POWER DESIGNS PRECISION POWER SUPPLIES

[F4]

I found one problem, but am unable to diagnose the transient I am seeing around what looks like the oven going on and off.

If you find your are out of spec at the 0000 calibration step, investigate R24.  I added 1k in resistance to that location and it brought my readings from +30mv down to +1.2mv. I have dialed that 1.2 down to 0s and am waiting on the unit to stabilize before dialing it in further.

As far as transients, has anyone ever investigated their cause?

This unit had 4 problems, these are the mitigation steps taken. 
1. The failed calibration with R24
2. Too much ripple solved by recapping
3. Voltage drift upwards, this was resolved incrementally from all steps but benefited the most from fix 4.
4. I spent several hours digging into the issue with transient voltages appearing on the output. The supply is supposed to stay stable at .0000, this one was doing that but had a lot of variance past 4 zeros, so 0.0000.001 to 0.0000.999. When the transient appeared it was sending the signal up to 0.0002.xxx every 20 seconds or so, some instances more intense than others. It also continued to voltage drift upwards with 0000 on the dials (no output).

I tested the transistor paths back to the power supply, saw the transients everywhere I put the scope leads:



The transients were even across pins on the power supply itself, which got strange as I thought it was the power supply that was the problem. I tool a look at the schematic again and noticed the heater circuit goes right from the 115v AC source, which maybe explains how the transient was seen everywhere:



Up until here I didn't realize the tap for the oven was at the transformer. I looked at the 2 caps across pins 9-11 on the oven, C9 and C14. I changed both caps and added 2 resistors to form more of a snubber on the feedback.  This worked, I ended up with 270 ohms, 75 ohms, and a 10nf ceramic cap, and 2 x 10nf to make the 20nf cap.

I still get the transient, but after several hours in stable conditions, the transient (from cold start) maxed out at 0.0000.400 versus 0.0002.000, a 5 fold decrease.




In summary, at rest, powered on, the unit is much more stable at 0000s, spends most of its time at 0000.010 or below, and after being stable 12 hours shows only two transients in a 15 minute test exceeded 0000.010, one hit .039 and the other .024.

Moving on to the 20v calibration, hopefully the stability translates there...

Edit: its a 220 ohm resistor, not 270.

[F4]

Following up again, I have narrowed the transient mentioned in a previous post down to the thermostat. I figured it was the thermostat or the oven bulb as they were the only things in the path other than the heating element itself. I converted both the power on bulb and oven bulb to LED to eliminate the oven bulb as a root cause.

(Attachment Link)

Monitoring pins 10 and 11 I observed the voltage stay close to 0, and on the scope monitoring the output I see the transient when the thermostat appears to signal for heat, but before the heater jumps to 120v across pins 10 and 11.

I couldnt find a picture of the thermostat anywhere, and eventually decided to open up the oven and see what was servicable. Turns out the thermostat and oven casing look tack welded together, or somehow soldered with the heater unit.  Not servicable.




I was thinking of changing the heating element on the can to one of these:

https://www.amazon.com/Coliao-Adhesive-45mmx100mm-Elements-Polyimide/dp/B0CM6KKHXX/?th=1

My problem is that I am not sure what a simple way would be to get this different heating element to run at a constant temperature from the oven.

Would these work?

https://www.amazon.com/gp/product/B0861C3ZNB/ref=sw_img_1?smid=A1THAZDOWP300U&psc=1

[PKeithM]

I can add a little history about this company if you are interested at all.
I worked at Power Designs Inc (Westbury NY) summers from 1973 to 1977, while going to school to for my BSEE.

I applied at the start of the summer 1973 to be a repair technician for new (power supplies that failed tests) and returned units.
Even though all I had was electronics in high school (three years), a Ham radio license and a 1st class commercial radio license (I was 18 when I started), the president/owner of the company, Herb Roth (sp?), hired me anyway after a series of interviews with many folks including the Chief Engineer Mischa Vamberg (sp?).  Herb's wife ran the front office. I can credit much of my career to the tutoring I got from Herb and Mischa on practical engineering and especially on feedback circuit design.

 During the week I worked as an assistant to Mischa helping him develop/improve some of the power supplies you folks are talking about and a lot of special project supplies (power supplies for Grumman jets, High Voltage/high current supplies for IBM research, NIM supplies,  and so on). I distinctly remember working on the temperature stability enclosures to improve voltage stability and reduce noise used in many of the supplies. I built hundreds of these as the zener diodes we received had to be selected by hand and then the assemblies were tested over weeks to meet the power supply specs. For some reason the ones I built worked really well and I was rewarded with getting to build a wholes years supply over each summer.

On weekends I worked with Herb at adjacent test benches fixing  supplies that the regular technicians were unable to troubleshoot (or just help him clear the backlog of testing new supplies). There was no scrapping out of spec supplies, we figured every one out.

Herb loved to talk on the weekends, and here is the story he told me how he got the initial money for starting the company.  Apparently, he was a radio technician (or worked with radios) during WW II and many years after the war a whole lot of surplus radio radio equipment was sold for pennies on the dollar and ended up in warehouses located around Canal street in Manhattan. Herb was aware that a certain class of vacuum tubes had grid wires that had a high gold content and he bought thousands of them (every one he could find) for very little money (nobody in the surplus chain realized the gold wire was there). Herb and his wife spent weeks breaking each tube and stripping out the gold wire. They then sold the gold and used the proceeds to start the company.

Herb was focused on exceeding published product specs. As you probably realize, that certain parts in the supplies were selected to meet very tight margins (more than the manufacturer was willing to meet). We had in house  built equipment we used to select the parts. If it survived over time, you would see this as various colored dots (actually cheap nail polish) on parts (mostly semiconductors) that were selected out for a specific set of characteristic depending on use (this varied by supply and even in a single supply). This practice was very prevalent in the higher reference supplies (those with the temp chambers) and the NIM supplies. So while the schematic may call out a part number, the internal documentation would include a specification of the colored dot(s) the part had to have to be used in that specific application.

[rsjsouza]

Thank you for the bit of history about the company.

However, are you sure the designers / owners weren't aliens?  The lore in this and other threads about the legendary quality of Power Designs is the following:

As someone else hinted in a past discussion, I am suspicious they were in fact from another planet and graced us with their designs...

Glad someone else believes this, since there is no internet history on them.  If something exists then it will be on the internet.  Since "they" do not exist but their products exist, something must be going on.  A logical explanation is aliens made them for some purpose that we do not know about.  Like help develop something for space flight that will take us to another planet.  Once this theory is proven then the value of these supplies will greatly increase.  I just hope humans will not become a food source like I think about every time I look at my power supply.  God forbid, the PD is not the alien (but just in case I use kind words when I talk to my power supply)

[F4]

I can add a little history about this company if you are interested at all.
I worked at Power Designs Inc (Westbury NY) summers from 1973 to 1977, while going to school to for my BSEE.

I applied at the start of the summer 1973 to be a repair technician for new (power supplies that failed tests) and returned units.
Even though all I had was electronics in high school (three years), a Ham radio license and a 1st class commercial radio license (I was 18 when I started), the president/owner of the company, Herb Roth (sp?), hired me anyway after a series of interviews with many folks including the Chief Engineer Mischa Vamberg (sp?).  Herb's wife ran the front office. I can credit much of my career to the tutoring I got from Herb and Mischa on practical engineering and especially on feedback circuit design.

 During the week I worked as an assistant to Mischa helping him develop/improve some of the power supplies you folks are talking about and a lot of special project supplies (power supplies for Grumman jets, High Voltage/high current supplies for IBM research, NIM supplies,  and so on). I distinctly remember working on the temperature stability enclosures to improve voltage stability and reduce noise used in many of the supplies. I built hundreds of these as the zener diodes we received had to be selected by hand and then the assemblies were tested over weeks to meet the power supply specs. For some reason the ones I built worked really well and I was rewarded with getting to build a wholes years supply over each summer.

On weekends I worked with Herb at adjacent test benches fixing  supplies that the regular technicians were unable to troubleshoot (or just help him clear the backlog of testing new supplies). There was no scrapping out of spec supplies, we figured every one out.

