Air Compressor not turning off.

[dumbuya]

Originally Posted by ozrien

The changes between 1.60 and 1.62 are slight changes in the status LED. There is no compressor related change between the two.

If your having trouble with your PCM, go to the self-test page of the roboRIO web-page configuration.
Jump to "Self Test"...
http://wpilib.screenstepslive.com/s/...ribution-panel

This will tell you...
(1) -if the PCM is enabled (robot is enabled)
(2)- if the pressure-switch is closed (pressure is not full)
(3)- if the compressor-output is on

(3) requires (1) and (2) regardless of which firmware is used.

Does the compressor turn off when the robot is disabled? If not then it sounds like the compressor ground input is shorting to robot ground somewhere outside of the PCM.

When the compressor is on, is (1),(2),(3) reported in the self-test?

Another easy check is to just disconnect the pressure switch. The PCM has two weidmuller connectors that normally go to the pressure-switch. When the pressure-switch signals not-full, it shorts the two leads together (relay). When the pressure-switch signals full, the two leads are open. So if you short the two BLUE PRESSURE SW. connectors on the PCM with a wire, instead of connecting to pressure-switch, PCM will behave as though the pressure-switch reads not-full. Remove the wire and PCM will behave as though pressure-switch reads full.

How do we get the old version of the firmware (1.60)? We know for sure the pressure switch is good. We did series of tests to isolate it. When we did the self test there was no single instance when the compressor is indicated full.
Right now we will try a new PCM and see what happens. We will try this tomorrow. I will keep you updated.

[jSchnitz]

Originally Posted by dradel

Ours won't even turn on!!

Do you have a solenoid anywhere in your code? Our experience was that the compressor needed that before it would run.

[WDB]

Originally Posted by ozrien

The changes between 1.60 and 1.62 are slight changes in the status LED. There is no compressor related change between the two.

If your having trouble with your PCM, go to the self-test page of the roboRIO web-page configuration.
Jump to "Self Test"...
http://wpilib.screenstepslive.com/s/...ribution-panel

This will tell you...
(1) -if the PCM is enabled (robot is enabled)
(2)- if the pressure-switch is closed (pressure is not full)
(3)- if the compressor-output is on

(3) requires (1) and (2) regardless of which firmware is used.

Does the compressor turn off when the robot is disabled? If not then it sounds like the compressor ground input is shorting to robot ground somewhere outside of the PCM.

When the compressor is on, is (1),(2),(3) reported in the self-test?

Another easy check is to just disconnect the pressure switch. The PCM has two weidmuller connectors that normally go to the pressure-switch. When the pressure-switch signals not-full, it shorts the two leads together (relay). When the pressure-switch signals full, the two leads are open. So if you short the two BLUE PRESSURE SW. connectors on the PCM with a wire, instead of connecting to pressure-switch, PCM will behave as though the pressure-switch reads not-full. Remove the wire and PCM will behave as though pressure-switch reads full.

We put the firmware version 1.62 back in to reconfirm.
The compressor STILL is not turning off -- while minutes before, the other (v1.6) PCM was working perfectly.

Also, after removing the 2 pressure switch leads from the PCM, we see this:

The self test completed successfully.

PCM is enabled.
Compressor is close-looped on sensor
Comp Is On
Pressure is not full

Checked also for damage, shorts, or stray wire bits, but it's clean.
Other than getting another PCM to try, I don't know what to do next.
Ideas?

[ozrien]

So judging from the Self-Test the PCM reports that the "Pressure is not full" despite the pressure-switch inputs on the PCM not being wired to anything. That is likely the root cause for WDB's symptom, and not firmware. This sounds like hardware and your "other PCM" likely works regardless of what firm is in it.

To further confirm it's not a firm issue, I put up 1.60 PCM firm (one-dot-sixty) at...
http://www.crosstheroadelectronics.c...ation-1.60.crf
... Flash 1.60 back into your bad PCM and I bet you will still find that the self-test still reads "Pressure is not full" despite having the PCM not connected to the pressure-switch.

For any given functional PCM : When the pressure switch is not connected, Pressure should read as "Pressure is Full", regardless of all other factors. This is deliberate so an unplugged pressure switch causes compressor to stop.

dumbuya, can you do the same check? Unplug the pressure-switch from the PCM and run the Self-Test to see if the pressure still reads "Pressure is not full".

At any rate dumbuya and WDB, I recommend contacting support@crosstheroadelectronics.com for next steps.

[WDB]

Originally Posted by ozrien

So judging from the Self-Test the PCM reports that the "Pressure is not full" despite the pressure-switch inputs on the PCM not being wired to anything. That is likely the root cause for WDB's symptom, and not firmware. This sounds like hardware and your "other PCM" likely works regardless of what firm is in it.

To further confirm it's not a firm issue, I put up 1.60 PCM firm (one-dot-sixty) at...
http://www.crosstheroadelectronics.c...ation-1.60.crf
... Flash 1.60 back into your bad PCM and I bet you will still find that the self-test still reads "Pressure is not full" despite having the PCM not connected to the pressure-switch.

For any given functional PCM : When the pressure switch is not connected, Pressure should read as "Pressure is Full", regardless of all other factors. This is deliberate so an unplugged pressure switch causes compressor to stop.

dumbuya, can you do the same check? Unplug the pressure-switch from the PCM and run the Self-Test to see if the pressure still reads "Pressure is not full".

At any rate dumbuya and WDB, I recommend contacting support@crosstheroadelectronics.com for next steps.

Confirmed. Reflashed the suspect PCM back to 1.6 and the same behavior was observed. Removing the pressure switch from the PCM should not allow the compressor to run, but it still runs. So it looks like a hardware problem with this specific PCM. Put in our other PCM in and it works perfectly.

We took a picture of the board near the pressure switch contacts and there might be some anomaly there (you'll have to zoom in a little), but we'll get a replacement PCM anyway.

Thanks Omar for your assistance!

[dumbuya]

Originally Posted by ozrien

So judging from the Self-Test the PCM reports that the "Pressure is not full" despite the pressure-switch inputs on the PCM not being wired to anything. That is likely the root cause for WDB's symptom, and not firmware. This sounds like hardware and your "other PCM" likely works regardless of what firm is in it.

To further confirm it's not a firm issue, I put up 1.60 PCM firm (one-dot-sixty) at...
http://www.crosstheroadelectronics.c...ation-1.60.crf
... Flash 1.60 back into your bad PCM and I bet you will still find that the self-test still reads "Pressure is not full" despite having the PCM not connected to the pressure-switch.

For any given functional PCM : When the pressure switch is not connected, Pressure should read as "Pressure is Full", regardless of all other factors. This is deliberate so an unplugged pressure switch causes compressor to stop.

dumbuya, can you do the same check? Unplug the pressure-switch from the PCM and run the Self-Test to see if the pressure still reads "Pressure is not full".

At any rate dumbuya and WDB, I recommend contacting support@crosstheroadelectronics.com for next steps.

No matter what test setup we did it shows on the self test " Pressure is not full". We found a fix now by swapping a new PCM and it works as intended.
However, I cannot tell you what we find to be the issue other than to say we might have had bad PCM.

One thing I could try is update 1.6 firmware in the old ( "bad") one and see if it is going to respond as intended.

Thanks for your urgent response to the issue.

[dumbuya]

Thank you to all for sharing your experiences with the PCM.

We swapped the PCM with a new one and it worked as expected.