AX2550 melting Ch 1

11 years 7 months ago #29524920 by timraymund
AX2550 melting Ch 1 was created by timraymund
We seem to be regularly melting Ch1 in our AX2550\'s and are not sure what we\'re doing wrong.

We\'re driving a 36 VDC trolling motor using three series connected Group 31 AGM 100 AH 12 VDC batteries. Power control is connected to a fourth, separate battery.

The motor controller is set to RS-232, separate A & B, current limit of 5 (52.5 A), acceleration = 2 (medium/slow) and trim adjust = 7 (none). It appears that we are using version 1.7b or 1.7.

The standard operating procedure is to update the power command to ch1 every second. The command is set to 0 or maximum power (usually 80%), always positive. Thus, over the course of 30 seconds or so, the controller may be commanded to apply 0 for a few seconds, then 80 percent power for 20 seconds, then 0 again.

Typically the motor controller reports 40 - 45 A on ch1 at 80%. Temperatures are usually around 32 C.

After operating for five or six hours, we sometimes see that that motor controller continues to communicate, accepts the power command, but suddenly no current goes to the motor. In one case, we observed that the current suddenly went up to 100+ A. In no cases did we observe sudden increases in temperature.

In almost all cases, inspecting the non-responsive motor controllers shows that the MOS melted off the rail, out of the printed circuit board, and occasionally the printed circuit board is burnt up around Ch1 cables.

Is there a particular diagnostic we should consider? In the one case where the printed circuit board did not burn up, is there something we should inspect or review? Is there something in our operating procedure that is wrong?

Thanks!
tim.

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524921 by cosma
Replied by cosma on topic Re:AX2550 melting Ch 1
There is no known problem specific to Channel 1. The amps you pull are totally within the controller\'s abilities. 32 degrees is quite cool.

If Ch2 is still working, try connecting your motor to it and repeat the tests.

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524922 by timraymund
Replied by timraymund on topic Re:AX2550 melting Ch 1
We did that on two of the controllers and burnt up ch2 as well. Typically ch2 is running a smaller motor at 20% maximum.

We also have a Futaba RC RX with RCE220 relays in line with the RS-232 to enable a switchable RC mode. Is there something in the RC wiring that might cause such serious problems in the controller? Is there a specific diagnostic we can perform to validate the interface with the RC RX?

I have seen several articles in the forum that seem to suggest that suddenly shutting down a large motor sometimes causes problems. That is, it appears that going from 80% to 0% directly with a medium/soft acceleration may create very high regeneration currents. (The ch1 motor drives a 600 lb autonomous water surface vehicle, with maximum speed around 2 m/s. The motor may spin for a bit after the 0% command, either due to motor momentum and/or water flow past the propeller.) Do you think that regeneration currents might be a problem? Is there a diagnostic we can perform to see if regeneration currents are burning up the controllers?

I should also mention that tests in air (virtually no load) generally go very well. The motors show maximum consumption of about 2 A.

Thanks!
tim.

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524923 by cosma
Replied by cosma on topic Re:AX2550 melting Ch 1
It is not impossible that regeneration causes some problems here. You may want to try to spin the propeller while the channel is at 0 to get an idea on the kind of force it takes to move it and whether that kind of force is present if the boat is moving with the motor halted.

When stopped, the controller essentially shorts the motor. It will however sustain a lot of amps (at least 200A) for a few seconds before anything happens. There is no current flowing back to the battery.

There is current flowing back when the controller command is almost 0 (say 30% to 5%) and motors are still spinning fast. If for any reason, the current cannot make it back to the power source, then voltage will rise until it does. There is a protection against such overvoltage condition but these protection are not 100% bulletproof. If the controller is solidly connected to a battery, the voltage should rise only a little. Note that different type of batteries may react differently. If Lead-Acid, there is no problem.

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524924 by timraymund
Replied by timraymund on topic Re:AX2550 melting Ch 1
Thanks - we\'ll try spinning it to see if it\'s likely to spin in the water.

In reviewing the logs, I find one place where ch1 current goes to 190 A for 2 seconds. The ch1 temperature goes up about 10 F in about 3 seconds. The spike in current corresponds to a reduction in the ch1 command from 80 to 0% power. There is no corresponding dip in the 36 V bus voltage.

What do you think is happening here?

From this point on, the ch2 current shows no current, though it is clear from the GPS logs that the vehicle is successfully using the channel to turn the vehicle.

Also, from this point on, the ch1 current shows many short duration spikes to 70 A. Previous to the event, 80% power yielded a very consistent, flat 20 - 25 A current.

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524925 by cosma
Replied by cosma on topic Re:AX2550 melting Ch 1
This is a bit odd in that if it was regen current, Amps would be of negative value. The controller can measure negative current and in fact uses it to limit the regen current in case of abrupt stop. It cannot report negative current and would show it as 0. Regeneration also tends to increase the battery voltage, although as mentioned, if the batteries are solid, that rise may not be noticeable.

Also, current limiting is quite fast, so 190A would not show, or if it does it would only for a very short time. Not to mention that you have your current limit set at around 50A.

Still, assuming these numbers are correct, this would imply a very suddent and sharp overload, almost as if there was something blocking your motor. Unless it is an internal failure to the controller which eventually resulted in an internal short.

It is unfortunatelly nearly impossible to tell what happened looking at the burned transistors after the fact.

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524926 by timraymund
Replied by timraymund on topic Re:AX2550 melting Ch 1
As it turns out, this latest controller appears unburnt and is in fact still working.

Is there something we should inspect or test that would help identify what we did to it?

Are there repair schematics? Test points?

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524927 by cosma
Replied by cosma on topic Re:AX2550 melting Ch 1
Schematics are not published and would not be of great help YOur best resource at the moment is the log. Can you please email us the file at This email address is being protected from spambots. You need JavaScript enabled to view it.

Also, insert 2K, 1W resistors between each of the 4 motor leads and ground. This will keep the transistor outputs at a firm ground reference when the output stage is totally off, as it would be in undervoltage, overvoltage or over temperature conditions. These are rare events that are harmless to the controller but we found that pulling the outputs to ground, rather that letting them float and reach unpredictable potential because of the environment, results in improved reliability in some extreme cases.

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524928 by timraymund
Replied by timraymund on topic Re:AX2550 melting Ch 1
We\'re working to add the resistors and I emailed the logs this morning. Thanks!!!

Please Log in or Create an account to join the conversation.

11 years 7 months ago #29524929 by timraymund
Replied by timraymund on topic Re:AX2550 melting Ch 1
We just received two new AX2550\'s. As we are interested in \"before\" conditions, we opened them up and took pictures.

In the picture, one can see four blue resistors standing vertically in the near the middle of the printed circuit board. Are these the 2K 1W resistors discussed previously in this thread? That is, do we still need to add external resistors as described? Would doing so be counter productive?
Attachments:

Please Log in or Create an account to join the conversation.

Moderators: tonysantoni
Time to create page: 0.223 seconds