Bench Power Supplies.

Centralised area for notes about construction projects. Each project has its own sub-forum. In the sub-forums will be topics relevant to parts of that project (e.g., there might be a topic on motor control). If you start a construction diary, just post in the forum with your project name as the topic, and a moderator will create a sub-forum for your project.

Moderators: Dave Moll, Steve Anderson

Bench Power Supplies.

Postby Steve Anderson » Thu Sep 02, 2021 4:24 pm

Power supplies, more specifically the 'bench' versions. Incredibly useful, abused and taken for granted...until they fail. I bought two 0-15V/3A versions in Hong Kong some 30 years ago, cheap, certainly not precision, but they did the job. I blew one up recently by trying to recharge a car battery. Wrong polarity. Goodbye...

Not worth trying to fix, and one on its own is pretty useless, so it's replacement time. Commercial versions are ridiculously expensive for what they are, so time to make my own. Power supplies are boring, but a necessary part of the whole.

Just like buying anything, what do I need? Those two supplies easily did 95%+ of all I needed for semiconductor work, I rarely (perhaps never) needed 3A, 1A would have been perfectly adequate. So 0 to +15V and 0 to -15V both at 1A would be a good start. Easy...LM317 for the positive, LM337 for the negative, though they start at minimum of +1.2V or -1.2V, that's acceptable.

I also thought it would be useful to add a few fixed voltage outputs too...
+3.3V/1A, Micros, memory, logic etc...
+5.0V/1A, as above...
+12V/1A, mainly analogue stuff...
-12V/1A, as above...
...those in addition to the variables. Not all currents delivered at maximum simultaneously though! All easy with 78xx's/79xx's, except the +3.3V, use another LM317 with two fixed resistors. To limit dissipation for the 3.3V & 5V regulators some power resistors in series with the unregulated input. Though a large heatsink may still be required.

Protection? The regulators have current and thermal limiting but no reverse polarity protection if trying to recharge batteries! But the regulators are cheap and easy to replace if you fry one. Fuses and diodes may help on the outputs...

However, there's one major issue I have with the variables...

PSU 1.gif
PSU 1.gif (23.3 KiB) Viewed 2688 times

A typical arrangement from the datasheet...

What happens if the pot(s) fail? And they will fail, they're electromechanical. The regulators will output the unregulated input supply, less a volt or two. What's the most common failure mode of pots? Open circuit wiper, maybe only over a small part of the track, but the damage is done. Your 5V load circuit is toast!

So maybe not LM317/337's, more research needed. What is required is an arrangement that when a pot fails the supply shuts down.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Bench Power Supplies.

Postby Harry Dalek » Thu Sep 02, 2021 4:56 pm

I would say swap with a multi position switch with fixed resistors but you would get a surge on the switch over if not done correctly ,perhaps a few regulators with fixed resistors then adjustment every five volts with a multi position switch .
How fine a voltage adjustment is needed half the time ? most circuits are 5 or 12 volts , but the danger of using a pot or trimmer with a variable supply regulator i do see your point !
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 4937
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Bench Power Supplies.

Postby Steve Anderson » Fri Sep 03, 2021 1:01 pm

Harry Dalek wrote:How fine a voltage adjustment is needed half the time? Most circuits are 5 or 12 volts...

True, but often I'm tasked with something that needs to work over a range of voltages, say from a battery supply. So the device may be used with a small 'brick'-type 9V battery (UK PP3, US 6LR61), however you want a reasonable life out of the battery. Say a 'give-up' voltage of 6.5V, by then the battery is considered exhausted.

The same applies to rechargeable batteries, the same type of battery can have a fully-charged voltage of over 10V, so that needs assessing too.

So I'd be looking for at least 0.1V increments, less better.

For mains-powered devices I have a Variac, the item might be designed for a nominal 240V but I check that it still meets specifications at 200V (which covers 220V countries) and doesn't catch fire at 260V (e.g. Malaysia can often have a supply of over 250V). Pro-rata 120V countries, especially parts of Japan. Japan is divided into 120V/60Hz and 220V/50Hz regions, though some prefectures are still on 100V!. Though with switched-mode supplies (e.g. a laptop) it doesn't matter, 90-250V, 50/60Hz.

I think I've solved the open-circuit pot problem, now I need to add some current limiting/short-circuit protection. It won't be as simple as the LM317/337 arrangement above, but not overly complex either.

Steve A.

Some battery discharge data follows..

PP3_US_CT.pdf
(83.69 KiB) Downloaded 105 times

PP3_GP17R8H.pdf
(139.99 KiB) Downloaded 116 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Bench Power Supplies.

Postby Harry Dalek » Sat Sep 04, 2021 12:02 am

Steve Anderson wrote:True, but often I'm tasked with something that needs to work over a range of voltages, say from a battery supply. So the device may be used with a small 'brick'-type 9V battery (UK PP3, US 6LR61), however you want a reasonable life out of the battery. Say a 'give-up' voltage of 6.5V, by then the battery is considered exhausted.


