UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Reply
 
Thread Tools
Old 21st Dec 2018, 10:11 am   #1
M3VUV51
Pentode
 
Join Date: Feb 2018
Location: Horncastle, Lincolnshire, UK.
Posts: 131
Default Powering op amps?

Hi all.

I'm after using an op amp as a current sense amp in a PSU, 30V max output. The op amps I have are JRC4558 IC's, the max supply voltage is 18V. i wanted to use them from the same power rail as the PSU, obviously I can't do that. Now the question is can I power it from say an LM7824 regulator? The differential voltage is 30V max ,can the voltage at the current sense resistor be higher than the op amp supply voltage? Hope that makes sense.

Cheers M3VUV.
M3VUV51 is offline   Reply With Quote
Old 21st Dec 2018, 11:04 am   #2
MrBungle
Nonode
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 2,689
Default Re: Powering op amps?

Opamp power supply rejection is pretty good so it’s probably good enough to power the opamp from a well decoupled Zener/resistor combo. You can drop hundreds of volts across a resistor here.

However this brings in other concerns like startup and turn off transients on the output.

That’s why you see a lot of linear power supplies using separate taps for the reference and startup circuits.

Power supplies are way harder than they look. Which is why I buy ‘em
MrBungle is offline   Reply With Quote
Old 21st Dec 2018, 11:42 am   #3
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 12,639
Default Re: Powering op amps?

For monitoring a current sense resistor up at a voltage outside the rails of an opamp you can "pot down" the voltage at either end of the resistor using matched pairs of say 10k and 1k resistors (11:1 stepdown) to get the absolute voltages at the opamp inputs safely within its common mode operating range. You can then "put back" the lost voltage with suitable gain in the opamp if required.

There are specific types of opamp for precisely this application, but they will likely be more precise and more expensive than needed for a one off design which can be tweaked to be "good enough".
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is online now   Reply With Quote
Old 21st Dec 2018, 1:55 pm   #4
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 11,178
Default Re: Powering op amps?

When you use the potting down trick you need to be sure to use very well matched resistors and an opamp with low offset voltage and current. The potting down factor worsens the significance of these parameters. Also, as you want to avoid much voltage drop in the main output, you'll use a low voltage current sensing resistor with only a small voltage drop at full current. This makes the opamp offset and resistor balancing act more difficult.

Can you run the inputs outside the supply voltages that you apply... well, actually you can with some, but there are conditions and gotchas you have to follow. What you can do is spelled out in the data sheet. You need to read it because they're all different and some have limited voltage ranges you can take the inputs over, some rather less than your supply voltages. If you're unfamiliar with opampery and the interpretation of their datasheets, Horowitz and Hill is a friendly and helpful book. THey also have tables of the main standard opamps.

To be honest, for the price of an opamp, it's better to use something that's mainstream rather than just something you happen to have. It can be a case of setting yourself up for trouble if you ever have to get a replacement in the future..

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 21st Dec 2018, 2:19 pm   #5
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,139
Default Re: Powering op amps?

Quote:
Originally Posted by MrBungle View Post
Power supplies are way harder than they look. Which is why I buy ‘em
Yes, me too, but even the ones you buy may be "less than perfect" .

B
__________________
We know there are known unknowns. But there are unknown unknowns – ones we don't know we don't know.
Bazz4CQJ is offline   Reply With Quote
Old 21st Dec 2018, 2:52 pm   #6
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 11,178
Default Re: Powering op amps?

Quote:
Originally Posted by Bazz4CQJ View Post
Yes, me too, but even the ones you buy may be "less than perfect" .
B
"If tha wants a job done reight, tha'as to do it thissen"

It takes the same amount of knowledge to look at an existing/purchased design and see if it's good as it takes to design a good one yourself.

There are some quite decent power supplies out there that qualify as vintage in their own right, surplus lab and industrial units. They may need some general fixing - mostly chasing dried out capacitors. There's also plenty of nasty units whose selling point was low price.

It's rewarding to build your electronics skills up to the level of understanding these things. You can pick good 'uns, you can fix them and you can design your own if you want to. What's more, the knowledge gained can be applied to lots of other types of circuitry. There's work involved and having to learn to visualise some new concepts, but it's do-able and worth doing. Once you've climbed onto the plateau, it then seems easy.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 21st Dec 2018, 2:54 pm   #7
MrBungle
Nonode
 
MrBungle's Avatar
 
Join Date: Jun 2016
Location: London, UK.
Posts: 2,689
Default Re: Powering op amps?

One buys old HP analogue supplies if they want a power supply that works

One studies HP if they want to know how to make a power supply that works

My favourites are the 6236B/6237B units for daily use.
MrBungle is offline   Reply With Quote
Old 21st Dec 2018, 3:03 pm   #8
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 12,639
Default Re: Powering op amps?

The JRC4558 is both mainstream and old school. It's effectively a dual 741. It's got the same +/- 18V supply ratings as most opamps of its time. It's decidedly nothing special nowadays in its capabilities but may well be adequate for the job with the one-off tweaking I mentioned. For not many pennies though, you could buy something with orders of magnitude better offset, CMRR, drift etc that will fit the same socket.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is online now   Reply With Quote
Old 21st Dec 2018, 3:17 pm   #9
Bazz4CQJ
Nonode
 
Bazz4CQJ's Avatar
 
Join Date: Apr 2009
Location: Oxfordshire, UK.
Posts: 2,139
Default Re: Powering op amps?

Quote:
Originally Posted by Radio Wrangler View Post
"If tha wants a job done reight, tha'as to do it thissen"
I really ought to restore the 'high-end' Kingshill PSU that I rescued from a skip. It works but was configured just to supply 25V at 25A. I think it has 12 2N3055's (originals) passively cooled, with an enormous multi-tapped transformer made in the UK by Albion. 19 inch rack thing, about 10 inch high, and therein lies the problem - I literally cannot lift it

B
__________________
We know there are known unknowns. But there are unknown unknowns – ones we don't know we don't know.
Bazz4CQJ is offline   Reply With Quote
Old 21st Dec 2018, 3:51 pm   #10
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 3,441
Default Re: Powering op amps?

In a variable PSU, it's normally much easier to measure the current in the OV line rather than the output line. I did once try to measure +ve current using an op-amp as a unity gain differential amplifier, but it needed tweaking for good CMRR and just "felt" like a poor solution.

Of course, we have to ask how good does it actually have to be?

In my case, it was for reading current on a 1.999 digital display, so 1mA resolution. That's fairly pedestrian in all honesty, but still I thought that the differential amplifier was not the right solution. Some power supplies only have 10mA resolution. Others are much better.

If this same op-amp is also being used to complete the loop when in constant-current mode, then poor CMRR will unfortunately reduce the output impedance.

This issue is one of many reasons why the "floating" topology was so popular back in the day - the I-sense resistor moved up and down with the rest of the control circuitry. It's perhaps less common now because power supplies seem to need computers and IP addresses rather than switches and knobs to sell today

If you don't have the option to rearrange the circuit to measure at ground, then a separate floating supply for the current metering (I assume metering is the main goal?) is actually quite elegant.

It's not fully documented by any means, but here's the PSU I referred to earlier, with the I-sense resistors moved to the ground rail. It's a dual PSU, and it certainly caused a bit of head-scratching at the time (I did the design in my spare time at uni), but it still works well 25 years later.

http://www.markhennessy.co.uk/dualpsu/dualpsu2.htm
mhennessy is offline   Reply With Quote
Old 21st Dec 2018, 4:18 pm   #11
kellys_eye
Heptode
 
Join Date: Jan 2014
Location: Oban, Scotland, UK.
Posts: 619
Default Re: Powering op amps?

What is the general opinion on PSU's built using the ubiquitous 723 regulator and series pass transistors?
kellys_eye is online now   Reply With Quote
Old 21st Dec 2018, 4:50 pm   #12
merlinmaxwell
Dekatron
 
merlinmaxwell's Avatar
 
Join Date: Jul 2008
Location: Maidenhead, Berkshire, UK.
Posts: 7,991
Default Re: Powering op amps?

There are lots of "high side" current monitor chips available https://www.maximintegrated.com/en/a...dex.mvp/id/746 for example.
__________________
Cats have staff, it's dogs that have owners.
merlinmaxwell is offline   Reply With Quote
Old 21st Dec 2018, 5:16 pm   #13
evingar
Octode
 
evingar's Avatar
 
Join Date: Oct 2003
Location: Newbury, Berkshire, UK.
Posts: 1,572
Default Re: Powering op amps?

Quote:
Originally Posted by MrBungle View Post
Opamp power supply rejection is pretty good so it’s probably good enough to power the opamp from a well decoupled Zener/resistor combo. You can drop hundreds of volts across a resistor here.

However this brings in other concerns like startup and turn off transients on the output.

That’s why you see a lot of linear power supplies using separate taps for the reference and startup circuits.

Power supplies are way harder than they look. Which is why I buy ‘em

I work with the things, both at "The Day Job" and at a part time repair operation.
I would say you are correct. It's relatively easy to get something working, less easy to properly engineer something that will work reliably long term, especially with switched mode stuff working from Line voltages.

Unfortunately, even with commercial stuff, I see underrated resistors (especially WRT voltage rating), electrolytic caps mounted immediately adjacent to heat sinks, 85 deg type caps fitted in units with poor cooling, poor Line transient suppression, badly designed snubbers (etc etc).

PSU's seem to be the poor relation in a lot of people's mind, and corners are cut accordingly !
__________________
Chris
evingar is offline   Reply With Quote
Old 21st Dec 2018, 7:53 pm   #14
mhennessy
Dekatron
 
Join Date: Jul 2007
Location: Evesham, Worcestershire, UK.
Posts: 3,441
Default Re: Powering op amps?

Parallel threads running on EEVBlog: https://www.eevblog.com/forum/projec...nse-amplifier/ and https://www.eevblog.com/forum/beginn...ering-op-amps/

There's slightly more information over there - looks like this is using a 723. No mention of metering. Which begs the question: why not use the internal current limit transistor? Is it because you're looking for more precision with the current limit? Perhaps variable, like on many bench power supplies? A bit more information would help enormously.

For starters, what is the intended application? Is it a fixed supply, or a variable supply (perhaps a bench PSU?). If so, what's the minimum voltage? And why do you think the 4558 needs powering separately, given that you can run it from 44V, which is (hopefully) more than the unregulated input to your 30V regulator? Perhaps because you need to go down to 0V? If you do need to adjust down to 0V, then you'll almost certainly need a negative supply rail - easy to derive from the AC from the transformer with a couple of diodes and caps - no need for a separate winding because of the low current requirement. And if so, why not do low-side sensing like I suggested? It will make things much easier.

Regarding a dedicated high-side IC, bear in mind that they are a specialist part that carries risk for something that is going into production. For small quantities or a limited run, then maybe, but Maxim are notorious for discontinuing really rather excellent ICs - that makes me nervous about using them for anything other than personal projects (and keep a few spares!). There's a lot to be said for using "jellybean" parts, like standard op-amps and transistors that have many other alternative parts.
mhennessy is offline   Reply With Quote
Old 21st Dec 2018, 8:13 pm   #15
McMurdo
Dekatron
 
McMurdo's Avatar
 
Join Date: Jun 2009
Location: Staffordshire Moorlands, UK.
Posts: 3,307
Default Re: Powering op amps?

can you work-in a small dc-dc converter for the ic? Might that be an option?
__________________
Kevin
McMurdo is offline   Reply With Quote
Old 21st Dec 2018, 9:21 pm   #16
Radio Wrangler
Dekatron
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 11,178
Default Re: Powering op amps?

Yes, that's JRC's answer to the MC1458, essentially a dual 741. My advice about picking mainstream parts was meant generally. With a standard footprint and jellybean performance it's safe enough.

Maxim make some rather good ICs and are one of the first into the market with new things. Unfortunately they're also the first out. I have spent too much of my time redesigning stuff to get around their obsoletion notifications... Avoid like the Plague. Mark's obviously had the same experience.

Zetex (now Diodes inc.) make high side current sensors as well.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 21st Dec 2018, 10:40 pm   #17
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 6,708
Default Re: Powering op amps?

Quote:
Originally Posted by kellys_eye View Post
What is the general opinion on PSU's built using the ubiquitous 723 regulator and series pass transistors?
They work, but are antiquated and involve vastly more discrete components (to deliver an indifferent performance) than is achievable these days at much lower parts-cost with current circuitry.

Even the semi-obsolete LM317 is better than the 723. This century we've all switched to LDO regulators to avoid the need for the input to be more than a Volt above the intended regulated supply.

See the Texas Instruments design-notes;

www.ti.com/lit/ml/slup239/slup239.pdf
G6Tanuki is offline   Reply With Quote
Old 22nd Dec 2018, 2:27 am   #18
M3VUV51
Pentode
 
Join Date: Feb 2018
Location: Horncastle, Lincolnshire, UK.
Posts: 131
Default Re: Powering op amps?

ive some hall sensors from an old ebike motor,what would be the best way to use one of those,would it be to put inside a loop of thick wire in series with the psu output?,cheers m3vuv.
M3VUV51 is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 12:03 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Copyright ©2002 - 2018, Paul Stenning.