User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Old fashioned fault finding....

Thu Jul 11, 2019 6:47 pm

Some may find this interesting :
https://www.tnmoc.org/notes-from-the-mu ... -the-diode
This is the machine that I emulate on my Pis so that I can develop software when I'm at home.
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

drgeoff
Posts: 9881
Joined: Wed Jan 25, 2012 6:39 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 7:25 pm

Very wise to replace both diodes. Until the mains fuse blew, the working diode would have been very stressed on alternate half cycles. Twice normal voltage and only the resistance of the transformer winding to limit the current.

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 7:39 pm

drgeoff wrote:
Thu Jul 11, 2019 7:25 pm
Very wise to replace both diodes. Until the mains fuse blew, the working diode would have been very stressed on alternate half cycles. Twice normal voltage and only the resistance of the transformer winding to limit the current.
Yes, which I why I was surprised they didn't both go. I don't know which fuse went first so there is no way to tell what the actual conditions were until they had both gone, but to be honest I was more concerned about uneven load currents due to differences in the characteristics between diodes manufactured 50 years ago and modern ones.

PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

drgeoff
Posts: 9881
Joined: Wed Jan 25, 2012 6:39 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 7:53 pm

PeterO wrote:
Thu Jul 11, 2019 7:39 pm
drgeoff wrote:
Thu Jul 11, 2019 7:25 pm
Very wise to replace both diodes. Until the mains fuse blew, the working diode would have been very stressed on alternate half cycles. Twice normal voltage and only the resistance of the transformer winding to limit the current.
Yes, which I why I was surprised they didn't both go. I don't know which fuse went first so there is no way to tell what the actual conditions were until they had both gone, but to be honest I was more concerned about uneven load currents due to differences in the characteristics between diodes manufactured 50 years ago and modern ones.

PeterO
There is no "uneven load currents". Both diodes are never conducting at the same time.

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:00 pm

Yeah, but they may have had uneven volt drops when conducting.
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
Imperf3kt
Posts: 2862
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:01 pm

The UPS consists of a 23Ah 26V Nickel Cadmium battery
When they were new a battery could power the computer for a few seconds, or with optional modifications to the power circuits for a minute or so.
That computer sounds like it uses a lot of electricity.
Time to replace it with a ~15W Raspberry Pi 4b?
55:55:44:44:4C
52:4C:52:42:41

drgeoff
Posts: 9881
Joined: Wed Jan 25, 2012 6:39 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:15 pm

PeterO wrote:
Thu Jul 11, 2019 8:00 pm
Yeah, but they may have had uneven volt drops when conducting.
PeterO
That wouldn't cause any problem. Compared to the 26 volts from the transformer any difference in diode voltage drop would have insignificant impact on the relative currents through the diodes.

drgeoff
Posts: 9881
Joined: Wed Jan 25, 2012 6:39 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:17 pm

Imperf3kt wrote:
Thu Jul 11, 2019 8:01 pm
The UPS consists of a 23Ah 26V Nickel Cadmium battery
When they were new a battery could power the computer for a few seconds, or with optional modifications to the power circuits for a minute or so.
That computer sounds like it uses a lot of electricity.
@imperf3kt
You are showing your age. Or rather, lack of it. :)

https://en.wikipedia.org/wiki/Elliott_803

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:28 pm

drgeoff wrote:
Thu Jul 11, 2019 8:15 pm
PeterO wrote:
Thu Jul 11, 2019 8:00 pm
Yeah, but they may have had uneven volt drops when conducting.
PeterO
That wouldn't cause any problem. Compared to the 26 volts from the transformer any difference in diode voltage drop would have insignificant impact on the relative currents through the diodes.
Transformer secondary voltage is nearer 60V peak at full load. The smoothing and regulation is not conventional for a low voltage high current supply by modern standards.
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
Imperf3kt
Posts: 2862
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:28 pm

drgeoff wrote:
Thu Jul 11, 2019 8:17 pm
Imperf3kt wrote:
Thu Jul 11, 2019 8:01 pm
The UPS consists of a 23Ah 26V Nickel Cadmium battery
When they were new a battery could power the computer for a few seconds, or with optional modifications to the power circuits for a minute or so.
That computer sounds like it uses a lot of electricity.
@imperf3kt
You are showing your age. Or rather, lack of it. :)

Correct me if I'm wrong (likely), but 23Ah at 26v is 598watt hours of juice.
Used in "a few seconds"

I know old computers used a lot of power, but that's a LOT
55:55:44:44:4C
52:4C:52:42:41

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:32 pm

Imperf3kt wrote:
Thu Jul 11, 2019 8:28 pm
Correct me if I'm wrong (likely), but 23Ah at 26v is 598watt hours of juice.
Used in "a few seconds"
I know old computers used a lot of power, but that's a LOT
Read the article (especially note Fig. 14). The battery is only "on load" during a mains failure. In normal use the battery charger is providing 26V and 44A , that's the real consumption !
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

drgeoff
Posts: 9881
Joined: Wed Jan 25, 2012 6:39 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:35 pm

PeterO wrote:
Thu Jul 11, 2019 8:28 pm
drgeoff wrote:
Thu Jul 11, 2019 8:15 pm
PeterO wrote:
Thu Jul 11, 2019 8:00 pm
Yeah, but they may have had uneven volt drops when conducting.
PeterO
That wouldn't cause any problem. Compared to the 26 volts from the transformer any difference in diode voltage drop would have insignificant impact on the relative currents through the diodes.
Transformer secondary voltage is nearer 60V peak at full load. The smoothing and regulation is not conventional for a low voltage high current supply by modern standards.
PeterO
So diode voltage mismatch is even more insignificant.

