Experiences with bench power supplies?

April 30, 2009 7:41 AM

What are your experiences with different brands and types of benchtop/lab power supplies? Do certain ones fail more often than others? Easier to fix? Better power conditioning?

I'm a home experimenter. The dinky supply I bought to try to save space died a quick death and I'd like to avoid repeating that. I don't need anything crazy. 0-30V, 0-3 amps, accurate, small, and something that won't die. Are Tektronix (PS281?) worth the money? I have an oscilloscope from them. Instek, Extech, Protek, Velleman? Is there a specific one that is particularly popular and a good deal? Used and new are ok. Cost is a factor ($100USD?).

posted by jwells to technology (8 comments total)

I'm a fan of the Instek 1830D (a Tektronix PS281 "look-alike") for cheap power supplies. However, it still around $175.

As an aside, Tektronix is really a measurement company. Agilent is the king of the power supply market right now.
posted by saeculorum at 7:59 AM on April 30

Of course, as a learning experience, it could be interesting to make your own power supply. You won't end up with something that has anywhere near the supply quality as even some of the cheaper-end bench power supplies, but will that really matter for what you're doing?
posted by saeculorum at 8:02 AM on April 30

I converted one of my old PC supplies as outlined here. For monkeying about on my little projects at home it works just fine. And if it stops working I've got three more power supplies I can convert (cast-offs from friends and family).
posted by le morte de bea arthur at 8:17 AM on April 30

We have a few BK model 1626A's that are 3A and 0-30V, looks like they are $229 right now. You might be able to get a used one around $100. We're pretty happy with them.

You can get good fixed 3A DC power supplies from digikey for around $50 with a much smaller form factor. You could build your own voltage divider with a couple of high res 10-turn pots.
posted by mbd1mbd1 at 8:27 AM on April 30

You could build your own voltage divider with a couple of high res 10-turn pots.

You don't want to make a variable supply with pots! You want to use three-pin adjustable regulators instead. The application notes on this linked page even gives an example of a constant voltage / constant current 1.2-30V, 5A "bench" supply.
posted by fatllama at 2:57 PM on April 30

Building it might be the best idea. Thanks! How could I get a digital display of the voltage and current levels with the kits or that LM317 diagram?
posted by jwells at 7:29 AM on May 1

You could get a digital multimeter pretty cheaply now, and it would be useful for lots of other bench work as well. Just make sure it measures voltage and current - not just voltage.

fatllama is right about the regulators for the power supply - I actually just got a handful for a project today.
posted by mbd1mbd1 at 3:18 PM on May 1

Digital panel meters (like this) for voltage are pretty good. Some may require a certain fixed voltage supply; you can use the fixed three-pin regulators (i.e. the 78xx series) for this. You often set the measuring range with jumpers and voltage dividers, and can therefore measure hundreds of volts while only supplying 5V or 10V to the meter. Some require, and yet others forbid, the meter power supply to be grounded in common with the voltage being measured---beware!

For measuring current... Well, I've never been happy. Sometimes I put a 0.1 Ohm power resistor in series with my output and measure that (potentially floating) voltage drop with another digital panel voltage meter. I usually favor regular old dc current "needle on a scale" current meters. For large (say, >10A) ac currents, there are meters that are pared with an ac transformer to bring your large current down to something like 5A which corresponds to full-scale on the meter. Or "clip-on" inductively coupled ac or dc current meters.

Books like (even previous years', and thus cheaper) ARRL handbook are full of great, well thought-out general purpose projects like power supplies and tutorials on everything from how to solder to how to wind your own inductors. On the downside, it's no Horowitz and Hill, but on the upside, it's no Horowitz and Hill.
posted by fatllama at 8:00 AM on May 3