6502.org Forum  Projects  Code  Documents  Tools  Forum
It is currently Thu Nov 21, 2024 4:54 pm

All times are UTC




Post new topic Reply to topic  [ 32 posts ]  Go to page 1, 2, 3  Next
Author Message
PostPosted: Sun Apr 28, 2024 2:25 pm 
Offline

Joined: Wed Jan 08, 2014 3:31 pm
Posts: 578
I am going to do a completely silly project and build some logic gates using some 5 volt relays. The data sheet says the coil resistance is 130 ohms and 192 mW power. I think this means that roughly 38 mA of current will trigger the coil, which is pretty small.

These relays already have a fly back diode built into their coil, so no need there. But I want to use two diodes as inputs to create an OR gate, and use that to a relay inverter to create a NOR gate. Now if these were transistors I would use switching diodes for this purpose, but relays aren't transistors. However, 38 mA seems small enough that switching diodes would work here too.

Does my reasoning seem sound?

If anyone is curious, I want to build something like a relay counter and make a lot of noise while using the logic outputs to turn on some blinkenlights. Basically a really slow and noisy output device with a 50s retro-future vibe.


Top
 Profile  
Reply with quote  
PostPosted: Sun Apr 28, 2024 3:01 pm 
Offline
User avatar

Joined: Fri Dec 11, 2009 3:50 pm
Posts: 3367
Location: Ontario, Canada
Martin_H wrote:
However, 38 mA seems small enough that switching diodes would work here too.

Does my reasoning seem sound?
What a delightfully silly project! :) And yes, switching diodes should be fine, as they're generally rated for more current than that.

I like the counter idea. However, it might take some monkeying to make it work. To form each flip-flop, I'm imagining you'll have two relays cross-coupled to form a multivibrator, is that right? And there'll be some capacitors whose values need to be experimented with...

-- Jeff

_________________
In 1988 my 65C02 got six new registers and 44 new full-speed instructions!
https://laughtonelectronics.com/Arcana/ ... mmary.html


Top
 Profile  
Reply with quote  
PostPosted: Sun Apr 28, 2024 3:27 pm 
Offline

Joined: Wed Jan 08, 2014 3:31 pm
Posts: 578
Dr Jefyll wrote:
I like the counter idea. However, it might take some monkeying to make it work. To form each flip-flop, I'm imagining you'll have two relays cross-coupled to form a multivibrator, is that right? And there'll be some capacitors whose values need to be experimented with...

Thanks. Yes, some relay flip flops will form the basis for the counter. You're correct that some capacitors might be needed, but the whole thing is going to take some trial and error as relay logic isn't something I have done much of.


Top
 Profile  
Reply with quote  
PostPosted: Sun Apr 28, 2024 5:46 pm 
Offline
User avatar

Joined: Thu May 28, 2009 9:46 pm
Posts: 8504
Location: Midwestern USA
Dr Jefyll wrote:
Martin_H wrote:
I like the counter idea. However, it might take some monkeying to make it work. To form each flip-flop, I'm imagining you'll have two relays cross-coupled to form a multivibrator, is that right?

You guys might find this hard to believe, but I am the inventor of a SR-type relay flip-flop circuit that has a single input.  It consists of two relays, four steering diodes and some resistors.  The circuit is designed so it consistently powers up in a known state, as its original application was in a traffic control system I designed many years ago.  I have the paper copy of the patent application for the traffic control system buried somewhere in my “archives” (US patent 3,893,097, which was issued to yours truly and the guy who was my boss at the time—he knew less about electricity than I know about performing brain surgery :D).  I'll see if I can dig it up and post the flop circuit.

The system in which the flop circuit was used was built with Potter & Brumfield R10 miniature cradle relays, which were available in 2 form C through 8 form C, and were, at the time, the smallest relays to have such a wider range of contact forms.  In those days (early 1970s), use of TTL-style logic circuits in traffic control systems was not yet accepted practice, so everything was electromechanical.  1N400x diodes (usually 1N4002) were used as logic elements where needed.

While the P&B R10 series continues to be available, there are plenty of other choices in small relays, such as from Omron and Panasonic.  Attached are some data sheets for some of my favorites.  Also, Schottky diodes may be used in place of the 1N400x type to get less forward drop and circuit heating.

Attachment:
File comment: Omron G5V1 Low-Signal Relays
g5v1_series.pdf [752.88 KiB]
Downloaded 39 times
Attachment:
File comment: Omron G5V2 Low-Signal Relays
g5v2_series.pdf [1.08 MiB]
Downloaded 37 times
Attachment:
File comment: Panasonic NC Series Relays
NC_series.pdf [323.32 KiB]
Downloaded 33 times
Attachment:
File comment: Potter & Brumfield R10 Relays
relay_r10.pdf [952.6 KiB]
Downloaded 46 times

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!