Herb loved to talk on the weekends, and here is the story he told me how he got the initial money for starting the company.  Apparently, he was a radio technician (or worked with radios) during WW II and many years after the war a whole lot of surplus radio radio equipment was sold for pennies on the dollar and ended up in warehouses located around Canal street in Manhattan. Herb was aware that a certain class of vacuum tubes had grid wires that had a high gold content and he bought thousands of them (every one he could find) for very little money (nobody in the surplus chain realized the gold wire was there). Herb and his wife spent weeks breaking each tube and stripping out the gold wire. They then sold the gold and used the proceeds to start the company.

Herb was focused on exceeding published product specs. As you probably realize, that certain parts in the supplies were selected to meet very tight margins (more than the manufacturer was willing to meet). We had in house  built equipment we used to select the parts. If it survived over time, you would see this as various colored dots (actually cheap nail polish) on parts (mostly semiconductors) that were selected out for a specific set of characteristic depending on use (this varied by supply and even in a single supply). This practice was very prevalent in the higher reference supplies (those with the temp chambers) and the NIM supplies. So while the schematic may call out a part number, the internal documentation would include a specification of the colored dot(s) the part had to have to be used in that specific application.

This was a great read, makes it more meaningful to tinker around with the units. I do indeed have some of those dots inside.

The boards, were they made in house or outsourced?  It's a nice experience soldering with the lugs.

Please do stick around, we'd love to hear more.

[Melt-O-Tronic]

Please do stick around, we'd love to hear more.

YES, THIS!

Most of us have a lot of older equipment that gives us great service.  These tangible artifacts represent our under-appreciated history of innovation,  People are recognizing that.  But rarely are the records & stories of the design & production evolution of this gear available to those of us who are interested.  There is a lot to be learned for our throwaway society right now about the experiences and genius that went into this gear; gear that paved the way to the successes we enjoy today.

[mawyatt]

Please do stick around, we'd love to hear more.

YES, THIS!

Most of us have a lot of older equipment that gives us great service.  These tangible artifacts represent our under-appreciated history of innovation,  People are recognizing that.  But rarely are the records & stories of the design & production evolution of this gear available to those of us who are interested.  There is a lot to be learned for our throwaway society right now about the experiences and genius that went into this gear; gear that paved the way to the successes we enjoy today.

Agree we tend to forget the great innovations and stories from the past. While an adjunct way back in early 2000s we tried to convince the university while on the Presidents Advisory Board to offer a course on engineering history, unfortunately this fell on deaf ears!!

Much of the great stories like Power Designs soon will be forgotten, hopefully others will come forward with important and interesting historical stories from our past engineering life!!

Best

[meteo]

Could you recommend adjustable power supplies up to 500-600 volts / 200 - 300 mA and up to 1000 - 1200 volts 100 - 200 mA POWER DESIGNS level. Voltage ripple is preferably up to 1-5 mV.
Needed for the anode voltage of radio tubes.

[Barbouri]

Just finished working on a somewhat rare Power Designs model 605 precision power source.

More pictures and blog at:
https://www.barbouri.com/2026/05/31/power-designs-inc-model-605-restoration/

[mawyatt]

Those old Power Designs Power Supplies bring back fond memories, our labs were full of them. Seems you've found one in prestine condition

Best

[zike]

Haven't fully digested this massive and ancient thread, but pleased to contribute where I can.  I recently documented fixing a number of failed Power Designs DC supplies, dating from ca. 1963 to 1985, over in the Repair forum. Here are links: 

TP343A (advertised non-working for parts, only $50, but actually tested fine)

6050A

2005 and 2005A

4050

5015A

They're also linked in the "database" thread.

[Xiphos]

I can add a little history about this company if you are interested at all.
I worked at Power Designs Inc (Westbury NY) summers from 1973 to 1977, while going to school to for my BSEE.

I applied at the start of the summer 1973 to be a repair technician for new (power supplies that failed tests) and returned units.
Even though all I had was electronics in high school (three years), a Ham radio license and a 1st class commercial radio license (I was 18 when I started), the president/owner of the company, Herb Roth (sp?), hired me anyway after a series of interviews with many folks including the Chief Engineer Mischa Vamberg (sp?).  Herb's wife ran the front office. I can credit much of my career to the tutoring I got from Herb and Mischa on practical engineering and especially on feedback circuit design.

