Need help pointing to strange video problem on SE/30

I am working on a buddies SE/30.

This one here seems to work, it boots up… sound works. etc.

Except this video issue. its like a whole block of video lines are out. if i had to guess i would say lines 0123(work) and 4567 are dead.

I did some continuity checks, from UE8 and UC6 - UC7 and only one line was out, I ran a wire and fixed that.

before i spend any more time on this can anyone look at this and help me shotgun the issue?

there was quite a bit of cap leakage around UE8.

there was quite a bit of cap leakage in general.

here is a picture of what i get on the Tube,.



thanks

Issue cant really be "shotgun guessed" by anyone else but yourself, Only reason why I say this is because every failure is going to be unique in which traces and chips are gonna be bad in your board.

But I digress...

But i think your right, your missing the upper nibble of the databyte. And an easy way for that to be the case, is the upper/lower select signal.

Yeah I'd agree with techknight, your upper 4 bits of the Vid signal are out. So your issue is centred around UC6 as Vid(4) to Vid(7) are stuck on low (screen black). It's possible that those four lines happen to be bad but somehow I reckon that it may be on the input side of the chip.

I would get the schematics and check every connection on UC6 including power / ground. I'd say you'll find something wrong there. If not, work your way back to the mux's UA8 to UD8.

Good luck !

Thanks. I'm spent on time right now.. i have other boards to do. 3 more se/30s jus came in. and a nice IIcx is waiting for service.

i do agree that the issue is something with UC6 or something that should be feeding UC6.

i appreciate your input as always. really!

hey guys the only thing i can think of is UC6 is bad... one other thing i would like to say it seems UG7 gets hot to the touch after not too long at all.. all the other chips around it do not even get warm.

i am wondering why... i swapped that chip with one from a donor board, it still gets hot like the original one. and still has video issue. I used my meter, and man its like every dam trace is proper, i can't find any issues...

i would swap out UC6, but its soldered, everything to it appears to be the same as UC7 i even checked voltages with the board on.

I cant swing any more time on this for now, my buddy is gonna have to spend some more hours with it.. Maybe he can clear the video up.

other wise, all works...

If UG7 gets hot, there's definitely a problem there somewhere. I guess you took out UG7 and checked all the pins for any shorts already. I'd even check if any of the pins are somehow connected to each other when they shouldn't be and that none of the pins are shorted to power or ground.

I'd also be careful not to damage those PAL IC's like UG7...not like you can buy more...hot = bad

Troubleshooting...if only there were more hours in the day...

PS: had a quick look: UC7 pin 1 receives it's clock signal from UG7 pin 15. Also UA8 pin 6 receives multiplexer input from UG7 pin 12. So there is some interaction between the possible problem sources. If one of the mux inputs is bad it would mean the vrams UC6 / UC7 wouldn't be addressed correctly. Would that result in the pins stripe? I have no idea right now...maybe worth checking those two connections though...

UC7 pin 1 receives it's clock signal from UG7 pin 15.

Click to expand...

i checked that and it doesn't get 000 continuity it gets 022

so i guess that means there is some resistance.

Also UA8 pin 6 receives multiplexer input from UG7 pin 12

Click to expand...

i checked that and it gets 000 continuity, so its good.

thanks man.

The 22 is actually correct. There is a 22 Ohm resistor in between so that's OK as well.

I guess you just have to start checking out UG7 and anything it's connected to...bummer!

Yeah with this stuff it's time, time, time,

and unfortunately I'v way over extended my time budget on this thing.

my time at home when i am not at work is really getting quite valuable these days.

between the re-caps and the kids… an occasional bite to eat, and some sleep in there some where… there aint much left

Mine is like that even without kids in the mix. The months go by like weeks.

And, with my sister having a kid, years seem to go by like weeks.

-J

PAL ics in those things do run quite hot. This is normal.

Didn't realise that, but good to know. Cheers!

I'v found the problem

in this rare occurrence the issues was the UE8 I/C its self.

i changed it with a parts se/30 board and the video not works fine.

i also fixed 2 other se/30 boards with the gangster stripe issue tonight.

so it looks like i'm on a roll.

Yeah with this stuff it's time, time, time, and unfortunately I'v way over extended my time budget on this thing.

my time at home when i am not at work is really getting quite valuable these days.

between the re-caps and the kids… an occasional bite to eat, and some sleep in there some where… there aint much left

Click to expand...

I forgot to add my normal day job in there

Nice one man. Those gangster pinstripes are in for a tough time

so far pin3 UE8 to pin3 UG7, its the common problem gangster stripe.

man weekends are cool i can get caught up!

Vias can be notoriously problematic on some boards. They are created by a relatively thin layer of plating applied to the hole drilled through the board. Contact area with the traces on each side is very small and over time thermal expansion, flexing of the board, and/or corrosion can damage the via and you get an open connection. There can be dozens and dozens of vias and all it takes is one bad one to cause a problem.

yes true.

it sure is a problem.

that cap goo + rot = bad.

looks like I have another case of Diarrhea Screen.



what were the pins to fix this again?

UK6 ## to ???## forgot.

hey check out these hill billy ram slots, they broke apart right when i tried to install a ram slot.

Ok i fixed it.

I copied genie_mac's image and it works.

Just like 68kbits did as well.

I made a few changes to genie_mac's image

Last one on by list...

No bong,

no boot,

no boot floppy/?

just this:



Could maybe this one be a network filter issue?

Looks and acts just like the board you sent to me to have a look at. All 64 of the ROM connections look good. All of the caps and connections from them are good.

hey maybe we should change the resistor networks and see if that helps my good man?

i know that is a issue, but that is about all i can shot gun right now , i know techknight has one of JDW's boards that is doing something like this to.

Can you do that good buddy…? I don't have any networks right now.

i could remove some from that se.30 parts board but that is extra annoying.

Also i think the one you have has a different video pattern on the screen, can you post a picture of that one's screen?

I'll see what I can do in the next couple of days.

hey check out these hill billy ram slots, they broke apart right when i tried to install a ram slot.[attachment=1]Screen Shot 2013-09-16 at 9.43.30 PM.png[/attachment]

Click to expand...

Looks like its time for some new SIMM slots all around. That cap leaked pretty bad too.

i think the cheapest i saw networks for was 1.68 a piece but you had to buy 20.

yeah for sure on those ram slots. never had slots break apart like that on me.

the leaky cap goo really tore this board up. - The funny thing is, now i have it all patched up it works great.

hey mcd, not sure if you wanted to do JDW's too.. i know techknight is mega busy these days

if you fixed them you would be the man and the hero!

It's pretty easy to test resistor networks, I wouldn't go replacing those unless the ones you have are bad. Usually checking them in circuit will give you a pretty good idea. Comparison with a working board will remove any doubt.

Its not the resistor network, at least not on JDW's board.

Check the ROM Select function from the GLU. Make sure the ROM is getting moved to 0000 on boot, JDWs is not. Why? bad GLU or possible bad co-processor preventing the main processor from running.

What we need is get bbraun or doug3 to make a custom firmware that boots the machines and tests all addressing/hardware functions primitively, and flash an LED a code where it fails.

The ROM kinda has that already builtin with the diagnostic mode. Grounding the VIA1 pin should work on the SE/30 too.