Sinclair reliability

General Chit Chat about Sinclair Computers and their Clones
gozzo
Posts: 452
Joined: Fri Jul 08, 2011 8:52 pm

Re: Sinclair reliability

Post by gozzo »

Yeah....if you've got any 4v3 zener diodes... :( BY127's I have plenty of.. :) I do also have a +2A PSU, but that's at the back of the barn behind all the boxes of components, laptops, other computers, valves, etc.!!! Also have a smaller 'laptop charger' size multivoltage thing, will have to see what the current rating is, that might do OK. :)
User avatar
1024MAK
Posts: 5118
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Sinclair reliability

Post by 1024MAK »

If no suitable zener diodes, you can use two resistors. One connects between the regulator common terminal and ground (RB), the other one connects between the output terminal and the common terminal (RA). I have the formula somewhere.
Ah, here...

Vout=Vref*(1+(RB/RA))

So RA=390R, RB=270R, assuming a 7805 with a common pin current of 3mA, output voltage is 9.2V

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
RetroTechie
Posts: 379
Joined: Tue Nov 01, 2011 12:16 am
Location: Hengelo, NL
Contact:

Re: Sinclair reliability

Post by RetroTechie »

I'd take an LM317 for that, and besides 78xx regulators cost peanuts. But just realized what a silly direction this thread took: why you need 9V? :? :?:

There was talk of powering with an external supply. That is: bypass onboard voltage regulation. For a ZX81, that needs +5V only.

For a Spectrum, that needs +5V, -5V (for those :evil: 4116 RAM chips), and +12V (for same, and video encoder). Where's a 9V in there? Or am I missing something here? (Spectrum 128K / +2 / +3 machines?)
gozzo
Posts: 452
Joined: Fri Jul 08, 2011 8:52 pm

Re: Sinclair reliability

Post by gozzo »

Theres a good reason for the 9v.. I was thinking of deactivating the oscillator transistor by shorting base to emitter, feed 9v(ish!) to the usual 9v in, feed the +12v and -5v to the +12 and -5 points on the board through low value resistors (10 ohm or so), or a 100ma fuse in each , and then see what happens.... I could feed +5v in to the +5v line but I'm not sure if the regulator would like it having +5v forced onto its output with nothing on its input - although I could probably link the input to +5v as well, just so there's something there...???!!!This is only for a spectrum. In the case of a zx81, bit different as would need to feed 11v into the '9v' in, especially if testing it with certain RAMpacks, due to the cheapo circuitry used! This is purely for testing purposes, not actual long term usage!! All this is to find out why the oscillator transistor overheats rapidly and destroys itself! Have had this happen before when I stupidly inserted an interface while the power was still connected to my spec 128. :oops: . the transistor failed and replacements failed instantly at switch on...I had to replace the oscillator transformer to cure it as it appeared to have been damaged - overheated? - there was certainly a whiff coming from it - when the original transistor went short circuit...I was very lucky to be able to get a replacement as this was some years after the 128 was current and the local TV/video repair shops supplier still listed them! :) ! What voltage checks I've been able to do)quickly) indicate the +5v and -5v line ok, but the +12v line low, I think either one or more 4116's drawing heavy current, or the LM1889 colour encoder doing the same, although I do get up a picture of sorts with coloured blocks all over, so that's probably ok! I know 78xx's are pretty cheap but I was trying to use what I've already got lurking around, without having to buy anything if I can help it !(got a few spare new-old-stock 4116's luckily - hope they're ok... :?: !)
User avatar
RetroTechie
Posts: 379
Joined: Tue Nov 01, 2011 12:16 am
Location: Hengelo, NL
Contact:

Re: Sinclair reliability

Post by RetroTechie »

gozzo wrote:although I do get up a picture of sorts with coloured blocks all over, so that's probably ok!
Yes, very likely a problem with the lower RAM, one of those 4116 chips (as so often with ZX Spectrums :x ). You could try feeling if one gets hotter than the rest, or (with board powered off) some resistance measurements between D and Q pins, and GND / 5V supplies (wouldn't know exactly which, try various pin combo's). Maybe there's one chip that deviates noticably from what you get for the other IC's.
Btw better move that problem to WoS forum, they've seen this pass by many, many times before. ;)
gozzo
Posts: 452
Joined: Fri Jul 08, 2011 8:52 pm

Re: Sinclair reliability

Post by gozzo »

Got a couple of duff memotech 16k packs as well...most likely the dreaded 4116's again, I suppose!!!!
Post Reply