 During the week I worked as an assistant to Mischa helping him develop/improve some of the power supplies you folks are talking about and a lot of special project supplies (power supplies for Grumman jets, High Voltage/high current supplies for IBM research, NIM supplies,  and so on). I distinctly remember working on the temperature stability enclosures to improve voltage stability and reduce noise used in many of the supplies. I built hundreds of these as the zener diodes we received had to be selected by hand and then the assemblies were tested over weeks to meet the power supply specs. For some reason the ones I built worked really well and I was rewarded with getting to build a wholes years supply over each summer.

On weekends I worked with Herb at adjacent test benches fixing  supplies that the regular technicians were unable to troubleshoot (or just help him clear the backlog of testing new supplies). There was no scrapping out of spec supplies, we figured every one out.

Herb loved to talk on the weekends, and here is the story he told me how he got the initial money for starting the company.  Apparently, he was a radio technician (or worked with radios) during WW II and many years after the war a whole lot of surplus radio radio equipment was sold for pennies on the dollar and ended up in warehouses located around Canal street in Manhattan. Herb was aware that a certain class of vacuum tubes had grid wires that had a high gold content and he bought thousands of them (every one he could find) for very little money (nobody in the surplus chain realized the gold wire was there). Herb and his wife spent weeks breaking each tube and stripping out the gold wire. They then sold the gold and used the proceeds to start the company.

Herb was focused on exceeding published product specs. As you probably realize, that certain parts in the supplies were selected to meet very tight margins (more than the manufacturer was willing to meet). We had in house  built equipment we used to select the parts. If it survived over time, you would see this as various colored dots (actually cheap nail polish) on parts (mostly semiconductors) that were selected out for a specific set of characteristic depending on use (this varied by supply and even in a single supply). This practice was very prevalent in the higher reference supplies (those with the temp chambers) and the NIM supplies. So while the schematic may call out a part number, the internal documentation would include a specification of the colored dot(s) the part had to have to be used in that specific application.

Your history is fantastic!  Thank you for sharing it.  Which are your favorite models of the precision power supplies?  Which do you think are the best built and design?  I.e., which models are the pinnacle of Power Designs supplies?

[zike]

Apologies for multi-post, Re: this I find my 5020 needs the 1k\$\Omega\$ single turn wirewound pot with axial pull switch used to set the current limit on 5020, 2005A, 2020B and so on.

Curious if anyone posting here has had the misfortune to part out one of these models? Sad as I'd be to hear it, occasionally this must happen; and perhaps the most honorable afterlife may be that of an organ donor. So says my driver's license, anyway.

For reference it's internal PD part no. 868057-1. Mine, a Clarostat, happens to be marked "B67034-1  200Ω" though in real life it is (was) actually 1kΩ.

-zike
 

[zike]

Apologies for multi-post, Re: this I find my 5020 needs the 1k\$\Omega\$ single turn wirewound pot with axial pull switch used to set the current limit on 5020, 2005A, 2020B and so on.

EDIT 7/20/26: NEW POT DIDN'T WORK OUT. DPDT CONTACTS FUSED WHEN CHECKING CURRENT AT FULL OUTPUT.

Located a substitute 1k\$\Omega\$ pot with pull switch. My 'new' 5020 is back in action, with all features intact (including "pull to set" current limit knob).

It was helpful to realize that there's no apparent reason (IMO) to have specified a 2 watt wirewound pot for R41. I don't know why they did, curious to hear any theories. From the schematic (which, of course, I had to redraw; attached), the pot carries at most half a milliamp, dissipates a quarter milliwatt, and doesn't particularly benefit from linearity, precision, or thermal stability. 

Broadening the search parameter space turned up an odd cache of Spanish linear-taper, quarter watt 1k carbon pots with integral 2A DPDT push-pull switches.  The seller "electropartsgalore" didn't have any docs, and they came unpackaged and unmarked; but visually, they appear to match Piher (now part of Amphenol) T-21 series. 

The 2A switch contact rating is marginal (it sees 2.1A), but it's a DPDT, so I just paralleled the sections. However the shaft and bushing are metric. Boosting the 6mm shaft to the 1/4" PD knob just took a rolled piece of .007" shim stock. But I didn't like the idea of wallowing out the 3/8" panel hole for a Euro 10mm bushing. Felt like defacing a sacred relic.

