ZX Spectrum Issue 3 repair

[Tezz]

Hi all,

I bought an Issue 3 Spectrum a few weeks back, the board was a bit of a mess with the DC socket removed and a reset switch badly installled in its place with the power supply stripped of it's barrel connector and hardwired directly to the board. I've removed all the hacked up mess, cleaned the board and installed a new DC socket. There's a couple of traces cut on either side of the board however that need to be reconnected (See attached photos). Would it be possible to use copper conductive tape across the cut instead of installing jumper wires? or, is there another alternative method to make the repair look more presentable and like the original trace?

thanks

cut traces

issue3_dc_cut_traces.png (399.71 KiB) Viewed 6803 times

[1024MAK]

I think copper tape would work, as long as you quickly flooded solder on it before the adhesive backing on the tape becomes affected.

But some fine copper wire (the tinned type) would also work if you flood solder over it. As once you have something to help with the surface tension of the solder, it will happily bridge those track cuts You may find that you don't need anything to help the solder to bridge the gap.

Mark

[Tezz]

Thanks Mark, I'll try soldering over some tinned copper wire cut to size to fill the gap, I tried flooding some solder on the top side of the board but it resisted bridging the gap. I was thinking last night about cutting some component leads to fill the gap but tinned wire as you suggest would be better.

[Tezz]

As I already have copper tape, patience got the better of me and I went ahead and carefully cut some tape and applied it to the gap. I quickly flowed solder over the joint and it went well. The tape on the underside of the board between the DC legs did lift a little on one side but solder was bridged across ok and I checked continuity on the meter. I powered on the board now for the first time and there's no sign of life so I have some more investigation to do next week.

[1024MAK]

The next step is to test all four DC supplies: the nominal +9V input to the 7805. The +5V, +12V and -5V (three of the four corner pins of the 4116 "lower" 16k DRAM chips). See the link in my sig for pin-out details.

If the -5V is missing (or between +0.7V and -3V), don't leave it on. Switch off as the 4116 DRAM chips may be damaged with this supply missing.

No +12V means the video section is not able to function.

If the keyboard is intact, press a key and see if the speaker produces key "clicks".

Mark

[Tezz]

Thanks Mark, I normally run through the initial tests following Alastairs (zxzigg) guide before powering on but I was eager to try it out after the DC input repairs. I've checked with the multimeter now this afternoon and right away the 7805 doesn't appear to be right. Testing its left leg (following the guide) it should rise to over 100Kohm on the meter (which it does on an issue 3b board I have here to compare) but instead the meter immediately shows 2.9 on the display and stays that way. Despite this seemingly faulty regulator I continued to check the 4116 voltages. the upper right pins do all show 0v, the lower right pins however also all show 0v so no 5v showing there. The upper left pins -5v display 412 on the display with the meter set to the 2000 ohm range. The lower left pins 12v shows 2.7 on the display with the meter set to the 20k ohm range. The inductor winding ends displays 00.7 on the 200 ohm range and also correctly shows open circuit between the windings. Diode test on the meter for TR4: positive to base, negative to other legs shows 823 and negative to base with positive to other legs show open circuit. TR5 positive to base and negative to collector shows 955 on the display, and then negative to the emitter shows 833 on the display (at least I think I've written the emitter/collector readings the right way around there).

So I suppose the first step to progress will be to change the 7805 with a new one (which I do have) and then redo the tests?

[1024MAK]

Remove the 7805 and repeat the tests on the input and output pads without a voltage regulator fitted. If okay, fit a new regulator. It is possible that the Spectrum was connected with the incorrect polarity in the past. This kills the 7805 voltage regulator.

The internal switching supply needs the +5V rail present before it will function correctly.

Mark

[Tezz]

Thanks again, I'll try that tomorrow. With the mess the previous owner made of the board with a reset switch in place of the DC socket and the power supply hardwired to it with some traces cut I guess I shouldn't be surprised that the regulator (or more) has been damaged. Perhaps he/she wired the power supply to the board the wrong way around.

[Tezz]

I've now removed the voltage regulator and repeated the tests, there doesn't seem to be any significant change from the previous readings of the 4116 voltages. The lower right (5v) pins all still show 0v. How should I progress from here?

many thanks.

[1024MAK]

I was referring to the resistance tests, not voltage tests.

With the power disconnected from the board, and the negative meter lead connected to a 0V connection, what results do you get on the pad for the regulator input (the 9V nominal side)?

And what do you get on the regulator output (5V) side?

Mark