Top
 Profile  
Reply with quote  
PostPosted: Sun Apr 28, 2024 10:02 pm 
Offline
User avatar

Joined: Thu May 28, 2009 9:46 pm
Posts: 8504
Location: Midwestern USA
BigDumbDinosaur wrote:
I have the paper copy of the patent application for the traffic...I'll see if I can dig it up and post the flop circuit.

I was able to find the original paperwork and did a quickie redraw of the flop circuit, illustrating it with 12 volt-compatible parts.

Attachment:
File comment: SR-Type Relay Flip-Flop
relay_flop.gif
relay_flop.gif [ 38.32 KiB | Viewed 1149 times ]

The original circuit operated on 48 volts DC—there are design advantages to using higher voltages in relay logic.  The diodes that were used were 1N4002 and the relays were P&B R10 units with bifurcated crossbar contacts for maximum reliability in near-dry circuits.  Comparatively speaking, the R10 relays are power hogs, using 900mw at rated voltage.  The Panasonic relays in the above schematic use 360mw at rated voltage, and will pull in on 200 mw.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!


Top
 Profile  
Reply with quote  
PostPosted: Sun Apr 28, 2024 11:48 pm 
Offline

Joined: Wed Jan 08, 2014 3:31 pm
Posts: 578
@BDD, thanks for the schematic and the data sheets.

I contemplated going with some 24 volt relays in a clear housing, which would both make louder switching sounds and be more visible while doing it. But I got a really good deal on some Meder 5VDC DPDT relays. Basically $1 each from Electronics Goldmine which occasionally has some screaming deals. I got some nice gear motors from a sale a while back.

So I bought 30, which should be enough to build various logic gates and then Voltron them together into counters or maybe an adder.

I resisted the temptation to buy the two hundred it would take to build a relay computer. That would be a 1 bit wide serial ALU machine using serial flash and ram. Most of the relay machines online use semiconductor memories as building RAM and addressing decoders using relays is not for the faint of heart.


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 29, 2024 1:29 am 
Offline
User avatar

Joined: Thu May 28, 2009 9:46 pm
Posts: 8504
Location: Midwestern USA
Martin_H wrote:
I resisted the temptation to buy the two hundred it would take to build a relay computer. That would be a 1 bit wide serial ALU machine using serial flash and ram. Most of the relay machines online use semiconductor memories as building RAM and addressing decoders using relays is not for the faint of heart.

I built a relay “computer” while in high school, which included some rotary stepping relays scavenged from old pinball machines.  It didn’t have memory like what we would consider to be memory today, but did maintain state to some degree.  It could add and subtract small numbers, but that was about it.

Building RAM purely with relays would consume a vast number of parts, which is why the early relay computers were so huge and needed government funding to get built—and had so little RAM.  Assuming the use of a flip-flop such as in my previous post, you would need 16 relays, 32 diodes and 16 resistors to build one byte of parallel RAM.  Needless to say, scaling that up to any degree would result in a major expenditure of physical space and money—not to mention the power that would be required to run it.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 29, 2024 5:25 am 
Offline

Joined: Mon Jan 19, 2004 12:49 pm
Posts: 983
Location: Potsdam, DE
Nothing like a little space heating while doing computing work :mrgreen:

Neil


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 29, 2024 6:31 am 
Offline
User avatar

Joined: Thu May 28, 2009 9:46 pm
Posts: 8504
Location: Midwestern USA
barnacle wrote:
Nothing like a little space heating while doing computing work :mrgreen:

The minis I worked on in the 1970s were space heaters...and noisy.  There was a ton of discrete logic in them and in the days before the widespread availability of 74LS parts, the heat just poured out of those contraptions.  :D

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 29, 2024 9:18 am 
Offline

Joined: Wed Jan 08, 2014 3:31 pm
Posts: 578
BigDumbDinosaur wrote:
Building RAM purely with relays would consume a vast number of parts, which is why the early relay computers were so huge and needed government funding to get built—and had so little RAM.  Assuming the use of a flip-flop such as in my previous post, you would need 16 relays, 32 diodes and 16 resistors to build one byte of parallel RAM.  Needless to say, scaling that up to any degree would result in a major expenditure of physical space and money—not to mention the power that would be required to run it.

The first relay computer I saw online was Harry Porter's https://web.cecs.pdx.edu/~harry/Relay/, and he used semiconductor RAM to avoid that explosive parts count. Even with that compromise it is a beast.

