Video and/or Power problem - Suggestions please!

[Harbars]

I have a rather frustrating fault which I have tried to narrow down, but am hitting the limits of my ability and patience!
For now I am stepping away from the board in case I make matters worse by doing something stupid (which may of course have already happened! )

Any suggestions as to what to investigate or try greatly appreciated....

Issue 5 machine, which until a week or so ago was working.
Now there is no video from the RGB (which goes to a SCART into a HDMI box into a TV).
Also no video from Composite, nor RF (although I can't be sure RF worked previously).

When I power up the machine, there is a signal from the RGB (the TV goes from "no signal") but just a plain black screen.
If I wait a few seconds and hit F1 or F2, MDV1 spins up - which leads me to believe it's booting fine, and there is a video or power problem.

I've taken the 8301 from the faulty machine and placed it into a known good board, and it works fine there.
I've taken a known good 8301 and put it in the faulty board and the same symptoms occur.
Actually I've done this will all the socketed ICs - with the same results.

Thus, I suspected some kind of power related problem....
IC28 (PAL encoder) gets rather hot, certainly much hotter than the known good board. I realize this IC is irrelevant when using RGB out, but it gets hot.
As does IC37 (+12 regulator) very hot. much hotter than in the known good board. I have replaced IC37 and no change. IC37 outputs 12.3V.

There is also a buzzing or hum which appears to come from C37. If I press it with my finger the buzzing quiets. I swapped C37 with the one from the known good board. Same symptoms.
I (think) I've checked everything else related to the power circuit, including the diode at the power connector (d25), and the other capacitors.

One weird thing with this machine is that the incoming AC voltage on the good board (taken from the back of the power connector) is 16.63. The same reading on the bad board is 15.60.
Same readings for D26.
This is apparently within "spec" based on the service manual, but it's with the same power supply, and I've even gone so far as to swap the power jack between the boards, and the same result. My theory being that perhaps there is something else contributing to the difference - and perhaps making IC37 work too hard. Might be complete hogwash.

Anyway, I'm pretty much stumped at this point, and would welcome any suggestions as to what else to check or ideas as to what might be the root cause - I'd love to get this board back up and running.

[Silvester]

What happens if you just remove IC28? (ie. just power up without it - it is stand alone circuit)

Perhaps CSYNC is being clamped by IC28 somehow thus no RGB screen (RGB & CSYNC come directly from ULA).

Also are you sure D25 is OK ?

[tofro]

Did you measure the board supply voltage, especially the regulated 12V? I guess you have some short circuit on the 12V rail.

Does/did the machine have anything connected to the expansion port? Often, there's just a bent pin in there that creates a short. Actually, there isn't much in the QL (the PAL encoder and the serial port drivers) that could otherwise create a significant load on the +12V to make the 7812 produce significant heat.

[Ruptor]

Voltage regulators can oscillate causing more heat if the tiny capacitors across there pins are missing but it is rare. It isn't being overloaded because it isn't shutting down.

[Harbars]

Many thanks for the tips. Some more information after following those leads...

Expansion port all looks good, no pins bent or any touching each other.

I removed IC28 altogether:
I get the same black screen
the C37 hum/buzz stops
the incoming AC voltage is up to 16.7v (15.6v with IC28 in the socket)
IC37 no longer gets hot it's input is 18.9v (15.8v with IC28 in)
IC37 output is 12.3v (same as with IC28 in)

The 12v rail is 12.3v at the output of IC37, the via north of C38 and on the positive side of C43, as is the 12v pin of the IC28 socket.

I put a known good IC28 (again, just in case) in and no change to the symptoms.
I had previously swapped D25 for one from a working board and had the same issues.

So it seems it's certainly something to do with the IC28 or it's supply...

Any other suggestions as to what to check out in terms of a possible short on the 12v rail. I feel like I must be missing something really simple and obvious!!!

[Silvester]

I removed IC28 altogether:
I get the same black screen

With IC28 removed you won't get anything on composite output (PAL), but RGB output should be OK since it comes directly from IC22 (ZX8301). Do you get display from RGB output? If that is OK then at least we have begun to isolate the problem to the area around IC28 (crystal X3 ??).

(I assume the swapped MC1377P still works in other machine, it itself is not suspect)

[Harbars]

Yup, all these tests are with RGB - I only tried composite and RF in the initial troubleshooting.

With IC28 removed I get the same signal - a plain black screen - no normal QL start up displayed, although MDV1 spins up after hitting F1.
If I put the removed IC28 in the known good board it works fine, doesn't get hot etc - so it seems the chip itself is OK, and there is something dodgy with the circuit or another component.

I also (with IC28 removed) swapped the 8301 for one from the known good board with the same result.

[Harbars]

Well it seems like it's a slightly more fundamental root cause... LOL.

There I was being a complete amateur and trying to investigate things with my terrible cheapo scope - checking the 8301 - all looks good, expansion port pins for RGB - check.
Then I move on to explore things in the vicinity of R49 which is one of the signals on the way to the 1377.
I'm using the can of the RF modulator as the ground hookup for the scope, and as I fumble around I notice a flicker on the TV and the briefest glimpse of the Minerva boot screen.

I quit faffing with the scope and put a bit of pressure on the top of the RF can, and hey presto the signal springs to life.

Lift my finger, black screen

Also if I rest the board on the mat, and bend the right edge up a bit - the signal works

I can't yet see visually any obviously broken tracks / dodgy joints etc on the board, nor have I ascertained if this also fixes the humming cap and dodgy IC28, but at least I know what the RGB issue is. Some contact(s) someplace like a nice little lift from underneath!!!

[Pr0f]

"Re sweat" the solder joints near power connector, larger caps, the RGB socket - basically just hold the soldering iron on the join until it melts and then pop the soldering wire on it to introduce a small amount of new solder and flux - let it mix in - a couple of seconds probably all that's needed. It could be a joint where the wire of a component is not actually making good contact with the solder around it. Sometimes you can see them with a magnifying glass or lower power microscope.

https://en.wikipedia.org/wiki/Soldering ... joint2.jpg

[Harbars]

Turned out to be the RF can. the black shroud was pushing against the bottom of the board, tilting it up at the front of the board and down at the back. pushed the shroud so it's in front of the board itself and all good.

Twist it slightly in the other direction - video works (and buzzing cap and heating up components stop). Push it down towards the bottom of the board, video goes black and buzzing/heat starts. Couldn't determine what it was actually shorting or interfering with, but that was the cause in the end.