Power Supplies - Of The Bench Variety...

Anything not specifically related to NBTV, but at least of some technical nature that might be of interest to NBTV members. Items for sale and links to retailers do not belong here.

Moderators: Dave Moll, Steve Anderson

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Thu Oct 06, 2022 12:14 pm

Harry Dalek wrote:Yes the pot part is a bit of problem...

I haven't studied the internals of the LM317/337 but I'm sure with some relatively simple rearrangement of the die it could be made such that if the pot goes open-circuit it shuts down - or at least resorts to the minimum output (1.25V) rather than almost the whole of the unregulated input. It's most unlikely that a re-design will ever be done as the current version is successful as is.

This flaw isn't the fault of the LM317/337 design but the inherent unreliability of the usually available (cheap) pots. A 'crackly' volume control is an annoyance, here it could spell disaster for attached circuitry. Sure, you can buy better quality pots, but the inherent weakness remains.

I think most would prefer an arrangement where you can slowly increase the voltage while keeping an eye on the current. Any faults, shorts, errors, whatever, should be obvious before frying any downstream components....especially if expensive.

This bench power supply does have a few fixed output voltages provided by the usual 78xx/79xx regulators in addition to these variable outputs. Once you're sure there are no major faults your creation can be transferred to those. - +3.3V, +5V, +12V, and -12V.

+/-15V is rarely used except in quality audio designs where the highest dynamic range is desired, sometimes +/-24V, usually a discrete design with no op-amps, or using specialist devices rated for that sort of voltage are specified...rare/expensive. (D. Self, a prolific audio designer in the 60s/70s, often used +/-24V supplies in his discrete designs. Much of his work was published in the UK 'Wireless World' magazine of that era.) Today, with the exception of the output voltage capability, there are op-amps that better those and similar older designs....there has been 50 years of development since then...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Tue Oct 11, 2022 4:43 pm

Re-reading Rod Elliot's remarks referred to earlier - he's right, it's not easy to design a blow-up proof bench power supply. I've been grappling with this on and off for a few days.

You simply can't cater for every scenario of potential disasters that may occur. But assuming it uses easily available and inexpensive components the fix and repair shouldn't be too onerous. That is one, perhaps two, of the challenges. There's always the human factors to consider, impatience and not thinking through what you are doing, and not checking the attached prototype or lash-up. The not-so-comforting response is, "Get over it. It happens." Not what you want to hear or think at the time.

Quoting Rod,..."It's not an accident that there aren't that many DIY projects for bench power supplies. Most people come to the realisation fairly quickly that it's a very expensive exercise, and that getting a fully working, reliable supply that does exactly what you need is not a trivial undertaking. "

I concur, so this is still on-going....though I wouldn't call it 'very expensive' considering how much use you should get out of it, but it's not insignificant....

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Wed Oct 12, 2022 8:32 pm

Some changes, using MOSFETs instead of Darling transistors as the series-pass elements, it does make some things simpler, though one area of the design is challenging in how to do it...more anon....

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Fri Oct 14, 2022 7:59 pm

I've got the calibration documentation for model of Fluke mulimeter I'm interested in, including the password. Now that reference source of volts/current/ohms needs calibration somehow before I can calibrate the meter itself.

If (let's say) the meter needs calibrating once a year, I don't know, but if you don't do it, the meter simply shuts down and becomes useless. Cars are already becoming like this if servicing via the dealer isn't done. They won't clear the flashing/beeping alarm unless they do the service and reset the car's computer. i.e. to clear the warning they must do it even if you have done the service/inspection already.

Today a voltage/current/resistance source with 4-digits of accuracy isn't impossible to DIY. Could be useful for other gear too, e.g. 'scopes.

More to come on this insidious 'maximizing the profits out of the consumer' trend.

Steve A.

Soon I can see your domestic refrigerator requiring temperature calibration every so often otherwise it shuts down. (Product Liability Laws - you ate some thawed chicken, because you didn't have your refrigerator serviced by us when due - it did warn you!)....especially with this IOT crap...