The Mercia relay computer https://www.relaiscomputer.nl/index.php uses over 2000 relays and has just 512 bytes of RAM. He also used diode ROMs with in relay decoders for the microcode and ROM. Besides the expense, where do you keep such a beast?


Last edited by Martin_H on Mon Apr 29, 2024 12:47 pm, edited 1 time in total.

Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 29, 2024 10:16 am 
Offline

Joined: Mon Jan 19, 2004 12:49 pm
Posts: 983
Location: Potsdam, DE
Fortunately, I have two rarely used bedrooms and respectable cellar space. Unfortunately, I have been forbidden from putting 'your bloody toys' in them :(

(Also I have to pay the power bill! Though I suspect they might be lower than Usagi Electric's valve systems.)

Neil


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 29, 2024 8:12 pm 
Offline
User avatar

Joined: Thu May 28, 2009 9:46 pm
Posts: 8504
Location: Midwestern USA
barnacle wrote:
Fortunately, I have two rarely used bedrooms and respectable cellar space. Unfortunately, I have been forbidden from putting 'your bloody toys' in them :(

“Bloody toys,” eh?  :lol:  Women can be such downers at times!

Quote:
(Also I have to pay the power bill! Though I suspect they might be lower than Usagi Electric's valve systems.)

When I was in the US Navy and attending electronics school at the Great Lakes Training Center (near Chicago), I had an opportunity to see the school’s ballistic computer, which was a tube (valve) monstrosity.  That contraption used so much power, there was a separate service drop to the building just to run it.  Word had it that when a tube burned out, the only way anyone knew about it was when the thing started producing nonsensical results.  :shock:  The challenge was in finding the burned-out tube...I seem to recall there were thousands of them.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 29, 2024 8:36 pm 
Offline
User avatar

Joined: Thu May 28, 2009 9:46 pm
Posts: 8504
Location: Midwestern USA
Martin_H wrote:
The first relay computer I saw online was Harry Porter's https://web.cecs.pdx.edu/~harry/Relay/, and he used semiconductor RAM to avoid that explosive parts count. Even with that compromise it is a beast.

It does make a mesmerizing sound as it computes.  I can see where one could set that sound to some light rock music.

On close inspection, it appears he used standard “cube” relays, which are relatively slow, as relays go.  If I were ever to become sufficiently deranged to undertake such a project, I would use cradle-style relays, which are three-to-four times faster (and more expensive).  Then instead of setting the sound to light rock, it would be fast jazz...or perhaps “Flight of the Bumblebee.”

Quote:
The Mercia relay computer https://www.relaiscomputer.nl/index.php uses over 2000 relays and has just 512 bytes of RAM. He also used diode ROMs with in relay decoders for the microcode and ROM.

I like how one of his goals is for his computer to be “portable.”  I suppose it would be...if you have a forklift truck to lift it and a railroad car on which to transport it.  :P

Quote:
Besides the expense, where do you keep such a beast?

My house has a full basement.  Unfortunately, about half of it is already spoken for by my wife.  :(  Also, I only have 200 amp electrical service.  Not sure I could run a relay computer from that and have enough left over to power the lights, refrigerator, washing machine, etc.

_________________
x86?  We ain't got no x86.  We don't NEED no stinking x86!


Top
 Profile  
Reply with quote  
PostPosted: Tue Apr 30, 2024 5:10 am 
Offline

Joined: Mon Jan 19, 2004 12:49 pm
Posts: 983
Location: Potsdam, DE
Ah, but the opportunity to make a solar powered relay computer, even if it could only run when the sun shines! At least you could go straight from the 28v or so the panels provide without having to bother with all that mains interfacing stuff :D

My office/workshop lives in the attic at the top of the house, with a small mechanical workshop in the cellar. It works pretty well, apart from the occasional Murphy's law of requiring a tool from one while working in the other...

Neil


Top
 Profile  
Reply with quote  
PostPosted: Tue Apr 30, 2024 4:34 pm 
Offline

Joined: Wed Jan 08, 2014 3:31 pm
Posts: 578
BigDumbDinosaur wrote:
On close inspection, it appears he used standard “cube” relays, which are relatively slow, as relays go.  If I were ever to become sufficiently deranged to undertake such a project, I would use cradle-style relays, which are three-to-four times faster (and more expensive).

Relays seem to have more varieties than transistors. They differ in voltages, number of contacts, AC versus DC, and now they even have different switching speeds!?

What I want are decorative relays with clear cases that make noise as they switch! Plus some blinkenlights.

At the suggestion of Dr. Matt Regan I took a look at Dieter Mueller's relay ALU which is here: http://www.6502.org/users/dieter/a3/a3_4.htm
It is a marvel of economical design.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 32 posts ]  Go to page 1, 2, 3  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 24 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: