Suitable Replacement for a µA723 IC

[Snarf81]

Hi all,
I need to get a replacement voltage regulator IC for one of my Philips N1700 vcrs and according to the service manual the part number is uA723CA.
I can't find the exact replacement but have seen some advertised for sale on eBay with LM723CN DIP-14 . My question is, is that suitable? It might be obvious to most people but not to me.

I've included a picture of the service manual page with the IC101 listed.

Many thanks
Rich.

[TonyDuell]

Is the original IC in a 14 pin DIP package (looks like most ICs) or a 10 lead round metal can? The '723' comes in both packages.

But other than that one '723' is much the same as another and there should be no problems in replacing it.

[Snarf81]

Hi Tony,
Thanks for the reply.

The original is a 14pin DIP. Ah right, so any 723 DIP 14pin will do? Phew.

[turretslug]

40-50 years ago (eek!) it was one of the go-to ICs for setting at the heart of a high-spec linear PSU, so all sorts of additional letters appeared regarding operating temperature range, package material, reference tolerance and so on but unless you intended to go to war or the Moon with it, there won't have been many considerations beyond price. I expect a bit of searching would throw up any fundamental differences that would be a problem.

[Scimitar]

A 723 is an op-amp, not a regulator.

[Maarten]

While it contains an opamp (error amplifier), it certainly is a regulator.

https://www.ti.com/lit/ds/symlink/lm723.pdf

[factory]

The data-sheet disagrees, definitely a voltage regulator IC;
https://docs.rs-online.com/b73a/0900766b80ea1a07.pdf

David

[TonyDuell]

I normally describe the 723 as a 'power supply kit' . It contains the error amplifier (essentially an op-amp) with a driver transistor on the output and also the voltage reference. It can be used to make both linear and switching regulators.

As others have said the prefix and suffix letters give the manufacturer, temperature range, package (plastic or ceramic DIP), etc. None of which matter for this application.

[Radio Wrangler]

The number 723 was in a series used by the semiconductor division of Fairchild Camera and Instrument. Other well known ones were the 741 and 725 opamps, but the series contained all sorts of analogue ICs. The 723 became themost successful voltage regulator controller chip of its day, and the 741 the most successful opamp.

Fairchild used the uA prefix on its ICs.

In that era, customers, especially those with government contracts wanted multiple sources of parts so that supplies would be secure, so semiconductor firms were encouraged to licence their parts out to rival companies even to the point of providing copies of the masks. This is how a design goof in the 555 timer chip appears in all makes of it!

So you'll find the 723 with national semiconductor's LM prefix as LM723 and a variety of other makers identified by their own house prefixes.

Multi-sourcing continued even without the impetus from the military. If you designed a really good IC and sat on the design, the others would design their own parts, not necessarily compatible so the customers would feel insecure, and you'd have competition. If you licenced it, then you got income from the licence even though you didn't get the full production flow.

The 723 was a very successful generic IC.

It could be used as a low current regulator on its own, but usually it appeared with a supporting cast of driver and power discrete transistors for higher current use.

Applications note circuits abounded. Ways to boost the current, to control higher voltages than its ratings would allow directly, adding current limits and so on, and so on. you could fill a book.

It could even be made to sel-oscillate as a sort of switch mode regulator (best avoided!)

The big gotcha with the 723 is that it's easily upset by the presence of RF power. So, if you use one in say an amateur radio station power supply, then take care over filtering and decoupling. The data sheets don't make this clear, but people soon found out.

All the makers tried to carve out a niche for their own improved variants... more accuracy, more voltage rating, different packages and this is where the suffix comes in.

Fairchild semi got bought by Schlumberger, sold on a few years later and I think they're currently owned bu On Semi which used to be part of Motorola. When the music stops....

David

[Electronpusher0]

Originally Posted by Radio Wrangler

This is how a design goof in the 555 timer chip appears in all makes of it!

You have got me intrigued, I have tried googling this and can find no mention of it, can you explain please?

Peter

[Radio Wrangler]

The 555 goof?

In the original bipolar version, the output stage goes unstable during transitions. So if you set it up to make or extend pulses, on one edge you can actually get a burst of short, fast, pulses before it stays at the new level.

Not a problem with analogue sorts of applications, but if you have it driving some sort of counter, you can get a lot more counts than you expected. Temperature dependent, too

Other makes of the 555 are exact copies right down to this behaviour

CMOS variants are OK.

David

[Electronpusher0]

Thanks, that explains it. The principle use I had for 555s was to hold the reset line low for a short duration on microprocessor circuits during power up.

Peter

[Red to black]

The original 555 also 'crowbarred' the supply lines briefly on transition of state changes at the output, this didn't happen with the Cmos 7555 series, the original 555 needed local de-coupling close to the supply pins IIRC, no evidence to support this to hand, just remember reading it somewhere.

[cmjones01]

Originally Posted by Radio Wrangler

It could even be made to sel-oscillate as a sort of switch mode regulator (best avoided!)

I think one of the popular 1980s home computers (the Commodore 64, I think, but I may be misremembering) actually uses a 723 like this!

Chris

[cmjones01]

Originally Posted by cmjones01

I think one of the popular 1980s home computers (the Commodore 64, I think, but I may be misremembering) actually uses a 723 like this!

I was wrong - it's the TI99.

Chris

[kalee20]

Originally Posted by Radio Wrangler

The 723 was a very successful generic IC.

The big gotcha with the 723 is that it's easily upset by the presence of RF power. So, if you use one in say an amateur radio station power supply, then take care over filtering and decoupling.

The other big gotcha, is that the current-limit function is rubbish! An internal transistor, to be connected across a current-sense resistor, that diverts the error-amplifier output, is all you get.

Works well enough as a current limit, as long as you keep the '723 in a temperature-controlled environment. If you don't, then limiting current is highly temperature dependent. Though, at least it varies the 'right' way, less current available at higher temperatures, which the sales guys would no doubt sell as a feature!

[Radio Wrangler]

Shorting power supply rail to ground in a 555 is real, it's part of the unwanted oscillatory action.

Oh, yes, I agree about the 723 current limit. The threshold is rough but it works. It's OK for having a rough current limit on a power supply which is never needed to do cunstant current modes. So OK for a 138v supply for radio gear, but useless for a lab type supply.

David

[TonyDuell]

Originally Posted by cmjones01

I think one of the popular 1980s home computers (the Commodore 64, I think, but I may be misremembering) actually uses a 723 like this!

Chris

So do a lot of older DEC power supplies. Some of my PDP11s use the 723 as a switching regulator, for example. Along with the HP9800 series of calculators/computers. And a lot more that I've not bothered to check.

I am not sure why it's 'best avoided'. I've never had any serious problems with the PSUs I've just mentioned.

[Snarf81]

Thanks for all of the replies. Very interesting reading material for me.

I've ordered a few from Cricklewood ( I always think of 'The Goodies' at the mention of Cricklewood- haha) Electronics. So I think they'll arrive in a few days.

I'll let you all know if it's cured the problem with the N1700.

Cheers
Rich.

[barrymagrec]

I used a 723 as a switching regulator over forty years ago on a film recorder footage counter - worked a treat.