Victor 888 "y" cable issues

[theawesome1730]

For our drive train motors we are running 4 CIMs off of 4 Victor 888s using "y" cables to run two in tandem. We keep experiencing a strange issue of one Victor functioning normally when only it is plugged in, but as soon as the second is plugged in the solid light goes dark and makes a high pitched screech. When like that the screeching Victor outputs -12 volts even with no control input. We've swapped them with 884s and they function as expected. Are our Victors bad (we tried 5 out of the 24 888s we have, so it seems very unusual) or are we doing things wrong? We are considering sending them back to Vex and opting for Talons. If you have any ideas please be sure to share. Thank you all in advance

[AllenGregoryIV]

We have been running Y cables on 888s for the past 4 weeks with no issues. They are on our drive train controlling 4 of the 6 CIMs.

[Jimmy Nichols]

Same here. Running Y cables on the drive train with no issues or screeching as you describe.

[Brian Ha]

We had something of the same issue. We took out the y cable and it did stop the issue. But we also fixed some shorts. Although from what I have read on chief delphi y splitter cables on victors is a no no.

[Jimmy Nichols]

Although from what I have read on chief delphi y splitter cables on victors is a no no.

Please point out some examples.

[theawesome1730]

We had something of the same issue. We took out the y cable and it did stop the issue. But we also fixed some shorts. Although from what I have read on chief delphi y splitter cables on victors is a no no.

I tested for shorts as I made the y cable myself. none were found and our victors returned to normal when running by themselves, just together on the y cable seems to make them malfunction. We did finally get 2 888s to work on a y cable together. So I'm thinking that we have a bunch of faulty 888s

[AllenGregoryIV]

I tested for shorts as I made the y cable myself. none were found and our victors returned to normal when running by themselves, just together on the y cable seems to make them malfunction. We did finally get 2 888s to work on a y cable together. So I'm thinking that we have a bunch of faulty 888s

Have you changed sidecars? I remember reading somewhere that victors due draw more current from the PWM cables than the Jaguars do. If for some reason your sidecar isn't buffering the PWM signals properly it could not be giving enough current.

[theawesome1730]

Have you changed sidecars? I remember reading somewhere that victors due draw more current from the PWM cables than the Jaguars do. If for some reason your sidecar isn't buffering the PWM signals properly it could not be giving enough current.

It is a brand new sidecar from 2013. we will try a swap, but I don't think that's it.

[Ether]

I remember reading somewhere that victors due draw more current from the PWM cables than the Jaguars do.

I don't have a Jag here to test, but here's some input current vs input voltage test data I ran on Victor back in August last year.

[AllenGregoryIV]

Alright, so I maybe I spoke to soon. We finally wired the real robot this evening and guess what happened to our lift side Y cabled victors. High pitch noise, trying to move but not at full speed. We had to take it off the Y cable for our demonstration. Switched to the Y-cable we have on the practice robot and it still has the problem. Haven't had to much time to diagnose. One side is working fine with the Y cable to other is not.

[theawesome1730]

Alright, so I maybe I spoke to soon. We finally wired the real robot this evening and guess what happened to our lift side Y cabled victors. High pitch noise, trying to move but not at full speed. We had to take it off the Y cable for our demonstration. Switched to the Y-cable we have on the practice robot and it still has the problem. Haven't had to much time to diagnose. One side is working fine with the Y cable to other is not.

Sounds exactly like what we had happen. Did you test the Victors by themselves later on? We set aside all the ones we thought we might have damaged so we can test them after stop build to see if they were faulty or if we were doing something wrong.

[Marc P.]

We noticed the same thing with 888s. My suspicion is the 888's may have significantly lower "deadband" than the 884s, making them much more susceptible to slight variations in PWM input. If the calibration is slightly different between the victors, one might be seeing "neutral" while the other sees "+1% power". It didn't seem to affect our performance much- full forward on the joysticks gave us the green LED, full reverse gave us the red LED, just the neutral position seems to be slightly off. Gently bumping the joysticks one way or the other would reverse which 888 didn't have the orange neutral LED, and on occasion when the joysticks came to rest, both would be orange.

We haven't tried calibrating them yet, but based on what we observed of the symptoms, it may solve the problem.

[Ether]

My suspicion is the 888's may have significantly lower "deadband" than the 884s

There's no "may" about it. They do.

[theawesome1730]

There's no "may" about it. They do.




Does that essentially mean that we have nothing "wrong" with victors on a y cable, but since the operating deadbands vary from Victor to Victor we had two in tandem that were not compatible? The "screaming" we experienced would be explained by the +or - 1% like Marc noted above with one victor making since I know that 888s normally make a high pitched sound under normal operation. I do not however understand how a victor running 1% output can put out 12 volts.

[Al Skierkiewicz]

Has anybody tried contacting IFI to check on this?

[RufflesRidge]

I do not however understand how a victor running 1% output can put out 12 volts.

Was that measured unloaded?

I agree that the cause of this is almost certainly the added load of a second Victor affecting the rise and fall times of the PWM signal enough to land just outside the 888s deadband. Try calibrating the Victors while they are attached to the Y cable and I expect the problem will go away.

[Ether]

Was that measured unloaded?

Here's a graph (http://www.angelfire.com/nd/edu/FRC/Vic884_1500_ohm_resistive_load.png) of an 884's output voltage vs input pulse width, with the output essentially unloaded (1500 ohms).

[Racer26]

I found when working with 4343 that we had to calibrate our Victor 888's on the Y cables to get them to behave nicely. Took a few tries to get "off" to be "off" and not "on just a hair"