Yes some times the idea does come in handy ,i was thinking perhaps a digital variable resistor circuit but even if useable here bit complex ....an idea the might be bit simpler an LDR over the fixed resistor but still have to control the light so an thing happens here in the same sinking boat .


For mains-powered devices I have a Variac, the item might be designed for a nominal 240V but I check that it still meets specifications at 200V (which covers 220V countries) and doesn't catch fire at 260V (e.g. Malaysia can often have a supply of over 250V). Pro-rata 120V countries, especially parts of Japan. Japan is divided into 120V/60Hz and 220V/50Hz regions, though some prefectures are still on 100V!. Though with switched-mode supplies (e.g. a laptop) it doesn't matter, 90-250V, 50/60Hz.

I think I've solved the open-circuit pot problem, now I need to add some current limiting/short-circuit protection. It won't be as simple as the LM317/337 arrangement above, but not overly complex either.

Steve A.

Some battery dis


There is another way that you know about that's simple and safe ! as in if it brakes it lowers the voltage and works with a small transformer with the primaries of the controlling transformer and main transformer in series to 240v may case
3rd last video below , you can control the main transformers AC out put by either as in the video with voltage or resistance on the controlling transformer secondary windings or magnetism shown .
The magnetic amplifier should not be forgotten because its an old idea hell i used the idea twice now and it works well either with control on primary or secondary of a transformer you want to control its AC ...great way to control HV AC.


youtu.be/DBX1-POuJMw
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 4937
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Bench Power Supplies.

Postby Steve Anderson » Sat Sep 04, 2021 12:20 pm

Novel demonstration, but not really applicable to this project. I need the fixed regulated outputs as well as the variable regulated outputs at the same time. In addition the regulators reduce hum/ripple to acceptable levels. I'm planning to use one transformer only as well.

But I can see some applications where that concept might be useful.

I did consider using a digital method, but I felt it was overkill for a power supply...

Steve A.

Added later...I might change my mind on only a single transformer, for the +3.3V and +5.0V outputs there'll be quite a lot of wasted power (heat) to get rid of...also perhaps increase those two outputs to 3A...(LM350s)...

I might also add a +2.5V and a +1.8V output as I've had a few enquires (though nothing definite) about future microcontroller work using them. It's easy to add them now rather than later...they probably would only need to be a few hundred mA though...more LM317s...I'm happy to use 317s, but not with a pot...fixed resistors OK...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Bench Power Supplies.

Postby Steve Anderson » Sun Sep 05, 2021 1:22 pm

Well I think I've cracked the +ve and -ve variable supplies with the exception of adding current limiting. They go to zero output upon pot failure. Nothing exotic, with perhaps the exception of TIP120 and TIP125 Darlington devices, but easy to find, or substitute with something similar.

The rest are all fixed output voltages, so a mixture of 78xx/79xx devices and LM317s for those fixed positive voltages not available in the 78xx/79xx series (1.8V, 2.5V & 3.3V). Though I can get all three of those low voltage regulators from Japan, exactly the same as the 78xx series, but that's no good for you guys...78018, 78025 & 78033s. They don't seem to have reached Western shores yet.

So, on with the current limiting for the variable supplies...and that's proving tricky if you want 'foldback' current limiting on a variable voltage supply, no problem with a fixed output voltage. Maybe I'll just go with the standard current limiting arrangement for the variables...

Keep in mind I'm more interested in protecting the supplies rather than the load...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Bench Power Supplies.

Postby Harry Dalek » Sun Sep 05, 2021 6:25 pm

Arrr thats good do post up when your happy with it ,its something i would add for sure for a bench supply as that's where your so likely to find it fail ..Very good Steve !
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 4937
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Bench Power Supplies.

Postby Steve Anderson » Sun Sep 05, 2021 8:21 pm

Harry Dalek wrote:Arrr thats good do post up when your happy with it ,its something i would add for sure for a bench supply as that's where your so likely to find it fail ..Very good Steve !

Well we've all heard 'scratchy' volume controls, and that's where or dust dirt has got into the pot, and/or the wiper and/or track has simply worn out. One instant open circuit. Disastrous in a variable LM317/337 circuit, and many others...LM350 to name but one...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Bench Power Supplies.

Postby Steve Anderson » Sun Sep 05, 2021 9:31 pm

I've come across a variable PSU circuit on the 'net that looks quite promising (why re-invent the wheel when you don't need to [1]). All discrete, easy parts supply, easy to modify/simplify for my needs. There are (as often) a couple of design flaws, but they're easy to deal with...easy to 'invert' the whole thing for a negative version...more anon...

Steve A.

[1] I'd forgotten how complex the requirements for a 'bench' power supply can be. All kudos to those who do it as part of their job and turn out a decent product...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Bench Power Supplies.

