1. Home
  2. Electronics Design
  3. Programmable vs. "knob" power supplies

Programmable vs. "knob" power supplies

Hi:

I returned my Protek 3044T power supply, which I didn't like because it's impossible to set low current limits precisely. Partly due to the

0.01A resolution meters, but also the knobs are linear. And at low currents there was a large 2x discrepancy between the setpoint and actual limit (actual lower than limit). Finally, the one I received at home buzzed when I turned it on because a transformer cover was not secure and not easy to fix.

I'm also likely to dump my Protek 1805 power supply on Ebay. Anybody want it for 60% of the $193.80 purchase price (not incl. shipping)?

At least Xantrex puts logarithmic current control pots in their "knob" type power supplies.

But also, I found that the Protek supplies get screamin' hot even at currents of about half the rating. This bugs me. I don't like designs that push the limits of component temperature.

I regret trying to save money with the Protek supplies. More than half of the time I try saving money I wind up sending junk back, wasting a bunch of time&shipping charges. Almost never do I regret buying something with a hefty price tag but clearly superior quality.

I'm now inclining toward programmable supplies only, because I'm getting hooked on the digital settability after getting an Agilent E3631A and a Xantrex XDL35-5 at work. Agilent user interfaces leave much to be desired, though.

The Xantrex is one precise power supply! And the user interface is very good as well. Monday I should give it a heavy load test to see how the heat is.

What I need is a logic level supply that can do at least 18V 5A, and a dual supply that can do at least 30V 2A.

I was considering replacing the Protek 3044T with a Xantrex XPL30-2D "knob" supply. This would be adequate for +/- rails for linear type circuit experimenting, and their log pots should avoid the low current settability problem of many knob-set supplies.

formatting link

Then splurge on the $815 Xantrex XDL35-5 for the logic supply. The thing that bugs me is that the Xantrex doesn't have a computer interface unless I pay $1025 for the XDL35-5P model. Ouch!

formatting link

For that money, it makes me consider instead to look at the Instek PSM-3004 or PSM-6003 which have standard interfaces for more versatility in terms of voltage and current range. But then again, I have more use for the Xantrex's 0.1mA lower end precision.

formatting link

So many choices; Ugh!

--
_____________________
Christopher R. Carlen
crobc@bogus-remove-me.sbcglobal.net
SuSE 9.1 Linux 2.6.5
Reply to
CC
Loading thread data ...

You probably won't go running out to get one like mine...

formatting link

This design features marginal regulation and output ripple. Its one real nice feature is a current limit cutout. You set it in .01 A increments up to 4 amps. When it senses the limit, it just shuts off. This all but eliminates the 'constant current cookout' when the circuit I'm designing screws up.

Luhan

Reply to
Luhan

If you ain't got a button you must be a knob ;-)

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.
Reply to
Jim Thompson

No, I probably won't. I built a dual 28V 5A supply using LM338K devices. I don't like the foldback limiting that wreaks havoc on anything which draws brief current surges. Also, no current limit setpoint at all.

I don't have time to build everything anymore, so now I'm willing to pay for test instruments and spend the rare hours here and there on something that can't be purchased.

Good day!

--
_____________________
Christopher R. Carlen
crobc@bogus-remove-me.sbcglobal.net
SuSE 9.1 Linux 2.6.5
Reply to
CC

I'm no fan of Protek, either. My workplace bought half a dozen of the same model (the exact model number escapes me, though it is a dual-polarity, analog-adjustable type with digital LED readouts) brand-new several years ago, and all fix of them failed within a year. We fixed them ourselves and "improved" them slightly, and they haven't given us much trouble since. But they do seem to break rather easily.

Reply to
Matt J. McCullar

Oh my, that's not a good stat. Have you any experience with Instek, or Xantrex?

Thanks for input.

Good day!

--
_____________________
Christopher R. Carlen
crobc@bogus-remove-me.sbcglobal.net
SuSE 9.1 Linux 2.6.5
Reply to
CC

I've used a XPL30-2D for a couple of years. Works well. The log current pot is approximated using 3 linear track segments and is very human friendly. Biggest drawback is voltage setting. The 'course' pot is single turn and runs 0-30V. It's soooooo easy to catch the knob and suddenly power those expensive 5V components from 20V. Voltage knobs now held in a mastic gunge. Unsightly but slows 'em down. Single 5 or 10 turn pots, would have improved the ergonomics no end. (Though as with most modern products, function is subserviant to form). john

Reply to
John Jardine.