...as so often, getting off topic...
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Sat Oct 22, 2022 12:58 pm

This bench power supply thing is getting closer in design. I now know exactly what Rod Elliot meant by "It's not as easy or as simple as you might think." It certainly isn't! There are many conflicting requirements and compromises to be made along the way. All the while keeping it reasonably simple and bullet-proof at the same time.

I hope to able to throw up a circuit diagram in a day or two...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Sun Oct 30, 2022 7:21 pm

Well, a day or two has gone....this is not an easy exercise!! Revised output arrangement as things are....note the revision number, many others have come and gone!

In addition the 'front end' has also undergone a vast change, though thus far unpublished...

Steve A.

More (for sure) to follow...

DC PSU 5.gif
DC PSU 5.gif (19.88 KiB) Viewed 5144 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Mon Oct 31, 2022 11:18 am

Part of the reason for the slow progress on this was/is trying to develop a discrete op-amp to get around the 30/36V limit imposed by most IC op-amps. It can be got around as shown above, but it's not ideal. I think I've succeeded, but it's early days...naturally it does make things more complex, but if you need an op-amp that can run from +/-30V (60V total), there's not much one can do.

Plans are to make it use +/-60V (120V total).

The origin of the design is here:-

https://sound-au.com/project231.htm

Therefore I do not claim originality for the concept...

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Mon Oct 31, 2022 4:30 pm

As this uses PWM to generate the variable DC output voltages, they need filtering to remove the PWM. The usual RC filters maybe aided by op-amps can be used, but here's a new (to me) and simple way that really does work...two resistors, two capacitors, an inverter and you're done....again I cannot claim originality for the concept...

https://www.edn.com/cancel-pwm-dac-ripp ... btraction/

Here's the practical implementation for this project...the two PWM signals are generated via the two PWM modules within a PIC microcontroller...as they are 10-bit PWM modules the output can be varied in 15mV steps from 0 to 15V....the output stage(s) have a gain of three...

Steve A.
Attachments
PSU PWM Filtering 1.gif
PSU PWM Filtering 1.gif (4.95 KiB) Viewed 5123 times
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Harry Dalek » Thu Nov 03, 2022 8:13 pm

Still watching Steve ,i can see your still putting it together on Paper or I mean PC hahahaha so far .
So Using PWM to control the voltage ..so i suppose your looking into what happens if you some how loose this ? or faulty glitchy signal to it .
What's the range out expected ? i can see above posts you want to improve top end in time .
Just see how it go's one of those things you need to build it and see if it behaves /
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Fri Nov 04, 2022 1:38 pm

When considering to build rather than buy a bench power supply I eventually drew up the following specification...

Variable output 1, 0 to +15V with 400mA or 60mA current limiting.
Variable output 2, 0 to -15V with 400mA or 60mA current limiting. Optional tracking of positive supply.
Fixed +5V, 1/1.5A.
Fixed +3.3V, 1/1.5A.
Fixed -5V, 1/1.5A.
I may add fixed +/-12V @ 1/1.5A too...maybe.

Remote control of output voltage on the variable supplies and on/off of each supply independently. This will allow some automation in testing and results recording...mainly for 'real work' rather than NBTV and the like.

As it's transpired the +3.3V and +5V outputs are 3A each, using a pre-built dual regulator board as shown previously for those two voltages. (My posting of Sep.18th in this thread).

Many might think that it's a bit 'lightweight' on current, but over many years of building stuff it's very rare that I've ever needed anything more. If you're planning building something which needs more volts or amps you're going to have to build a power supply for it anyway, so you might as well start there. Generally I use a bench power supply for breadboarding smaller chunks of circuits rather than complete projects. Testing 'proof of concept' in new designs one could say.

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Harry Dalek » Fri Nov 04, 2022 6:07 pm

Thanks Steve i was not sure on the range of the supply .
I suppose you make it to your wants needs.
Every ones different on what they want ,it can get out of hand trying to make some thing that can to do every thing low voltage projects HV for Valve work CRTs so on .
I think its best make it as you are thinking for you want and make some thing else for the other needs.
Again see how yours evolves ,are you thinking panel meters or just get a working board up and no casing proof of concept ?
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Sat Nov 05, 2022 12:27 pm

It'll be cased and I hope 'reasonably presentable'. Metering though is a whole different issue which I haven't spent much thought on as yet. To have enough individual front-panel meters to cover all those different supplies would be ridiculous, it would make the thing large and expensive at around US$10 each per cheap analogue meter. I may have one which is switchable between different supplies, but most of the information will probably come out as a serial data stream which can be displayed and/or recorded on a PC - part of the automation I mentioned previously. This I need to spend some time on, as ever, more to come....

