23-07-2018, 06:58 PM
To be honest, probably not.
The LM317T is excellent for the price, and also excellent for the intended application. Yes, there are better regulators out there, but for a transistor radio - originally designed to run on batteries - the advantages would be wasted. And the price would be a lot higher.
I still use the LM317T for many things. Where possible, it's best to avoid circuit designs that require great things from a voltage regulator. When I did my industrial year at university, I was asked to look at a circuit designed by a student who did his gap year previously. It had to take the output from a DAC and turn it into a current (up to a few 10s of milliamps) - but it had to be very stable. His circuit involved 4 op-amps, and the current was strongly dependant on the applied voltage. This in turn required an extremely expensive voltage regulator IC, but still, the very small temperature drift was noticeable (required an internal temperature sensor that fed into the software IIRC). All in all, a mess...
I took a quick look at this and replaced it with just 2 op-amps. It didn't care what the supply voltage was, nor the temperature. It saved the company a fortune.
Here's the rub. This character had returned as a graduate engineer by now, and was now a full time employee. I was helping him out. Subsequently, I found out how much credit I'd been given for this redesign. Absolutely none whatsoever - he passed it off as all his own work, and took full credit for my work. I learnt a valuable lesson there. Not least, how dumb the management was for not suspecting, given that he was a pretty hopeless analogue design engineer... And for approving the circuit in the first place, frankly.
The LM317T is excellent for the price, and also excellent for the intended application. Yes, there are better regulators out there, but for a transistor radio - originally designed to run on batteries - the advantages would be wasted. And the price would be a lot higher.
I still use the LM317T for many things. Where possible, it's best to avoid circuit designs that require great things from a voltage regulator. When I did my industrial year at university, I was asked to look at a circuit designed by a student who did his gap year previously. It had to take the output from a DAC and turn it into a current (up to a few 10s of milliamps) - but it had to be very stable. His circuit involved 4 op-amps, and the current was strongly dependant on the applied voltage. This in turn required an extremely expensive voltage regulator IC, but still, the very small temperature drift was noticeable (required an internal temperature sensor that fed into the software IIRC). All in all, a mess...
I took a quick look at this and replaced it with just 2 op-amps. It didn't care what the supply voltage was, nor the temperature. It saved the company a fortune.
Here's the rub. This character had returned as a graduate engineer by now, and was now a full time employee. I was helping him out. Subsequently, I found out how much credit I'd been given for this redesign. Absolutely none whatsoever - he passed it off as all his own work, and took full credit for my work. I learnt a valuable lesson there. Not least, how dumb the management was for not suspecting, given that he was a pretty hopeless analogue design engineer... And for approving the circuit in the first place, frankly.
