad’s USB port.
-
the ELS board’s power LED is dim when attached to a defective Launchpad but increase to normal levels when the ELS board is removed from the Launchpad’s headers.
Based on what I see in the pictures, there could be several problems including:
-
Physical twisting of the Launchpad PCB.
-
Launchpad wire(s) short to ground via the hex head screw or the aluminum standoff.
-
Floating ground between the white mounting plate and the panel itself. I can see the black ground wire from the AC safety ground (green) to the panel – slightly obscured by the split wire loom. However, there is no ground to the white mounting plate. The two screws holding the white plate in place will not provide a solid ground. The motor drive may or may not provide a ground to the white panel.
-
Floating ground on the Launchpad. (Yes, this contradicts the previous point about shorting to the standoff). Yet, the Launchpad and the DC power supply do not appear to be deliberately grounded in this system. It may be useful to run a wire from a Launchpad ground to the main chassis ground. You could test this by measuring the resistance from the Launchpad ground to the chassis.
-
There may be other unknown and likely intermittent issues such as a damaged encoder, the ELS board, the display, or wire. These intermittent problems can be very difficult to troubleshoot.
Caveat: I am not qualified or certified in UL safety standards for industrial control panels. However, the above-mentioned grounds appear to be a safety problem along with the floating wall wart power supply. These resources may help to get you started:
-
1770-4.1, Industrial Automation Wiring and Grounding Guidelines (rockwellautomation.com)
-
Fundamentals of Grounding in Industrial Automation and Control Systems - Technical Articles
Sorry I couldn’t point to anything specific. Again, I recommend reaching out to James Clough and other member of the community. Perhaps they have additional information about this equipment. Also, I do hope that other will offer suggestions on this thread.
Best Wishes,
APDahlen
APDahlen,
I have continued to attempt to address these issues and ended up bringing the whole setup to an electrical engineer that I work with. He initially suggested what you have mentioned in your last post, and after a quick look, found a short between 3v3 and ground. He said he will go more in depth to see what the issue may be that caused it, but as of Friday he was fairly convinced it was delamination, and suggested I may have just received two boards from the same batch. Thanks again for the help, I will be sure to post the findings.
Thanks,
Matt
Thank You, Matt.
I truly appreciate the update and hope you get the machine working soon.
Sincerely,
Aaron