Slaves disabled on ECG8 - anyone else?

[dmoore]

Ok...In my never ending quest to squash every last bug in my display I was trying to figure out why 504 of my vertical pixels in my display on universe 5 stopped working. At first I though it was something with the TP3244, so I pulled it and put in a spare unit - still nothing. So I pulled both the spare and the "production" TP3244 and put them into my test rig which is a little 2ft strip of 6803. I then used the TP3244 test output to confirm that both of them were working properly. So then I though... wiring or maybe power supply. So, I pulled out the power supply and replaced it with the one used in the bench testing - no go still. Then I tried using the same setup but with the test strip (outside on the display) - no go still. Then I though... must be LSP. So I go inside, stop the show and pull up the TP3244 test program and attempt to test output on universe 5 (slave 5) with my bench test rig and it didn't work. Then I'm thinking, the TP3244 program must really only output to universe 1 even though it says it outputs to universe 1-16. Then after some more testing I discover something I had seen in the past - where my slaves in the ECG show as --INACTIVE--. Hmm...

So, I think the problem must be with the ECG, so I shot this video to show what I was seeing so everyone can see that I'm not completely crazy! Wondering if anyone else has seen these --INACTIVE-- or some other disabling of slave outputs. The ECG8 is running 1.3.1 if that helps.

Video:

ECG Configuration Issues

 

[Pyrotech]

I'm not at all sure if this is helpful to you, but my inactive slave on DR4 was traced to a cold solder joint.

Ed gave me these instructions:

The lack of an id match is frequently caused by that VCAP being bad/wrong. Do you have a voltmeter (or does santa have that too)? If you compare the reading from the left side (near the IC) of that CA3 and compare it to the values on CB3, CC3, CD3 (the other slaves caps) I think you will find our problem. You can hook-up the negative side of your VOM to one of the test points. TP1 up near the power input connector is the common ground test point.

There did prove to be a significant difference on the inactive slave.

Obviously parts numbered differently on your set up, but is it a possibility?

Ian

 

[dmoore]

There did prove to be a significant difference on the inactive slave.

Obviously parts numbered differently on your set up, but is it a possibility?

I'm not sure but I can sure try this. What's weird is that the show from the prior night was fine but when the show started the next night (without any changes or adjustments in to the PC/ECG) it wasn't working. I've also had it at times stop working and then start working on just slave 5. Let me take a look.

Thanks,
david

 

[mrpackethead]

Almost certainly hardware on the ECG-8...

 

[dmoore]

I took the ECG out of the mounting box and flipped it over looking for anything wrong and found that pin 28 of output 5 was no soldered at all. So I soldered that up and then re-touched all the solder joints on slave 5,6,7. When I powered it up again, it was working fine.

Of course my question is - why was slave 5, that was working just fine without the soldered pin, seemingly cause problems with slave 6? Are slaves 5 and 6 linked in some way? Why would it work fine for awhile and now all the sudden become an issue?

So far this does appear to have resolved my issue. Sorry for the confusion - so much new DIY stuff and software!

 

[Pyrotech]

Thats good news.