Postby Steve Anderson » Sun Sep 05, 2021 10:07 pm

Here's the original, the included website is worth more than just a quick look...
bench-power-supply-compressed-1200x635.jpg


I'll paste up my initial thoughts/modifications tomorrow...my initial thoughts are, "What if the pot goes open circuit? Are we back to square one?" ...more anon again tomorrow...

Initial thoughts are, if the pot goes o/c it shuts down...but early days...sure, it's a lot more complex than a 317 plus a pot and a few other parts, but it's fail safe. And that I prefer...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Bench Power Supplies.

Postby Harry Dalek » Mon Sep 06, 2021 7:18 pm

Steve Anderson wrote:
Harry Dalek wrote:Well we've all heard 'scratchy' volume controls, and that's where or dust dirt has got into the pot, and/or the wiper and/or track has simply worn out. One instant open circuit. Disastrous in a variable LM317/337 circuit, and many others...LM350 to name but one...

Steve A.


Yes any one with any gear from the 50s will know the pots have had it ,pity we can't do all this with variable capacitor the shaft would have to wear out before it brakes have not come across this problem yet even with age.
Dam that friction on normal pots !
The electromagnetic spectrum has no theoretical limit at either end. If all the mass/energy in the Universe is considered a 'limit', then that would be the only real theoretical limit to the maximum frequency attainable.
User avatar
Harry Dalek
"Fester! Don't do that to 'Thing'"
 
Posts: 4937
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Bench Power Supplies.

Postby Klaas Robers » Tue Sep 07, 2021 2:41 am

But still if the bottom contact on the resistance trace of the pot P1 opens, then you have suddenly the whole output voltage (of that range). I have seen that happening with a pot.

But a bad contact of the wiper is much more likely.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1612
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

Re: Bench Power Supplies.

Postby Steve Anderson » Tue Sep 07, 2021 5:19 pm

Klaas Robers wrote:But a bad contact of the wiper is much more likely.

Very true, it's rare for the track itself to fail. Here I intend to use 'conductive plastic' pots, very little wear and fully enclosed. More expensive by a big margin than carbon types, but here it's worth it...even though the supply defaults to zero output if the wiper fails...

Choose a reputable supplier, Bourns in the example below...standard 1/4" or 6.35mm shaft...you can see they're smaller than the usual carbon types...a bonus if you're pushed for space...though not in this case...

For presets I prefer the enclosed versions rather than the cheaper 'open frame' style, in fact I abhor the 'open frame' ones, as reliable as the British summer. Even though they're inside the gear, the enclosed ones are peace of mind, and preferably they have Cermet tracks rather than carbon.

Steve A.

Bourns pot.jpg
Bourns pot.jpg (6.02 KiB) Viewed 2516 times
open-frame-presets-DO NOT USE.jpg
open-frame-presets-DO NOT USE.jpg (32.21 KiB) Viewed 2512 times
Attachments
Bourns preset.jpg
Bourns preset.jpg (14.47 KiB) Viewed 2512 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Bench Power Supplies.

Postby acl » Tue Sep 07, 2021 10:05 pm

Steve,

This is the answer to all my bench power supply needs. A cheap and cheerful module driven from an old PC power supply.

https://www.youtube.com/watch?v=IJFBvMMltQw

IMG_3454.JPG


Regards Chris Lewis
acl
Just nod and pretend you understand me
 
Posts: 289
Joined: Thu Mar 08, 2007 7:38 am

Re: Bench Power Supplies.

Postby Steve Anderson » Wed Sep 08, 2021 4:32 pm

Yes, that is an option, and I did consider doing just that as I have 5-6 laptop supplies left over from a previous (non-NBTV/SSTV) 'real-work' project. (18.5V/3.5A). Use one for the positive, one for the negative. Though I would have to check the output is totally isolated from the input ground, but that can be fixed easily if needed.

So, why don't I plan to use them? A large proportion of my work is audio-centric, so I'm reluctant to use any form of switched-mode supply when you're dealing with things such as microphone pre-amplifiers. However, for most this idea should work happily and certainly for logic devices it makes total sense.

Maybe I'll build this with the usual linear transformer/bridge/cap/regulator arrangement, then replace that lot with one or two of these 'bricks' and see what I end up with. Switched-mode supplies have improved over the years, better regulation, better noise filtering and so on. So maybe I'm doing them a disservice...

In addition, as the video mentions, they're already certified for safety, which is why I used them in the first place...though check that the certification isn't false/fake...not unusual for stuff from a certain country not far from where I am...though if you buy them in the west you're unlikely to get ripped-off...

Steve A.

A quick check, many suppliers in the UK, other countries much the same, so it shouldn't be a problem...and of course the 120V/240V issue goes away...

As a footnote, a client needs 1.8V supply micro-based system, contract completed tomorrow (I hope), glad I included 1.8V into this...though probably only 200-250mA as a guess...maybe less...these lower voltages (1.8V/2.5V) have been on the horizon for a while now...here we go...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 4888
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Next

Return to Construction Diaries

Who is online

Users browsing this forum: No registered users and 1 guest

cron