OTOH 3/8" is darn close to the nominal M10x0.75 thread root diameter; so I compromised. Without redrilling, I just ran an M10x0.75 tap through the hole, then screwed the pot's bushing through from the inside. Washers and jam nuts on both sides allowed clocking the terminals to suit wiring.  The panel ID remains 3/8",  should anyone want to fit an OEM pot in future (presuming they first figure out that this alien pot is threaded in...).

I still have no good theory for what happened to the original R41, or to the fried current control amplifier U1B.

At the point R41 gave out, I was setting up to chase what appeared to be 3mV of ripple on the output; the pot blew before I got to the bottom of it, but whatever it was it's gone now. Maybe I had a ground loop or something. Now I measure less than 20 \$\mu\$VRMS and 100 \$\mu\$Vpp ripple on top of 20VDC output with 1A load. The FFT shows 3.5 \$\mu\$VRMS at 60Hz, 1/10th that at 180 Hz, and nothing measurable at 120. You need some A-list grounding and shielding to make a battery look that clean in my lab.

Also as mentioned in the "repair" post, this unit has a really nice resistor set; against a known good Agilent 34401A, output stays within \$\pm\$100ppm of dial setting from 5V to 50V (10x better than spec). This one's a keeper.

One peculiarity: the 10mV vernier seems kind of nonlinear. Around 60-80% (nominally adding 6-8 mV to output) it's out by almost 0.6 mV, if calibrated dead nuts at zero and 10. I wonder if that 10\$\Omega\$ 10-turn helipot is bad? And down another rabbit hole we go....

Zike

PS eBay seller above messaged me that as of  early July 2026, they have at least 150 more of those Spanish pots.

[zike]

One peculiarity: the 10mV vernier seems kind of nonlinear. Around 60-80% (nominally adding 6-8 mV to output) it's out by almost 0.6 mV, if calibrated dead nuts at zero and 10. I wonder if that 10\$\Omega\$ 10-turn helipot is bad? And down another rabbit hole we go....

Just wondering if someone else with a 5020, 4010 or C500 (or other model with the 10-turn 10\$\Omega\$ vernier pot) could pop the bottom off (unpowered) and measure this pot with a 4-wire ohmmeter?

Shouldn't matter too much, but for consistency, supply dials should be set to a high voltage. Resistance with the vernier dial at 0%, 50%, 80% and 100% would be very instructive.

Briefly, on my 5020 both output voltage error and 4-wire R vs. vernier dial are perfectly consistent with a phantom 0.75\$\Omega\$ in series with the pot wiper (with best fit of 9.7\$\Omega\$ for the pot itself, and additional 0.05\$\Omega\$ for each end terminal). This produces an error of +450 mV \$\mu\$V with the vernier at 8 mV nominal, assuming that both zero and 10mV extremes have been trimmed exactly.  The pot is not noisy or intermittent, and the reconstructed phantom resistance is consistent at all positions; so it seems unlikely to be a degraded or dirty contact. What else might cause this?

I've located the modern descendant of this Beckman Helipot at Digikey (special order, $170 each!), and Newark listed an ETI alternate for "only" $83, but it would be good to know if this is just 'the way they are;' such a phantom resistance would not affect use *as a potentiometer* (as opposed to a rheostat), so long as minimal current is drawn off the wiper.  Working against this theory is the QA cert found in the older 5020 manual scan, that shows very accurate vernier tracking from 0 to 10mV.

Even if my pot is somehow bad, as long as it's stable I think it could be resolved by just clipping the short to the opposite end of the pot. The residual resistance at minimum is nulled out by the zero control, and the total span is normalized by R77 and Q11. The only possible worry is railing the supply if the wiper "skips" or goes open; but this can be prevented by a clamping diode across the pot.

Thanks in advance for any advice or opinions!

zike

[Vgkid]

Take a picture of the potentiometer.

[zike]

Take a picture of the potentiometer.

