ZX Spectrum issue 1 testing

Post Reply
Tezz
Posts: 173
Joined: Thu Mar 09, 2017 8:28 pm

ZX Spectrum issue 1 testing

Post by Tezz » Thu Aug 09, 2018 2:55 pm

Hi all,

After years of trying to find one, I finally bought an issue 1 Spectrum last year along with an original Sinclair RAM 32 board. The Spectrum itself was in nice physical condition and after a very careful opening of the case I was able to detatch the original keyboard membrain which was still intact without damaging it. It hasn't been tested at all so I wanted to make sure to do the initial tests before attempting to power it on. I've never worked on an issue 1 before, are there any different steps I should take testing the board compared to the later Spectrum boards?

The RAM 32 board was separate and both of its connectors had some broken pins, one of the connectors had a poor repair with a thicker wire soldered offset to the original pin by the original owner. I'd like to replace both the connectors. I haven't yet found anything suitable in the UK although I did find one that could possibly be suitable on digikey but they are extremely expensive:

https://www.digikey.com/product-detail/ ... -ND/473245

I suppose I should look at dealing with the RAM 32 once I know that the Spectrum itself is functional.

Here's a photo of the worst of the two connectors with its previous poor repair of several pins
RAM_32.png
RAM 32 socket

User avatar
1024MAK
Posts: 2042
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum issue 1 testing

Post by 1024MAK » Thu Aug 09, 2018 4:30 pm

With regards to the issue one ZX Spectrum, it depends on what you want to do...
Do you want to keep it as original as possible, or bring it up to the last service manual status?
What do I mean by this? Well, the electrolytic capacitors under the heatsink are a required replacement. And you could try to bring the on board power supply circuitry up to the same level as detailed in the service manual.

So, in order for you to decide, here are the known problems with early ZX Spectrums:-
  • Poor quality of construction (my issue one had lots of poor quality solder joints).
  • Two design faults with the first ULA [ULA5C102E] (hence the dead bug modification), with the second ULA bug not coming to light until after the next ULA version was in production (issue two boards had the spider modification). If you want full functionality, both modifications can be fitted.
  • The electrolytic capacitors under the heatsink suffered due to the heat. So should be replaced with 105 degrees types.
  • Throughout the production run until the issue 5 boards, Sinclair continuity tried to make the on-board power circuitry more reliable. The version used in non-modified early issue boards would not work correctly if the nominal "9V DC" input was greater than 11.9V (the "9V" mains PSU is an unregulated type and when not at full load, the output voltage will be higher than the rated voltage, especially if the mains voltage is slightly higher than normal). If the on-board power circuitry does not work correctly, the eight 4116 (or equivalent) 16k bit DRAM chips can be damaged due to the -5V supply not being generated.
  • Not in the service manual, but it's recommended that the electrolytic capacitors in the on board power circuitry section be renewed.
  • A 22uF 16V (or better, a 25V type) should be soldered to the +9V and 0V/GND connections near to the DC power input socket (if not already fitted).
  • There is also some recommendations on some other components, but these are not critical.
  • Do read and check the service manual in case I have forgotten anything...
  • And lastly, when you do power it up, you may need to adjust VR1 and VR2 so that the correct colours are produced.
My issue one board (I don't have the case or keyboard) also has two modification wires on the underside, but I don't know if this is due to track damage, missing/incomplete tracks or something else. They are connected to the resistor network for the 5 pin keyboard connector.

Mark

User avatar
1024MAK
Posts: 2042
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum issue 1 testing

Post by 1024MAK » Thu Aug 09, 2018 4:38 pm

The expansion RAM board "turned-pin" connectors are hard to find. It is slightly easier to find the single in-line type. Or if you don't mind the contacts in the sockets on the issue one ZX Spectrum board from being bent, you can use normal square pin long pin-headers.

Mark

Tezz
Posts: 173
Joined: Thu Mar 09, 2017 8:28 pm

Re: ZX Spectrum issue 1 testing

Post by Tezz » Thu Aug 09, 2018 9:08 pm

Hi Mark,

Thanks for the reply and advice, I'd love to keep it as original as possible although I think it would be best to make the ULA modifications and replace the capacitors, particularily the ones under the heatsink as you suggest so that the issue 1 is usable rather than a museum exibit. The ULA modifications are non destructive so that should be no problem in any case. I noticed that some people recently have gone to the trouble of replacing the capacitors with light blue ones that resemble the originals, I think they're Vishay branded which I think is a nice idea.

Could I do the initial tests in the same way as the later boards such as testing the voltages of the 4116 RAM?

Here's a photo of my board attached.
motherboard.png
motherboard

User avatar
1024MAK
Posts: 2042
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: ZX Spectrum issue 1 testing

Post by 1024MAK » Fri Aug 10, 2018 1:51 am

So, looking at the photo you posted...

Your board already has the dead bug modification. The spider modification was not normally fitted to issue one boards, but it can be fitted.
Your board also has the 22uF capacitor already soldered to the +9V and 0V/GND connections near to the DC power input socket.

In terms of voltage checks, I suggest the following:-
First test the "9V" power supply with ONLY a multimeter connected to the output. Take note of the voltage (and while there, check the polarity). If it is 12V or less, but above 10.5V, then it's not going to harm the Spectrum.
If the voltage is higher than 12V, you may want to consider testing it on a test load. I use 12V car filament lamps (4W, 6W, or 10W).

When you first power the ZX Spectrum up, test and record the following voltages:-
+9V (available on pin 1 of the 7805 voltage regulator and on the edge-connector).
-5V (available on pin 1 of any of the 4116 DRAM chips and on the edge-connector).
+12V (available on pin 8 of any of the 4116 DRAM chips and on the edge-connector).
+5V (available on pin 9 of any of the 4116 DRAM chips and on the edge-connector).
IMG_7625.PNG
4116 DRAM pin-out
If the -5V rail is outside the range -5.5V to -4V, switch off at once.

The +9V should be between 10V and 11.9V
The +5V should be between +4.75V and+ 5.25V
The +12V should be between -10.8V and 13.2V

Mark

Tezz
Posts: 173
Joined: Thu Mar 09, 2017 8:28 pm

Re: ZX Spectrum issue 1 testing

Post by Tezz » Sat Aug 11, 2018 12:36 pm

Thanks again Mark, that's great. I'll run through the tests over the weekend. I'm using the original unregulated Sinclair PSUs with my Spectrums and Zeddies although they were all serviced and renewed last year. I will make sure to check the supplies under load with the meter before plugging any in the issue 1 once all the tests look ok :)

Post Reply