Loss of Serial Comms without Power Supply

[whip_tracker_cracker]

I have an interesting issue that I thought I'd see if y'all have ever ran into. I've got a serious of cm-5/bioloid kits being used by various students in a lab. All of them have created robot arms in which they run feedback controls through using an external computer. The cm-5 contains a bridge program that takes the serial comms into the cm-5 and translates it to ax-12 language. Anyways, the programs have been working perfect with the wall outlet PSU or on-board charged battery. Recently, I can now only get them to work with the wall outlet PSU. Let me dismay a few things I have checked. Yes the fuses are intact! That was the first thing I checked. The batteries are fully charged and mantain the proper voltage under load. I can operate the robot from my PC and give each motor and angle with the PSU plugged in. If I also have the battery plugged in, I can remove the PSU and the battery has no issus holding the motors still. But my Comms goes away. I've tried this also with the standard robot terminal. I can communicate with the CM-5 only with the PSU and not with just the battery. It seems to be a really strange behavior and it is not just due to one cm-5 but something they all share. I am using a bridge program that is modified off of the program that pedro created as a bridge program. The serial comms are the same, I've just modified extra cases for handing angles, speed, continuous rotation,etc. Any help would be greatly appreciated. I'm planning to run the cm-5 and arm off of mobile robots that control the arm through their serial.

[i-Bot]

Did you try swapping the cable and any USB serial/ PC interface ? I wonder if you are missing the ground connection which is only made with the PSU.

[whip_tracker_cracker]

Did you try swapping the cable and any USB serial/ PC interface ? I wonder if you are missing the ground connection which is only made with the PSU.

I'm using the serial cable directly into the PC. I've done multiple PC's with Multiple cables. The same effect is prevalent. This makes me wonder if in fact there was an issue with or something up with the serial programming. Again, my serial communication just drops away without the wall outlet PSU. Battery power still works, but I can no longer send the needed commands to change motor angles,etc.

[i-Bot]

Do the LED still light on the CM5 OK ?

When you disconnect the wall outlet PSU, do you remove plug from CM5 ?

The schematic for the PSU is here. You may want to probe about with a meter. Also check what you program does witht the pins associated with PSU.

http://robosavvy.com/Builders/petej/AX- ... rBoard.pdf

[StuartL]

My guess is the power supply for the RS232 driver chip has gone. Perhaps a burned out track?

[whip_tracker_cracker]

For all of those who are curious, or may run across this issue, I figured it out. I purchased all 5 basic bioloid kits from Trossen Robotics and would have no problem with them for a while until they suddenly have communication issues. After going over everything suggested, of which I do appreciate your input, I finally decided to swap out the serial cable the runs to the cm-5. The cables look fairly thin so I figured as a last ditch effort I would try that. Sure enough, with a better cable, instant success. It seems that although with power supplies from a wall outlet on both sides they share a common ground so the serial comms works, even if the ground through the comm cable is shot. If I go onto battery supply with either the computer or the cm-5, the bioloid communication stops working. However, I noticed that if the computer and cm-5 both shared a common battery supply then the communications would work. thus it is simply a grounding issue. A better cable seems to have fixed all of my problems. Thanks again for your input.