Atari STE only boot in High Res

[Mr Nours]

Hello Hardware Gurus,

I recently bought to a musician an Atari STE with an SM 124 monitor. Seems to be working fine, but when i go back home i plugged it on my low res monitor and... White screen, then black screen.
Plugged it back to the sm124, worked fine.
So this one just boot in high res!

I tested it with the diag cart, all seems OK ( timing, cpu, ram, sound... ).

So i opened it, it seems tehre's some odds things on the motherboard but don't know if this could be related.

I post you some pics here, perhaps you could help me find the origin of the problem.
Got some readings, some says it could be a faulty u504 chip, some other says to replace the MFP chip...i'm a bit lost






U504 has a patch wire...


C230 has some kind of piggyback stuff toward the MFP...

[Arne]

The HW patches are factory made. Mine has them, too.
First of all: what tools (multimeter, scope, soldering-station,...) do you have at hand and are familiar with?
Second: can you read the 1040STE schematic?

[Mr Nours]

Hello Arne,
Thanks for the reply.

I've got an iron soldier, a multi-meter and have moderate skills with them.
I can hardly read the STE schematics...seems very complicated to me to understand.

Well, you got me, i'm an hobbyist!

Is this resolution problem frequent on the STe range?

[Arne]

Is this resolution problem frequent on the STe range?

I really don't know.

Switch off the STE, disconnect the power cord. The PSU can hold hazardous voltages even if switched off. So better wait 15mins until starting your measurements.
On your last photo you see resistor R215. Check the resistor's value. It should be in the 4.7K range.
Then check if the resistor's right pin has continuity to U209 (The 74LS86 just above it) pin 12. The trace is on the top and visible on your photo.
Come back here with results.

[spiny]

>So this one just boot in high res!
>I tested it with the diag cart, all seems OK ( timing, cpu, ram, sound... ).

Does the diag cart boot to medium res when you have the colour monitor connected ?

[joska]

Musician's STE -> maybe moved around a lot -> monitor socket may be worn/broken. Check that the mono detect line isn't permanently grounded.

[Mr Nours]

On your last photo you see resistor R215. Check the resistor's value. It should be in the 4.7K range.
Then check if the resistor's right pin has continuity to U209 (The 74LS86 just above it) pin 12. The trace is on the top and visible on your photo.
Come back here with results.

Arne, here are the results :
R215 value = 4,66 kΩ
The continuity test is ok (beep!) between r215 right pin and U209 pin 12 just above.

Does the diag cart boot to medium res when you have the colour monitor connected ?

Don't know ( Black screen ), but The ST is still responsive, and if i blindly select some test ( for example sound test ) they seems to work ( i hear sound ).

Check that the mono detect line isn't permanently grounded.

How do i do that? I will search where is this line...

Thanks for all your answers

[Mr Nours]

Deleted

[Mr Nours]

Deleted

[Arne]

Arne, here are the results :
R215 value = 4,66 kΩ
The continuity test is ok (beep!) between r215 right pin and U209 pin 12 just above.

Then the monitor detection should work, especially if you can blindly handle the diag menu in colour mode (sound).
So I guess the problem is more output related i.e. Shifter to monitor out plug.

[joska]

What kind of monitor are you testing with? A modern LCD may give you some hints, e.g. it would probably say something like "no video" or similar if there's no sync signals or they're out of range, but only display a black screen if the syncs are OK but there's no RGB output.

Also, what kind of cable are you using? RGB or composite? Looks like you have a french STE with no modulator, then there will be no composite signal either.

[spiny]

Don't know ( Black screen ), but The ST is still responsive, and if i blindly select some test ( for example sound test ) they seems to work ( i hear sound ).

Check that the mono detect line isn't permanently grounded.

Colour diag cart should look like this:

IMG_20200218_124137246aaa.jpg

and Mono like this:

IMG_20200218_124150395aaa.jpg

If you get noting on the colour, but it does boot, then I'd agree with Joska, the mono detect could be stuck on mono. Are there any obvious bent pins on the monitor port ?

[Mr Nours]

Diag Cart on sm124 :


On the sc1425 screen i got nothing but black screen.
Same on a modern lcd screen ( "no image" ).

So I blindly forced the STE to go into med res with the memory examiner.
( by typing E then W ff8260=1 )
And got this :



I assume that there are some other registers to set to get a proper display. That was just for testing if the ST could emit a med-res signal...
And got this with the same maniuplation on the modern LCD :


No obvious cable bent on the connector side, just to be sure i reflowed the pins. Same story. Checked some continuity, seems ok. What is the pin responsible for the low-res detection?
Is there a way to force the hardware to go low res ?
I use a RGB scart cable for the tests.

Again thanks for all your messages.

[joska]

Ok, what you see is a monochrome framebuffer displayed in med-res. So video output appears to be working correctly, but TOS always detects a monochrome monitor.

When pin 4 on the monitor socket is grounded, TOS use monochrome mode. If it's not grounded, TOS will only allow colour modes. Disconnect the monitor, and see if there's continuity between U405 pin 6 and ground. If there is, there's either a problem with U405 or the monitor socket.

