User avatar
ubermick
California, via Cork

Posted Thu Mar 11, 2021 10:02 am

Heya folks - so picked up an A1200 from the UK which has a bit of an issue. Seller mentioned that the display had a bit of a brown tint but was present only in some modes. Asked him about it, and he said that his research led him to believe it was the VIDEO-DAC chip VP101-3BA, a $5 part but required SMD soldering skills which he didn't have. That said, he said it was barely noticeable and only present in some modes in workbench so the repair wasn't even necessary unless I desperately wanted to use those. Fair enough, went ahead with it.

Arrived yesterday and fired it up, and I kinda feel like the "bit of a brown tint" was undersold a wee bit. Here's kickstart screen of the A1200 on the left, compared to that of the normal purple in my A500+ on the right.

Image

So despite him saying that it wasn't really noticeable, erm... I kinda notice it. So was wondering if any of the resident tech gurus can offer insight into this - is it REALLY a $5 chip replacement, or something more? The board was recapped a few years ago, so don't THINK it's that.

And more to the point, any SMD-handy Amigans in the USA (I'm in Northern California) able to help out with a repair?

User avatar
obitus1990
New Orleans, LA, USA

Posted Thu Mar 11, 2021 4:13 pm

Same results on both the composite out and RGB?

Could be the DAC, could be a bad resistor, could be a bad trace due to improper cleanup after a recap (and continuing corrosion after the recap), worst case could even be a LISA issue (not likely).

Acill (Paul) is in California, Port Hueneme (don't know if I spelled that correctly), and you could check with him. If he's too busy, I can take a look at it.

User avatar
ubermick
California, via Cork

Posted Thu Mar 11, 2021 6:48 pm

obitus1990 wrote:
Thu Mar 11, 2021 4:13 pm
Same results on both the composite out and RGB?

Could be the DAC, could be a bad resistor, could be a bad trace due to improper cleanup after a recap (and continuing corrosion after the recap), worst case could even be a LISA issue (not likely).

Acill (Paul) is in California, Port Hueneme (don't know if I spelled that correctly), and you could check with him. If he's too busy, I can take a look at it.
Thanks, Kirk - I've reached out to Paul to see if he's up for taking a look at it (his website says he's only doing recaps at the moment, hopefully he can take a peek though, if not I'll be annoying you again!)

I haven't tested it with composite - I genuinely have nothing around the house that accepts it! (Oh wait, I might have an old AV unit I can pass it through...)

User avatar
ubermick
California, via Cork

Posted Fri Mar 12, 2021 1:07 pm

So diving down the oul' internet rabbit hole, as I tend to do...

The previous owner had actually posted up elsewhere about this issue a few weeks prior to him selling it. (Again, he did disclose this issue, so not claiming he sold me a faulty machine. Well, he did - but did so openly, and I bought it knowing it!) Per him:
In PAL hires, lores, interlaced, super hires etc. The grey has a brown tint to it (pic on left). Not the same as my A600 or A500. However, if I choose productivity or dblpal then I get the nice pure grey (pic on right) I’m expecting. Photos were taken from an lcd via an Amiga video out to ossc but the results are same using a scart lead straight into a commodore 1081 monitor. Any idea what might be causing this?

Also the palette colour for the grey background is set to 170/170/170 in all instances so it's not that. Also my A500 and A600 were tested using the same cables and are fine.

EDIT: Same off colour tint on the composite output as well (can only try the PAL modes I guess on this; DBLPAL the monitor didn't like)
So does indeed replicate in composite. I pulled out the multimeter and checked the RGB resistors (R232A/B/C and R231A/B/C) which showed all six were the correct values (25ohm for R232 and 50 for R231) so don't THINK they're the issue, although obitus' point about a bad trace could also be an issue. (Board looks GREAT tho, recapped by Retropassion on 10/20 so just six months ago... which makes me think its a recent happening, since surely he'd have seen that in testing after the recap). Looking at this video here,

https://www.youtube.com/watch?v=8xlG9Ln5o30

Does make me worry about the Video DAC. (And others have said the same thing.) So from my completely idiotic standpoint, the original diagnosis LOOKS to be close to the mark, and that its a $5 chip plus labour. (Or it could be something else entirely. I can cobble together breadboards, but when it comes to actual boards then I'm WAY over my head!)

User avatar
obitus1990
New Orleans, LA, USA

Posted Fri Mar 12, 2021 3:41 pm

Testing resistors (or any other passive component) in place (i.e. without desoldering them) with a multimeter will often yield unreliable/incorrect results.

User avatar
ubermick
California, via Cork

Posted Mon Mar 15, 2021 3:52 pm

obitus1990 wrote:
Fri Mar 12, 2021 3:41 pm
Testing resistors (or any other passive component) in place (i.e. without desoldering them) with a multimeter will often yield unreliable/incorrect results.
The board is winging its way to Acill as I type. I have little doubt it will return in perfect working order!





Return to “Hardware”