PCM Brownout

[BOSS]

We are having a slight issue with our PCM see below. Has anyone else had this issue see below.

We have noticed during this build season at times that the PCM would alternate between green and then flash red and yellow for a second or two and then return to green. Initially we didn’t really pay much attention to this as it didn’t seem to be causing issues during initial testing. However once we completed the robot and were relying on solenoids and the compressor heavily it has become a serious issue.

The issue in the PCM is occurring only when our motors are running at full speed in either direction and then the joystick is released quickly causing the motors to stop suddenly. As soon as that occurs the PCM lights will alternate red/orange and then return to green. While this is happening the compressor will not activate and essentially all functionality on the PCM is nonfunctional until the unit apparently resets and returns to green. If we take it easy on throttle release this issue will not happen.

We have done the following.

1. Checked and rechecked all wiring on the entire robot

2. Relocated the robot wireless radio over 10’ from the PCM

3. Replaced the Canbus wiring

4. Temporarily moved the PCM power input to a standard 20 amp output on the PDP

5. Increased the 18g wire feeding power to the PCM to 16G

6. Replaced the module with a spare from last year’s robot

7. Ensured the battery in the robot is fully charged and is also a new battery.

8. Ensured that firmware is up to date on all components.


All of these attempts do not change the issue.

As a last test we moved the PCM power input off the robot and went directly to another robot battery outside of the robot not connected to the PDP. This step obviously cannot be maintained for competition but absolutely made the PCM function 100% correctly.

This issue seems to be some type of voltage change issue to us (brownout, spike, etc)

We cannot think of any other things to try to resolve this issue.

[nickbrickmaster]

Have you used the self test function to get any additional information?

Is the 20A blade fuse secure?

How many cims are on your drive?

[gpetilli]

It may be that your battery voltage is dropping low enough that the PCM can not generate the correct voltage to drive the solenoids. It is probably real. This year in particular with many teams not correctly adjusting gear ratios for 8" wheels, there have been lots of brown outs.

Keep in mind that the short circuit current of the CIMs is 133A per CIM (practically limited to about 90A). If you have 4 CIMs, thats roughly 400A. The battery internal resistance is about 0.022ohms so it drops 8.8V leaving only 12-8.8V or 3.2V at the input to the PDB. This is an extreme example, but you get the picture. I believe the driver station software allows you to plot the battery voltage even during a match.

One effective solution for the exact problem you mention where you brown out when changing direction quickly is to add a voltage ramp to drive code. If you are using the SRX (or older Jaguars), this can be configured directly - otherwise you need to add a ramp to the commands to the speed controllers. I think we targeted 48V/sec or 0-12V in 1/4 sec.

[kan123]

I can answer the questions also since I am also on this team.

We have check the 20 amp fuse.

We are 4 cim drive. For testing this issue we have been on blocks and have also removed breakers for 3 of the cims, so we are powering one cim while on blocks. The issue remains.

We will certainly look at the voltage ramp as a possible solution.

What is puzzling is this never happens when accelerating rapidly, it only occurs when decelerating rapidly.

We appreciate the assistance!!

[Daniel_LaFleur]

Originally Posted by kan123

I can answer the questions also since I am also on this team.

We have check the 20 amp fuse.

We are 4 cim drive. For testing this issue we have been on blocks and have also removed breakers for 3 of the cims, so we are powering one cim while on blocks. The issue remains.

We will certainly look at the voltage ramp as a possible solution.

What is puzzling is this never happens when accelerating rapidly, it only occurs when decelerating rapidly.

We appreciate the assistance!!

What is the voltage being reported back to the drivers station.
The dashboard should give you a graph of that info.(I know you said you checked the battery, but this info will help diagnose the issue)
A good fully charged battery should report >12.8 VDC. If the battery is reporting lower than 12.2 VDC then the battery may be not fully charged or may not be good. Even new batteries may not be good.

What type of drivetrain are you using?
Going from full forward to stop may cause the CIMs to stall, depending on the mass of the drivetrain. This will be made worse by reducing the number of CIMs to 1. A single stalled CIM should not cause an issue, but knowing this again will help diagnose the issue.

When the compressor first kicks on it may draw more than 20 amps (inrush current). This should not trip the snap action breaker since that current draw is for such a short period of time but you might want to check the current of the compressor.

[gpetilli]

Originally Posted by kan123

I can answer the questions also since I am also on this team.

We have check the 20 amp fuse.

We are 4 cim drive. For testing this issue we have been on blocks and have also removed breakers for 3 of the cims, so we are powering one cim while on blocks. The issue remains.

We will certainly look at the voltage ramp as a possible solution.

What is puzzling is this never happens when accelerating rapidly, it only occurs when decelerating rapidly.

We appreciate the assistance!!

It is hard to induce brownout while on blocks - especially with only one CIM fuse active. Decelerating rapidly makes sense since the drive needs to overcome the inertia of the robot - but not while on blocks.

If you are sure about brownout on blocks, it sounds like a mechanical binding. I would start by putting all fuses back and using the drive station software to plot out the current to each CIM independently. Is one current different than the others? Currents should be say 6A per CIM full speed on blocks and spike to maybe 30A when changing direction (more if driving). I am guessing at these numbers so if someone can try this on a working bot (ours is in the bag) and report back - that would be good.

Are your axles square with the frame? Belts, chains or gears? KOP chassis or custom? While on blocks, can you turn an unpowered wheel or is it bound up?

We did have a problem with one of four gearboxes once where one CIM had evidently been dropped on its output shaft and it had a slight bend which caused the gears to over engage. Disassembled that gearbox probably 10 times till we found the root cause. It does sound like some sort of assembly issue - maybe an extra spacer someplace binding things up?

[kan123]

We are looking through the logs to determine voltages at various times. More testing on that bot will have to wait obviously. We have checked this on last years bot this morning and the issue is there also. We did not rely on pneumatics as heavily last year. We didn't notice this last year. Drive train is lighter on last years bot for sure

The drive train is 4 cims, 2 driving a gear box on each side with 3 - 8" wheels on each side

This issue occurs with or without the compressor connected. The PCM can be at a state of powering the compressor or not, the issue is the same.