AX1500: 37.0 volts = Undervolt error?

I\'m having issues with my AX1500.

I\'m trying to get it running closed loop position mode with RC input and, in fact, I did have it running that way for a very brief period earlier this week.

But today when I went back to working on the project, it powered up with the status light on solid.

Trying to talk to it with Roborun was an exercise in frustration. Most of the time I got a lot of invalid data errors and the controller information would not fill in, despite the \"Controller Communication OK\" message.

I was able to talk to it on the console tab. So after a little manual reading, I sent it the ^89 command to find out what the fault is:

^89
04
+

So that seems like bit 2 is set, thus an undervoltage error. A meter on the battery terminals reads 37.2 volts. I tied the previously floating Pwr Ctl input to B+, but not see any improvement.

More manual reading leads me to the ?e command:

?e
AC
4E

By my math, this means:
Battery Voltage: 55 * 172 / 256 = 37.0v
Internal Voltage: 28.5 * 78 / 256 = 8.7v

8.7v seems low for a 12v rail. Is this fixable?

Still no clue what is going on here. This morning I powered it up, and still had the solid status light. I stuck my meter on the input pin of the L7805, and it was fluctuating between 8.7v and 9.1v. This is on a 36v lead acid battery pack, reading 37v.

So I pulled the controller out and moved it back to the bench. Put it on my power supply at 24v, and the status is doing the single-flash RC mode indicator. The voltage on the input of the L7805 is 12.0v. Then I slowly turned up my supply to 37v. At this point, the input pin of the L7805 was reading 10.1v, but status indicator is still flashing the single flash, RC Mode indicator.

For this kind of troubleshooting, it is always best to connect the controller to a pc running the roborun pc utility.

From the run time, monitor the internal voltage (which is the 12V at the input of the 7805). It should be around 11.8V when the power supply is 13V or higher. If it is less than 11V, this means the controller may be damaged.

If it is less than 10.5V, the controller will be in an undervoltage condition.

I\'m not so sure the data comping out of the serial port is all that reliable.

I wasn\'t able to reproduce the under volt error condition today, but I tried to talk to it with two different computers with two different USB->Serial adapters, and none of the results were consistent or reproducible.

One went like this:
Start RoboRun.
Controller Found
Reading...
Popup that says:
Run-time error \'6\'
Overflow
Roborun shuts down.

Another time my AX1500 was identified as follows:
Controler ID: AX500-SC
Software Rev. v1.9g ?31/02/0
Hardware Code: 3-1-0 2 FET per Leg Single Channel

Another time, the console screen just showed gibberish. (Screen shoot attached)

Yet another time it recognized the controller, and I was able to correctly identify the controller, and give me battery and internal voltage readings. Battery voltage match what the power supply was feeding it (37v) and Internal voltage read at 11.7v. I got my hopes up, but then it started telling me nonsense like Battery Voltage 0v and Internal Voltage 22v.

This looks like a case of bad serial connection. We\'ve seen this type of problem when the ground connection on the 15-pin connector is not properly made and then ground somehow makes it via the Ground terminal and power supply, picking up a lot of noise alone the way.

Check also the voltage on the -12V test point

I\'m not finding the -12V test point. Where about should I be looking? I see the +5v, +12v and a bunch of others.

I\'ve been switching back and forth between two serial cable, to try to eliminate that factor. On of the is the one supplied by RoboteQ, and the other is one I made by pulling the RC1 input off the RC cable, and then connecting a DB9 to it.

I just continuity tested them both. The factory cable is, not surprisingly, wired correctly. However, I did find that on my home brew one, I have DB-9 Pin 5 to DB-15 pin 6! So that could be something. I will have to fix before doing more trouble shooting.

It does seem that data is getting to the controller. In Hyperterm, I send 10CRs, and the LED blink changes to the 3 blink pattern. Sending %rrrrrr causes it to go back to the one blink pattern. Results in Roborun are similar, but it seems to quickly switch back to the 3-blink serial configuration quickly after sending the %rrrrrr. This was noted before I discover the cable wiring discrepancy, and I don\'t know which cable was in use at the time.

Fixed my home brew cable. No difference. Neither one seems to get much back from the controller. Entered commands aren\'t echoed back, but based on the changes in LED blink pattern, I think they are getting received.

Measured voltages at test points:
-12v => 1.2v
+ 5v => 5.0v
+12v => 11.9v

The 1.2V on the -12V test point is definitely wrong.

It looks like one IC got damaged. Please contact us at This email address is being protected from spambots. You need JavaScript enabled to view it. for instructions on how to repair yourself or on how to return for repair.