Why is the form superior with the coarse/fine scheme instead of a

10-turn. This constantly bugs me about most linear supplies. I put a 10-turn on my power supplies. I suppose they save a few $. But at a large cost in function, as you say. Why don't designers tell their bosses to shove it, and argue for a superior product?

Have you checked if the XPL30-2D appears possible to modify with 10-turn pots to replace the coarse V and I pots (abandoning the fine V)? Protek had the pots on a PCB mounted to the front panel, so such a mod was hopeless.

I tend to be disappointed by the slightest deficiency in things, making me wish I had just spent more for something better, or considering the "perfect" design to build myself.

I'm getting close to considering the $1290 dual XDL35-5T for the +/-15V stuff. Though I don't know why they didn't bother to make the 3rd output programmable and more substantial. If they had made it 0-6V at

5A or something, it would be an Agilent E3631A killer for sure.

So I tend to consider the XDL35-5T really only a dual. Maybe try a cheaper $375 B&K 1785B perhaps for the logic supply.

The ultimate setup would be the XDL35-5T and the Instek PSM-3004.

Good day!

--
_____________________
Christopher R. Carlen
crobc@bogus-remove-me.sbcglobal.net
SuSE 9.1 Linux 2.6.5
Reply to
CC

Sadly, (lke the Protek) the pots are mounted on a big front panel PCB. Could drill through I suppose but there's about a dozen tracks that would then need linking through. A hopeless case :( john

[...]

Last year, mail ordered a standard, mini tower, PC case. Front panel a swirl of artistic, post-modern, sculpted mouldings. Can't though f*****g eject a floppy disc. Disc has to be teased out of the artistic mouldings using long nose pliers. The PC and programming world seemingly survives with secondrate as a norm. Yet a powertool company (as a real world example) would only last months if a similar attitude to users needs was adopted. john

Reply to
John Jardine.

Hello Chris,

Whatever you buy test it for load change spikes. A client of mine got burned by that, big time. I actually had to come out there.

What surprised everyone was that my old Russky power supplies did not exhibit that problem at all. These also were difficult to adjust but I swapped the pots for 10-turns plus nice mechanical counters.

John Larkin had once pointed these out:

formatting link

No idea how they deal with load changes, hopefully John will let us know some day when he has run the supplies through the paces.

--
Regards, Joerg

http://www.analogconsultants.com
Reply to
Joerg

I started building a design like this after seeing something similar on the net (software controlled regulation), but during testing I decided that the response time to shifting load current was just never going to be fast enough with a microcontroller. Have you done any torture tests on your supply to see how well it regulates under variable load?

--
Ben Jackson AD7GD

http://www.ben.com/
Reply to
Ben Jackson

Hi John,

In my experience, this can usually be fixed by unscrewing and then *carefully* re-mounting the drive bracket into a *working* position prior to tightening the screws. The problem seems to be that tolerances are really sloppy on a lot of these cases, so while there is *some* screw position that makes everything align, it's almost never obtained when someone just arbitrarily screws everything together.

Interestingly, companies like Dell seem to do a good job at proving that case design *can* be both mechanically robust and inexpensive -- it just requires (apparently) above-average designers.

Reply to
Joel Kolstad

Hi,

I am a seller of pre-owned test equipment and to me, the choice is obvious...............Xantrex !!!!!!!!!!!!!!!

Martin

"CC" a écrit dans le message de news: snipped-for-privacy@news1.newsguy.com...

Reply to
Martin Plourde

I'm definitely not going to go with Protek again. But can you say anything about Instek? The PSM series dual range supplies look really good on paper, except for being massive.

--
Good day!

________________________________________
Christopher R. Carlen
Principal Laser&Electronics Technologist
Sandia National Laboratories CA USA
crcarleRemoveThis@BOGUSsandia.gov
NOTE, delete texts: "RemoveThis" and
"BOGUS" from email address to reply.
Reply to
Chris Carlen

agree it's a stupid design. hit the eject button harder (works sometimes) or consider replacing the drive, I get 3-4cm here, or just rip that stupidf plastc bit off, it serves no useful purpose.

Rhino's been making makita knock-offs for a few years now...

Bye. Jasen

Reply to
jasen

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.