horizontal line interference on screen

[Marcus7]

Good Evening.

I bought a ZX Spectrum 48k Issue 4s motherboard off of Ebay early this month.

It had 2 missing IC's, IC1 and IC12. I have since added the ULA 6c001e-7 and an TMS4116-15NL.

Also I have replaced the electrolitic capacitors, and changed the voltage regulator for a new switching mode reg. done the compisite mod (100uf cap). also had to replace IC15 due to being faulty which was flagged up by the diag board. The upper ram are all TMS4532-15NL3 which I replaced the faulty one with the same type 15NL3.

Now everything is passing when Diagnosic board fitted.

But when powered on and with the Sinclair research message at the bottom of the screen, I am seeing faint dark horizontal lines apearing and disapearing on sreen. I am not sure what is causing this?

Screenshot of dark line



Photo of the issue 4s board



I also have a video file of this but unsure how to attach.

I have tried another LM1889N and changed the um1233 modultor but still get the same dark lines appear and disapear.

Have tried 3 other motherboards Issue 3b, 6a and another 4s using the same compisite cable and LCD TV (Samsung SyncMaster 2333HD) and they are ok, not seeing any dark lines.

The only test equipment I have is a multimeter (Robin AR4003) and a Retroleum "X-ROM" External ROM board diag board for ZX Spectrum and a Maplins N93CX Switching mode power supply 1-20v at 0-5A.

I'm still new at fault finding.

Any help would be much appreciated.
Thank You

Marcus7

[mark8bit]

is the ULA in a socket ? , if so you could try the ULA from one of the other boards just to confirm it's not that (being very careful not to bend/snap it's pins)

[Tiger]

Try to change your switching power supply ...

[1024MAK]

@Marcus7 is the horizontal line stationary, or does it move up or down?

You mention a Maplins N93CX Switching mode power supply. Are you using this to power the 4S board, or are you using an original UK1400 Sinclair PSU?

The best way to include videos, is to create a Youtube account (free), upload it to Youtube. The copy the link to the video here.

Mark

[hatman72]

I had something very similar to this and it was due to the power supply I was using

[mark8bit]

I have had similar with amstrad cpc’s but it would move the culprit was cheap ali/eBay power supplies .. never again

[jamesh]

But... There is MuRata OKI regulator installed (modded for horizontal mounting, from Retroleum?). That is supposed to "iron out" most of the PSU imperfections, isn't it?

Anyway, @Marcus7, trying another PSU is a sane idea and you are likely already have the one that will work fine. It is likely to have centre-positive plug, though. Perhaps, getting polarity changer (the one like this or this or this one) and trying another PSU may be a quick way to move further.

[1024MAK]

The +12V supply for the video section and the 4116 DRAM is derived from the DC input via the DC/DC converter based on TR4, TR5 and the coil, not via the 5V voltage regulator (whatever type that is).

Mark

[jamesh]

Oh, right, you mean that part ? Never really paid much attention to this, all power rails were alright on the board I have so I never really examined it closer.

Which brings us to the next question: what is the practical upper limit for the input voltage? Apparently, modern regulators can handle upper 20'sV when converting to 5V but I doubt this circuit will be able to cope with that much power. I may be wrong, but I am afraid not many sources mention that fact when they talk about contemporaneity replacement of the PSU.

[1024MAK]

The upper limit of the maximum input voltage depends on a number of things:

• For early issue boards (issue one, two, issue 3) where the DC/DC converter circuitry has not been upgraded to the latest (last) official Sinclair version, I advise not to let the input voltage exceed 11.8V. Otherwise there is a danger that the DC/DC converter may stop producing a -5V rail (which the 4116 DRAM will not like)

• Then there is the heat output of the 7805 (assuming it has not been replaced with a modern DC/DC converter device/switching regulator). As the input voltage increases, so will the amount of heat that the 7805 produces, which increases the heat from and the temperature of the metal heatsink.

• Finally, there is a ‘hard’ limit of the voltage rating of the electrolytic capacitors connected to the DC input circuit. If the capacitors that are fitted are rated at 16V, then you don’t want to go higher than 15V really.

Mark

[Marcus7]

Hi Guys,

by mark8bit » Thu Nov 16, 2023 7:59 pm

is the ULA in a socket ? , if so you could try the ULA from one of the other boards just to confirm it's not that (being very careful not to bend/snap it's pins)

Reply
I have today tried using the NEBULA chip by Retroleum, but still still seeing the lines.

by Tiger » Thu Nov 16, 2023 8:19 pm

Try to change your switching power supply ...
by jamesh » Fri Nov 17, 2023 3:08 pm

But... There is MuRata OKI regulator installed (modded for horizontal mounting, from Retroleum?). That is supposed to "iron out" most of the PSU imperfections, isn't it?

Anyway, @Marcus7, trying another PSU is a sane idea and you are likely already have the one that will work fine. It is likely to have centre-positive plug, though. Perhaps, getting polarity changer (the one like this or this or this one) and trying another PSU may be a quick way to move further.

Reply
Just tried again today using the uk1850 power supply from my spectrum 128k toastrack, still no change.

by 1024MAK » Thu Nov 16, 2023 10:41 pm

@Marcus7 is the horizontal line stationary, or does it move up or down?

You mention a Maplins N93CX Switching mode power supply. Are you using this to power the 4S board, or are you using an original UK1400 Sinclair PSU?

The best way to include videos, is to create a Youtube account (free), upload it to Youtube. The copy the link to the video here.

Mark

Reply

Have tried with a UK1850 psu today.
The Lines appear to be random in places on screen but do seem to appear and disappear and reppear in a downwould fasion.

Screen could be ok for 10 to 20 secs then see a couple of lines for a couple of seconds then ok again.




Looking at the caps they appear to be from 35v to 50v.

Regards

Marcus7

[Tiger]

Since swapping the power supplies doesn't help, the problem seems to be on the board itself. The coil seems to be a replacement - the "Samsung" coils of the 4S boards look different. Presumably, this replacement coil can be influenced by the frequency of the Murata circuit regulator and that's why the 50 Hz interference strip occurs. Maybe a shield between the voltage regulator and the coil will help




On the other hand, such interference can also be caused by the cable between the spectrum and the monitor...
I'm assuming you disconnected the 5V line from the UM 1233 modulator and there is only the 100µF Cap connected to the signal out jack ...

[1024MAK]

Apart from the TV and the ZX Spectrum, what else is connected to the TV, or powered up near the TV, the cables and the ZX Spectrum?

The Murata DC/DC converter will be operating at a much higher frequency than the coil, hence although it’s possible that one is interfering with the other, I’m not yet convinced that this would result in the symptom described.

A video may help.

Mark