"Beckman RWV-100-3C-10T"

apparent match

possible alternate

[zike]

Even if my pot is somehow bad, as long as it's stable I think it could be resolved by just clipping the short to the opposite end of the pot. The residual resistance at minimum is nulled out by the zero control, and the total span is normalized by R77 and Q11. The only possible worry is railing the supply if the wiper "skips" or goes open; but this can be prevented by a clamping diode across the pot.

Following up, I haven't sourced such a pot, but I did confirm by isolating the existing one that my theory (phantom 0.75 ohms in series with the wiper) was correct. Also confirmed that severing the short between the wiper and the dead end of the pot, and then readjusting R42 (zero) and R77 (10 mV cal), completely resolved the issue; output voltage now corresponds within 140 uV or 1.5% of setting throughout its range, vs. 440 mV or 7% at best trim before.  I did add a 1N914 diode clamp in case the wiper goes open, although the contact appears to be in good shape.

Still puzzled about that phantom 0.75 ohms but it is what it is (and appears stable).

Incidentally there was another complication I had overlooked: PD specified a peculiar tolerance band of +0/-10%. If the resistance span exceeds 10 ohms over the wiper travel, current source Q11 setting the 10 mV range can't cope, as it can only source, not sink. Besides costing a fortune, half of available pots with "normal" ±5% tolerances will not work. So it's fortunate I could rewire this one to compensate.

In the "taketh away" department: turns out the switch on that Piher current limit pot I was so proud of can't take the 1/3 Joule stored in C21 at full output. Testing the "pull to set current" feature at 50V welded the contacts. Oh well. Don't do that.

I swapped in a regular old 1k pot, and will just have to short the output with a screwdriver to check current limit, like the farm boy I truly am. Le mieux est l'ennemi du bien.

[zike]

In the "taketh away" department: turns out the switch on that Piher current limit pot I was so proud of can't take the 1/3 Joule stored in C21 at full output. Testing the "pull to set current" feature at 50V welded the contacts. Oh well. Don't do that.

I swapped in a regular old 1k pot, and will just have to short the output with a screwdriver to check current limit, like the farm boy I truly am. Le mieux est l'ennemi du bien.

OK, OK, I just couldn't ignore that itch. With encouragement from members of the repair forum, I fitted a MOSFET to do the "check current limit" output shorting deed without blasting the fragile Euro switch contacts. Details here.

[zike]

For anybody following this saga, it happened again, this time on a 2020B that came in last week. Am I the only person afflicted by self-immolating R41/S4 pots? Am I doing something wrong?

To recover I'm again implementing the metric Spanish pot with a MOSFET switch, as I did on the 5020.

It also needs a 1 \$ \Omega \$ vernier pot R47, if anyone knows a good source.

Incidentally aside from faults this new-ish 90's era unit carries interesting signs of "value engineering," perhaps signaling PD's eventual decline. For example, the number dials are not engraved; there's a silver-colored decal stuck on each wheel with printed numbers. Factory soldering and loom routing is meh, certainly nothing like the NASA-grade construction found in my older units. More seriously, the LM399 precision reference has here been replaced by an unremarkable consumer-grade LM329D (!) I will be interested in the drift when I get it back in order.

Apologies for multi-post, Re: this I find my 5020 needs the 1k\$\Omega\$ single turn wirewound pot with axial pull switch used to set the current limit on 5020, 2005A, 2020B and so on.

[zike]

Another successful repair, service notes here. More sources for the oddball pots, musings on mid-90's PD craftsmanship, and bonus LM329 exoneration included free of charge.

For anybody following this saga, it happened again, this time on a 2020B that came in last week. Am I the only person afflicted by self-immolating R41/S4 pots? Am I doing something wrong?

To recover I'm again implementing the metric Spanish pot with a MOSFET switch, as I did on the 5020.

It also needs a 1 \$ \Omega \$ vernier pot R47, if anyone knows a good source.

[zike]

... More sources for the oddball pots ...

Just confirming that the Surplus Sales Nebraska item 101290, panel bushing mount pot is a drop-in replacement for the 1k ohm current control R41/S4 on the 5020, 2020B, and C500, and for R35/S4 on the 2005A (Power Designs p/n B68057-1).  They are partially prewired for some specific instrument (easily reversed) and provided in sealed mil-spec foil envelopes with NSN numbers. My samples date from the 1980's.