Also, I see from the schematics that U211 is involved in the mono detect circuitry. In your STE a chip (U800 (I believe, poor quality scan) in the schematics) has been piggybacked on top of this chip. There could be something there. Maybe a short that has not been detected previously if the previous owner only had a mono monitor.

[Mr Nours]

When pin 4 on the monitor socket is grounded, TOS use monochrome mode. If it's not grounded, TOS will only allow colour modes. Disconnect the monitor, and see if there's continuity between U405 pin 6 and ground. If there is, there's either a problem with U405 or the monitor socket.

PIN 6 is grounded whether the screen is connected or no. Should i change u405?

Also, I see from the schematics that U211 is involved in the mono detect circuitry. In your STE a chip (U800 (I believe, poor quality scan) in the schematics) has been piggybacked on top of this chip. There could be something there. Maybe a short that has not been detected previously if the previous owner only had a mono monitor.

This piggyback chip appears in other 1040 Ste motherboard, but it seems strange to me, checked the pins, seems ok... no excessive solder or whatever...

if the previous owner only had a mono monitor.

You're right, he only have a mono monitor.

Do you know if i could force the low res mode?

[neanderthal]

Pin6 on U405 like real grounded or logical?..If real then try to find out where its grounded.If logical might be chip itself(U405)..I read this mostly cause busted my U405 on a machine testing something ages ago.And only afterwards reconized it has to be a 74S version..So I had to use a pullup on the pin for used some HC version replacement.

[joska]

PIN 6 is grounded whether the screen is connected or no. Should i change u405?

Interesting. This is a 74LS257. Pins 5 and 6 are inputs, and the output from these are on pin 7. Pin 7 goes to the mono detect circuit, pin 5 is always 5V and from what I can see of my poor quality STE schematics U405 pin 6 is connected directly to the monitor port pin 4 (mono detect) and nothing else. So if pin 6 is floating (not connected to anything as would normally be the case with a colour monitor) the high pin 5 would cause pin 7 to be high. If a mono monitor is connected pin 6 would be low (earthed) which would cause pin 7 to be low as well. But in your case pin 6 seems to be always low.

If this was my STE I would have separated pin 6 from the motherboard (cut pin or cut trace between U405 and monitor socket) and measured again on both sides of the cut. One of them should still be grounded, that's where the problem is.

[neanderthal]

And if truly grounded like almost 0 ohm style,I think that trace is totally visible on the solder side of board.Maybe check it out with magnifying glass or something first.That is pin 4 on monitor connector to pins 2 and 6 on U405.I have found stranger things with just visual fault tracing.

[frank.lukas]

This is a 74LS257

It is a 74S257 or a 74F257 ...

[Mr Nours]

Here's the backside of the u405 area.
I tested pin4 of the video connector by inserting a wire in it.

I got continuity between video pin4 and u405 pin 2 and 6 and ground.

Resistance between video pin4 and ground is about 1,6 Ohm.



If no one ha other idea, i think i will cut the trace to see where the ground could be connected to pin 4...
Do a faulty u405 could cause a ground connection to pin 2 & 6 ?

Again thanks, it is very interesting. I learn a bit on how to track a fault...

[Mr Nours]

Frank Lucas, the U405 is labelled as a a 74S257N...

[frank.lukas]

Pin7 from the 74S257 must be change the Level from high (+5V) to Low (0V) when you connect a SM124 to the Monitor Port or disconnect the SM124 ...

ste_video.png

[neanderthal]

About 1.6 ohms or so?,Trixy.depends a bit on your probes and such.Rather easy to get some ohms when using connectors and so.Pity I cant remember what had to change my U405,exept that got the same style of problem when used a HC style chip.Then again pins are exposed directly to external so a direct charge into it can bust the ovp in the input(if any).Test measuring direct on the pins, almost 0 ohms sounds a bit low for chip fault? Then again,could be.Otherwise you looking at a odd mobo short or something.Main thing,you know what causes it

[Mr Nours]

Some measurement with monitor unplugged :

U405:
Pin 1 = 5.02 v dc
Pin 2 = 0v
Pin 3 = 0.924 v
pin 4 = 1,374 v
pin 5 = 5,02 v
pin 6 = 0v
pin 7 = 191 mV
pin8 = 0v

So clearly i got no 5V on pin 7.

Could i solder a wire between pin 1 and 7 to force the ST to go in low mode?

[neanderthal]

Some measurement with monitor unplugged :

U405:
Pin 1 = 5.02 v dc
Pin 2 = 0v
Pin 3 = 0.924 v
pin 4 = 1,374 v
pin 5 = 5,02 v
pin 6 = 0v
pin 7 = 191 mV
pin8 = 0v

So clearly i got no 5V on pin 7.

Could i solder a wire between pin 1 and 7 to force the ST to go in low mode?

Hmm,,would most likely just cause machine to behave real odd,the select (pin 1)is used for switching between external and internal main clock.I think the idea was to use some sort of genlock or something synchro via
injecting external 32Mhz into mono pin?,I've never seen anything that uses it tho.

Never mind since the problem is that mono mon detect input should not be in the range of 0-2 ohms when measured with a
multimeter.The question is that if it is U405 itself or something in the trace on mobo to connector.
You didn't see anything around the trace?,small solder blobs or micro thin copper between trace and ground?