• Please review our updated Terms and Rules here

Sinclair Spectrum 48k issue 6A 12v drops to 10.5v when video Ram installed, but supplies 12v when video ram removed. RAM okay.

P

pocoaust

Member
Joined
May 31, 2022
Messages
35
G'day,

I have a Sinclair Spectrum 48k which I have ALMOST repaired. I have changed CAPS, replaced Video Ram and it boots diagnostic ROM, but in black and white, passes first RAM test then fails on all subsequent tests. I have pinned it down to the 12v being generated by the DC-DC circuit. It is measuring 10.3V at video ram, -5 and +5 are good. If I remove the 4116 ram the voltage goes back up to 12v and the screen displays Colour. I have replace the TR4 and TR5 but they are aliexpress sourced, so I am wondering if they could be dodgy. I have heaps of tested 4116 ram and tried 3 different batches, same issue.

Does anyone know what voltage i should see on the output of the coil?

I have measured the resistance and they seem okay, no shorts and continuity on each coil, but could the coil be failing under load give me such an issue?

Anyone experience similar issues?

Thanks....
Paul in Victoria Australia
 
Is your AC voltage dropping at all under the observed voltage drop? The regulation of the +12v DC from 12v AC seems to be just two diodes. Perhaps they are failing under load?
 
Thanks for the reply,

AC is fine. Used an original Spec supply that works with other machines and also tried a bench supply, no difference. It is dropping down only on the 12v side, 5v, -5 etc are fine.

Any other ideas?

Thanks
Paul
 
I am only getting 10.8V from the coil output? Has anyone measured the voltage at this point?
Thanks
Paul
 
@pocoaust It might be worth trying to measure the current drawn from the +12V voltage regulation (or boost conversion) circuit to see whether it is particularly high.

You can use a multimeter for that, but it's a bit more complicated than simply measuring resistance or voltage. It may help to throw together a quick test rig with some wires to form the needed loop and some resistors to prevent overloading the multimeter.

If any of the ram chips are socketed, pulling one out and using the socket as a test point may be one way to go about it.

Alternatively: Some internet sources propose that one or more ram chip may be shorting to ground in some fashion that results in drawing enough current to cause a dip in the voltage. And another possible cause could be bad switching transistors (TR4, TR5) in the DC-DC conversion part of the circuit.

P.S.

^ there may be some useful information in this video
 
Hi Paul,

10.5v is low, but not entirely outside of Spectrum voltages... A bigger question is whether you have -5v and it's stable.

Do you have a retroleum board or similar test ROM system? They can convey a lot of information on a non-working board as long as the ROM is operating.

The ZX Spectrum has a small oscillator that generates both the 12v and the -5v supply.

If the -5v line is suppressed for any reason, then you'll get a problem on the 12v line... Also the video modulator uses the 12v line.

David
 
You are measuring voltages using what (a multimeter or oscillsocope) and whare are you measuring the voltage across?

The oscillator produces 12V AC, -5V, +12V and +12VA.

The -5V and +12V are used by the DRAM and must be within +/- 5% of the rated value.

The -5V is derived from the output of the oscillator, which is taken from the same point as the +12V (which is the 12V AC point).

When you say you are only getting 10.8 V from the coil output - what are you measuring it with and between what two (2) points.

This is likely at a high frequency - and if you are measuring it with a multimeter - it may be giving you a false reading.

Dave
 
You must log in or register to reply here.
Back
Top