ZX Spectrum Issue 3 repair

Tezz
Posts: 186
Joined: Thu Mar 09, 2017 8:28 pm

Re: ZX Spectrum Issue 3 repair

Post by Tezz »

thanks Mark I should have mentioned that I tested the voltages first thing. They seemed to be within the expected range so I moved on to powering the board. I'll desolder the other 4 lower RAM and install sockets next and see how that goes. I have a few replacement modules so I'll try that first and if all good I'll see if I have enough good 4116s within what's left and install them instead
Tezz
Posts: 186
Joined: Thu Mar 09, 2017 8:28 pm

Re: ZX Spectrum Issue 3 repair

Post by Tezz »

I've removed the 4 soldered lower RAM chips and installed sockets. I've tried a lower RAM module and I now get coloured blocks with garbage on the output (see attached photo). The new sockets have a lip on them whereas the originals don't so the module doesn't sit straight but it should be still making contact. There's actually a jumper wire installed between two pins of the 4 existing lower RAM sockets so perhaps a bad job was done by the previous owner. I assumed the sockets were Sinclair installed as this issue 3 board was inside an "Upgraded" Spectrum+, I can't remember if Sinclair let you DIY upgrade your system with the + case though so I might have made an incorrect assumption.

I suppose I should take the module out next and continuity check all the lower RAM pins.

Edit: I checked some continuity between pins, the jumper is showing continuity and there are no bridges to any of the adjacent pins so that seems ok. I took out the RAM module and installed the 4 new and tested 4116 chips and the 4 chips that were originally socketed on the board and I got a different screen of garbage with no colour present. I really need 8 good tested 4116s to take a lower RAM fault of of the equation.

I mentioned C44 earlier and funnily enough the outer casing of the capacitor is coming away, I should really change that next to be sure.

issue_3_fault2.png
issue_3_fault2.png (433.22 KiB) Viewed 5073 times
Back of the board before installing the 4 sockets
Back of the board before installing the 4 sockets
back_board.png (470.8 KiB) Viewed 5073 times
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum Issue 3 repair

Post by 1024MAK »

Sinclair sold a lot of the Spectrum+ cases as upgrade kits... (these did not have serial numbers stamped into the plastic).

You may find ANY issue board in a Spectrum+ case! In theory, it is possible that a Spectrum+ is a actually a 16k Spectrum!

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum Issue 3 repair

Post by 1024MAK »

It would be wise to check ALL of the tracks to the 4116 DRAM chip sockets. All the 4116 DRAM address pins should link across to each and every 4116 DRAM, and to the ULA, and to the two multiplex chips.
The data in should link to the data out pin on each chip, then go to the ULA data bus, and to the resistors that link each data line to the CPU data bus.
Check also the /RAS and /CAS control lines. Also check for short circuits between each pin and each track...

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum Issue 3 repair

Post by 1024MAK »

Oh, and note that the schematic is not 100% accurate. Two or more data lines can be swapped, as the RAM does not care which bit it stores. Sinclair routed the tracks for the easiest layout rather than exactly to the schematic.

Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
Tezz
Posts: 186
Joined: Thu Mar 09, 2017 8:28 pm

Re: ZX Spectrum Issue 3 repair

Post by Tezz »

Thanks Mark, I'll work my way through those tests in the morning.
Tezz
Posts: 186
Joined: Thu Mar 09, 2017 8:28 pm

Re: ZX Spectrum Issue 3 repair

Post by Tezz »

OK, morning is here :) I've now checked all the DRAM address pins across each DRAM and that was all good, I've checked every pin for shorts and that was also ok. I checked data in to data out pins, all good. RAS and CAS between DRAM and ULA is good. The DRAM address inputs A0-A5 to the ULA pins A0-A5 gives no connection (I'm using 200omh on the meter btw) should that be the case? The two multiplex chips IC3 and IC4 I've not checked yet, IC3 is socketed IC4 soldered. I have the upper RAM upgrade chips which includes 2 SN74HCT157N chips, would this (or any of the 157s) work in IC3/4 too if they indicate a fault? They are currently SN74LS157N
Last edited by Tezz on Sat Jun 03, 2017 12:57 pm, edited 1 time in total.
Tezz
Posts: 186
Joined: Thu Mar 09, 2017 8:28 pm

Re: ZX Spectrum Issue 3 repair

Post by Tezz »

looking at the schematic I've checked continuity between ULA A0-A6 and the relevant pins of IC3/4 and they gave a brief reading on the meter before showing open circuit
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum Issue 3 repair

Post by 1024MAK »

Tezz wrote: Sat Jun 03, 2017 11:57 amThe DRAM address inputs A0-A5 to the ULA pins A0-A5 gives no connection (I'm using 200omh on the meter btw) should that be the case? The two multiplex chips IC3 and IC4 I've not checked yet, IC3 is socketed IC4 soldered. I have the upper RAM upgrade chips which includes 2 SN74HCT157N chips, would this (or any of the LS157s) work in IC3/4 too if they indicate a fault? They are currently SN74LS157N
Apart from when there are resistors in circuit, yes the 200 ohm range is the correct range. Switch up to the next range (2000 ohm or 2k) to test the circuits with resistors.

For testing for short circuits where there should be no circuit, use the 2000 ohm or 2k or 20k range.

The seven DRAM address lines A0 to A6 (not to be confused with the CPU address lines, which are physically different circuits and tracks) should go from the DRAM chips to the ULA (pins 5 thro to 11) and to the 74LS157 chips (IC3 & 4) via resistors (R17 thro to R23, all 330 ohm). Just like the data lines, easy routing of the board was done, so these address lines may be swapped around!

Also check the /RAS to IC3 & 4 via R32 (100 ohm).

Yes, 74HCT157 "should" work (not tested by me). As one of these has been changed, look carefully for any solder or track problems here.
IMG_4697.JPG
Issue 3 4116 DRAM & ULA
(538.3 KiB) Downloaded 357 times
Mark
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum Issue 3 repair

Post by 1024MAK »

Note- Post above edited.
ZX81 Variations
ZX81 Chip Pin-outs
ZX81 Video Transistor Buffer Amp

:!: Standby alert :!:
There are four lights!
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb :!:
Looking forward to summer later in the year.
Post Reply