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:44 pm

drgeoff wrote:
Thu Jul 11, 2019 8:35 pm
PeterO wrote:
Thu Jul 11, 2019 8:28 pm
drgeoff wrote:
Thu Jul 11, 2019 8:15 pm

That wouldn't cause any problem. Compared to the 26 volts from the transformer any difference in diode voltage drop would have insignificant impact on the relative currents through the diodes.
Transformer secondary voltage is nearer 60V peak at full load. The smoothing and regulation is not conventional for a low voltage high current supply by modern standards.
PeterO
So diode voltage mismatch is even more insignificant.
The documentation suggests that they were matched pairs. I bought four and chose the two with nearest volt drops.
Regulation is achieved with a "Transductor" in series with the transformer primary,
Anyway it doesn't really matter...
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
rpdom
Posts: 15361
Joined: Sun May 06, 2012 5:17 am
Location: Chelmsford, Essex, UK

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:54 pm

Imperf3kt wrote:
Thu Jul 11, 2019 8:28 pm
I know old computers used a lot of power, but that's a LOT
They certainly did. One I worked on had a couple of ammeters on the wall which used to read around 100A at 208VAC when the computer was running. That was before we added another room full of disk cabinets.

There were a set of motor-generators in the basement to convert from our local 415VAC 50Hz to the 208VAC 60Hz that the computer needed.

Oh, and I helped dismantle another computer later where the 5V power connections were stainless steel bars about 30mm by 15mm section.

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Thu Jul 11, 2019 8:58 pm

We have to be a bit careful which machines we run together as we have caused fuses in the building substation to blow in the past ! I suspect they may have been a bit old but still....
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
Imperf3kt
Posts: 2862
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Old fashioned fault finding....

Thu Jul 11, 2019 9:25 pm

PeterO wrote:
Thu Jul 11, 2019 8:32 pm
Imperf3kt wrote:
Thu Jul 11, 2019 8:28 pm
Correct me if I'm wrong (likely), but 23Ah at 26v is 598watt hours of juice.
Used in "a few seconds"
I know old computers used a lot of power, but that's a LOT
Read the article (especially note Fig. 14). The battery is only "on load" during a mains failure. In normal use the battery charger is providing 26V and 44A , that's the real consumption !
PeterO
So that comes out to 1144W, still a lot of power.
55:55:44:44:4C
52:4C:52:42:41

W. H. Heydt
Posts: 10981
Joined: Fri Mar 09, 2012 7:36 pm
Location: Vallejo, CA (US)

Re: Old fashioned fault finding....

Fri Jul 12, 2019 1:04 am

Imperf3kt wrote:
Thu Jul 11, 2019 9:25 pm
PeterO wrote:
Thu Jul 11, 2019 8:32 pm
Imperf3kt wrote:
Thu Jul 11, 2019 8:28 pm
Correct me if I'm wrong (likely), but 23Ah at 26v is 598watt hours of juice.
Used in "a few seconds"
I know old computers used a lot of power, but that's a LOT
Read the article (especially note Fig. 14). The battery is only "on load" during a mains failure. In normal use the battery charger is providing 26V and 44A , that's the real consumption !
PeterO
So that comes out to 1144W, still a lot of power.
Last I looked you could buy 1500W PSU for desktop PCs. They probably go higher by now.

It all kind of depends on what you're used to.... There was an MG set in the basement of the EE building at UC Berkeley where the DC output side was 1000v at 980A.

User avatar
Imperf3kt
Posts: 2862
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Old fashioned fault finding....

Fri Jul 12, 2019 1:47 am

True, but what does this system do that uses 1200W vs what does a 1500W desktop PC do?
Wikipedia claims it uses 3.5kw.

It's kind of like comparing a scientific calculator that runs on solar, and fits in your pocket, to a simple addition / multiplication calculator that runs on solar and is the size of a bus. One clearly does a lot more than the other in the same power budget.


Ultimately what I'm getting at is; why is such a system still in use when a Raspberry Pi is more than capable of emulating it?
55:55:44:44:4C
52:4C:52:42:41

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Fri Jul 12, 2019 5:27 am

Imperf3kt wrote:
Fri Jul 12, 2019 1:47 am
Ultimately what I'm getting at is; why is such a system still in use when a Raspberry Pi is more than capable of emulating it?
TNMOC.png
TNMOC.png (13.17 KiB) Viewed 1104 times
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

User avatar
Imperf3kt
Posts: 2862
Joined: Tue Jun 20, 2017 12:16 am
Location: Australia

Re: Old fashioned fault finding....

Fri Jul 12, 2019 8:31 am

Wait, it's a museum piece?
What sort of programs do you make for that.
55:55:44:44:4C
52:4C:52:42:41

User avatar
PeterO
Posts: 5066
Joined: Sun Jul 22, 2012 4:14 pm

Re: Old fashioned fault finding....

Fri Jul 12, 2019 8:42 am

Imperf3kt wrote:
Fri Jul 12, 2019 8:31 am
Wait, it's a museum piece?
:roll:
What sort of programs do you make for that.
All sorts... Lately I've been using it to create and draw mazes
https://www.youtube.com/watch?time_cont ... XaSCHJYJDw
PeterO
Discoverer of the PI2 XENON DEATH FLASH!
Interests: C,Python,PIC,Electronics,Ham Radio (G0DZB),1960s British Computers.
"The primary requirement (as we've always seen in your examples) is that the code is readable. " Dougie Lawson

Return to “Off topic discussion”