As for higher voltages for tubes/valves or even CRTs, I don't really see the need for a 'bench-style' power supply as the requirements are so many and varied. Each variant is is so different to what went before and probably what will follow. They were made in the past but unless you tend to specialise in one 'theme', say domestic radios or audio hi-fi amplifiers, and all-encompassing tube/valve bench supply is a complex, demanding and expensive project. Not one to be undertaken lightly...

As much as I enjoy tube/valve stuff it's rare that I work with them these days. The last project I built using glassware was a Nixie clock driven from the GPS time data, but aside from the Nixies all the rest of it were semiconductors....and that was probably 20 years ago! It included a PIC microcontroller which added 0700 hours to the UTC/GPS time-stamp and also corrected the roll-over after midnight for the time and date. I'm approximately 100 degrees east on ICT (Indo-China time, no daylight saving).

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Harry Dalek » Sat Nov 05, 2022 8:38 pm

I got 2 panel meters from Ali express some time back yet to use some are rather cheap easy fix .
https://www.aliexpress.com/af/panel-met ... 1105013409
Attachments
Screen 00017.bmp
Screen 00017.bmp (408.8 KiB) Viewed 5025 times
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: 5364
Joined: Fri Sep 26, 2008 4:58 pm
Location: Australia

Re: Power Supplies - Of The Bench Variety...

Postby Steve Anderson » Sat Nov 05, 2022 8:49 pm

I'm not keen on digital meters on power supplies, much preferring the older style analogue versions. here's Rod Elliot's take on the same subject:-

Pretty much the only meter I'd consider for a power supply is analogue, particularly for current. They don't have the apparent accuracy of digital panel meters (DPMs), but they are both immediate and intuitive. Just like having to press buttons to select voltage and/ or current is unacceptable for a bench supply, so are meters that you have to read, but can't do so if they're changing rapidly (especially current). Voltmeters can be digital, since they show the preset voltage, and this usually doesn't change other than under the user's control, or if the optional current limiter becomes active.

If you're using the meters for monitoring in this situation the analogue versions are vastly more human-friendly....

Plus many digital versions cannot measure negative voltages/currents relative to their own supplies, i.e. in some circumstances they are not easy to use or simply impossible...generally they're Micky Mouse devices, they look pretty, but otherwise useless...and their supply range is limited and draw power. An analogue meter doesn't, or at worse a few mV, or mW.

Don't get me wrong, I would never go back to an analogue multimeter, but with the digital versions you're aware of their shortcomings in the back of your mind when using them...

...the old adage, "Just because it's digital doesn't mean it's necessarily better".

Steve A.
User avatar
Steve Anderson
"Fester! Don't do that to 'Thing'"
 
Posts: 5360
Joined: Fri Mar 30, 2007 10:54 pm
Location: Bangkok, Thailand

Re: Power Supplies - Of The Bench Variety...

Postby Klaas Robers » Sun Nov 06, 2022 5:26 am

I agree with that Steve. Digital meters give more (digits) information than you need. Supply voltages never have to be precise better than +/- 0.5 volt. If your application doesn't work at that voltage, something else is faulty.

And indeed, current may vary during operation. A digital meter cannot follow that and in the mean time you can't read what it indicates. Analogue meters can't follow it as well, but at least they give you some "about" indication.

A practical problem might be that it becomes more and more difficult to obtain analogue meters. We are dependant on what the industry needs. If they don't need analogue meters, there are very soon no more analgue meters for us. I experienced that with varable (tuning) capacitors. No more available. Same reason.
User avatar
Klaas Robers
"Gomez!", "Oh Morticia."
 
Posts: 1656
Joined: Wed Jan 24, 2007 8:42 pm
Location: Valkenswaard, the Netherlands

PreviousNext

Return to Off Topic

Who is online

Users browsing this forum: No registered users and 13 guests

cron