Strange Color Problems with Cadet Modules

[simon]

Hello I’ve currently built one of each Cadet except V, VI and X and have been experiencing a strange problem with color output.
First of all the colors don’t seem to correspond to the inputs labeled on the RBG Encoder (I’m using this demonstration as a reference). The VCO output, when plugged into Red on the RGB Encoder makes green, into Green a desaturated red, and into Blue a yellowish-green.
Also the colors seem to gradually become increasingly desaturated over time after I power up the modules until around 10-20 minutes after power up the output of the encoder suddenly becomes completely black and white. After turning the eurorack power off and on again the output is back to the original color and this process repeats itself.

These problems seem similar to the ones described in this post:

Getting black and white output from Cadet II: RGB Encoder

I reflowed some points on the RGB Encoder, and checked everything around the rbg to composite encoder and the crystals. The parts placed seem to be correct and i got something like this: [02] The image quality is kinda crap sorry about that, i was in a rush. Those are not white bars they are rather reddish/pink. But im still getting only this output, it does not matter in which color i plug the vco. Next up im going to check everything again according to the BOM and Schematic and reflowing e…

I’m using the correct region setting on the RGB Encoder and Sync modules, using a Doepher A-100 power supply, an lzx sync distribution cable, and have the output of the Encoder going directly to a CRT.

I’m wondering if this is salvageable, if I should get another PCB and parts and try again (afraid I’m starting to make things worse the more I try to resolder stuff as I’ve been trying to fix this for a while) or if there’s someone I could buy a properly-functioning pre-built RGB Encoder from or pay to build one.

Attached are photos of the RGB Encoder as I imagine that’s probably the module that’s malfunctioning. I’m happy to provide more detailed photos of any part of the board or photos of another module if people think it could be a problem other than the Encoder. Thanks!

[simon]

Also according to someone in that thread I linked the top RCA jack of the RGB Encoder is a black and white output and the bottom is a color one? Both jacks appear to have the same washed-out color output on mine

[csboling]

@simon

someone in that thread I linked the top RCA jack of the RGB Encoder is a black and white output and the bottom is a color one

That might have been me, and this is not correct - both outputs are color and should be basically identical.

Sorry to not be more helpful than that. Nothing jumps out at me from your photos. The behavior you describe is odd, sounds maybe temperature dependent? Not sure exactly what that might suggest looking at.

[transistorcat]

Just a hunch here - Have you tried running the module (and the sync generator) in NTSC mode rather than PAL (Or the other way round if you’re in ntsc country)?
Do get the same error, a different error, or no error?

[reverselandfill]

Yes , check the PAL / NTSC switch position (of both the Encoder and Sync gen)
sometimes a TV monitor also has this switch. make sure it is in the correct position.
Also check your cables. try a different set as a test

All the parts locations look ok, I referenced it against mine
Some solderjoints look like they could use some reflowing. pic 4 near the 7805 and some of the lm6172 pads.

[simon]

Thanks for the replies everyone!

PAL / NTSC switch positions were correct on both the Sync and Encoder module.

I tried flipping them to the incorrect position (PAL) just to make sure it wasn’t a problem with the switch and when I powered up the eurorack it the image was black and white (immediately, instead of starting in color, fading and then switching to black and white)

So it doesn’t appear to be a problem with the switch either(?)

[transistorcat]

The fact that there’s a difference in the failure mode between ntsc/pal is interesting, but sadly not at this point very enlightening.

The fact that B&W video still works means that the error more or less must be in the yellow area here:

I’d do a sanity-check with a voltmeter at the middle terminal of the switch, but i wouldn’t have very high hopes of that being the issue.

Looking at the pictures,
That translates to this bit

and this IC:

As well as the smd IC and the switch.

It’s hard to tell from the pictures, but there might be some cold joints in these areas:

The top left one is the 5V regulator, this is probably not your problem, but a little extra solder on that pin can’t hurt.

Top right - This is the inverter that drives the crystals - I don’t think these are to blame, but some of those joints look like they may not have proper solder wetting (Make sure the soldering iron connects properly with both the pad and pin, hold it steady for around 3-4 seconds until the solder sticks to both the pad and pin all the way around the pin)

Bottom left: This is one of the crystals - there is something funky going on with the soldering here. If you’re unlucky, you’ve managed to short it to the plane beside it. Either way both pins look like they could use some careful cleaning, inspection, and re-soldering.

[Temp_hill]

Oh I see i forgot to post an explanation to my post.

The problem wasn’t with my RGB Encoder, but with a blank ATMEGA chip on my Sync Generator.
See this topic here.

I don’t know if this is any help but i got a programmed Chip from LZX and my problems were solved after swapping them out. I had tested my RGB Encoder with a Visual Cortex to get this insight.

[Robbertunist]

@simon, did you find a solution to the problem?
I wonder if it was a problem with a blank ATMEGA88 as @Temp_hill suggested?