LM393 comparator output does not attain 'high'

Nigel Goodwin said:

You've not read the post properly - it was uA709 PLUS discretes - basically the 723 was based on an original PSU design using a uA709 and plenty of extra components, the more integrated design giving better performance and using many less parts.

To be fair the 723 is a serious antique now

Click to expand...

That antique just keeps plugging along.

Also, I did read it properly, both the post and the attachment. I asked the question to get some clarity for others who weren't around when those parts came out.

ak

The 2nd gen component improvements and better differential matching and with bandgap ref.

Hi,

Yeah the good old ua723 I remember very well. I loved it back in the day because it was a voltage regulator that could easily be current boosted, and it also had that temperature compensated zener inside (not all packages had that though). I loved it, but that was back in the 1970's and maybe a little in the early 1980's.

Today it gives way to a plethora of new parts made for voltage regulation and especially gives in to the trend to move away from linear regulators in general. It's a shame in a way it was a good chip, but today we have parts much better than any zener could ever be in the form of voltage reference diodes. Back then the best you could get was an IC with a voltage regulator inside and a heater that would keep the temperature above ambient, enough to that it would always be at a constant temperature. These guys were expensive too, at least $10 per IC chip unless you found a sale somewhere. They were also big too, much, much bigger than the SMD parts we see today.

I still love linear regulators though, and I have one I built a long time ago that has some special features like very, very fine output voltage adjustment. I needed that for some tests I was doing back then. Most of the power supplies we see today have a fine voltage adjustment, but it's usually not fine enough for some things. If I wanted 12.002 volts output I did not want 12.001 volts nor 12.003 volts, I wanted 12.002 volts, and I wanted it to be steady over times of perhaps hours. It's hard to get that even with todays better voltage regulators because there are some inbred faults.
Even the notorious LM317 and similar have what might be a serious fault: the voltage reference IC part is inside the same packages as the power handling components like the series pass transistor. That means than when the device heats up, the voltage reference also heats up. The voltage reference itself has an amazing temperature characteristic, but when heated and used to reference the output voltage it could still result in a 2 percent (or more) tolerance over the temperature range. It's hard to spot that from the data sheet too so it's often overlooked. Luckily this is not an issue for everything though, just those applications that have to work over a wide range of temperatures and don't have to be "very very accurate", just "accurate".