Question for Mark Re ULA internal regs.

Discussions about Sinclair ZX80 and ZX81 Hardware
Post Reply
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Question for Mark Re ULA internal regs.

Post by Moggy »

Or anyone else who might have an answer. I mention Mark as it touches on something he once pointed out re Ferranti series 2000 ULA internals.

As some may or may not know a lot of the heat given out by the ULA is due to the internal regulators dotted around the die having to sink what is to them the fairly high voltage of 5 volts used by the other chips down to a somewhat smaller voltage for its own use, the remaining voltage converted and given off as heat.

My question is this.

If feeding the 81's regulator with 7.5volts make that device run cooler would it be possible to say lift the ULA's 5volt and ground pins and feed it with say 3volts using a separate power supply? I know it would be messy to look at and cumbersome but I'm only interested in the theory which if plausible I would be willing to rig something up if only to prove the point like I did with lardo's question about fan cooling.

Also I assume Sinclair might have thought of this but decided on the cheaper and simpler 5Volt fits all design rather than a lower separate supply for the ULA.
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Question for Mark Re ULA internal regs.

Post by 1024MAK »

Actually, they effectively did feed the ULA with around 2.8V to 3.5V for the type of ULA used in the ZX Spectrum. But here they had just enough pins so that in addition to the supply for the “core”, they also could have a separate +5V supply for the input pin circuitry and the output drivers for the output pins, hence making it 5V TTL compatible (for use with a NMOS Z80, NMOS ROM and NMOS RAM chips).
And I wrote about it just today :mrgreen:

With the ZX81 ULA, they did not have enough pins for this, hence the single +V supply pin. [Side note, I never did understand why they included the keyboard input lines in the ULA for the ZX81, or the ZX Spectrum, having one extra inexpensive 74LS chip to handle the keyboard would have freed up four or five ULA pins].

It may well be possible to reduce the input voltage to the ULA slightly, and still keep everything else happy if the signals from ULA can still reliably get above the minimum of 2V which is the minimum level for TTL logic chips.

It may also be possible, if the ROM, RAM and Z80 are replaced with CMOS types, for the whole board (including the ULA) (but not including the modulator) to be run at a lower voltage.

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.
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: Question for Mark Re ULA internal regs.

Post by Moggy »

Excellent as always Mark.

I feel a project coming on!! :lol:
nollkolltroll
Posts: 325
Joined: Sat Sep 27, 2014 8:02 pm
Location: Stockholm, Sweden

Re: Question for Mark Re ULA internal regs.

Post by nollkolltroll »

Sounds really interesting, maybe a 3.3V CMOS rebuild would be possible?
/Adam
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Question for Mark Re ULA internal regs.

Post by 1024MAK »

This was posted on Spectrum Computing forums:
rastersoft wrote: Fri Apr 23, 2021 1:31 pm I found two scanned PDFs from Ferranti with design guidelines and info about their ULAs. I think that they are quite interesting.

https://archive.org/details/FerrantiULA ... 5/mode/2up

https://archive.org/details/FerrantiQ.R ... 1/mode/1up
(Link to post)

Notice the supply voltage range. Also notice the rated temperature range.

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.
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: Question for Mark Re ULA internal regs.

Post by Moggy »

Thanks for this Mark very interesting.

It seems that at least three volts is do-able and the 125 deg max temp spec is interesting considering that heat-sink purveyors reckon that without said sink the die is running at 50 odd deg which is way below max for this device and belies the myth of heat even being a problem and that clumsy handling of interfaces when under power etc is the main cause of ULA destruction.

I would add that I have no interest in converting my zeddies to run their ULA's under voltage as I don't believe heat is the problem many panic over, rather I'm in one of those "what would happen if I did this?" moods and like my fan cooling experiment I don't mind giving it a whirl. :lol:
User avatar
1024MAK
Posts: 5101
Joined: Mon Sep 26, 2011 10:56 am
Location: Looking forward to summer in Somerset, UK...

Re: Question for Mark Re ULA internal regs.

Post by 1024MAK »

Did you also see the power dissipation figure of 650mW (that’s over 0.6 Watts of heat) for a typical High Speed Array (HSA) ULA type, which is the type used in a ZX81/TS1000. It also gives the current consumption of 130mA at 5V (also 0.65W if the ULA is not sourcing any current out of it’s pins).

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.
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: Question for Mark Re ULA internal regs.

Post by Moggy »

Sadly it's a bit academic as it didn't work. :(

Tried at 3v nothing, 4.5v gave thin vertical lines but oddly though 5 volt from a separate supply just gives a blank screen,which leads me to believe that there may be a timing issue as well when it comes to switching on. Perhaps it comes down to all the chips getting voltage together.

Ah well NEXT!!!!


Forgot to say that all worked well when put back together.
Post Reply