Author Topic: POWER DESIGNS PRECISION POWER SUPPLIES  (Read 330217 times)

0 Members and 1 Guest are viewing this topic.

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #350 on: October 26, 2014, 01:44:52 pm »
I've got replacement OpAmps on the way. In the mean time I'm just going to start unsoldering one leg of everything until I find something broken I guess... Maybe start with the transistors, then diodes, then caps?


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #351 on: October 27, 2014, 05:34:25 pm »
This one is baffling. It almost sounds like the behavior you see when the input to something is floating.

BTW: Check out this beautiful 2010 rebuild on eBay. I especially like the power cord mod he did.

http://www.ebay.com/itm/Power-Designs-2010-Precision-Power-Supply-Unique-/111493092318

The seller says he's here on EEVBlog frequently, so for all I know it might be one of you guys on the thread here. I hope he gets what he's asking for it, but I suspect the market of potential buyers is too small.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #352 on: October 27, 2014, 09:33:46 pm »
Wow, very nice looking! A bit too pricey I think IMHO though.

And yeah, it is baffling. I almost suspect a ground issue or something. (Judging by the fact that when I connect the ground lead from my scope probe to the negative side of C3 things get better; it also happens if I touch the case of Q1.)


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #353 on: October 28, 2014, 03:34:27 am »
So I tested a bunch more stuff this afternoon (out of circuit), still nothing. Then I ran into a new problem! I had a jumper wire attached to the DC- terminal and it accidentally touched the case of Q1, which caused a spark. Now the thing is completely out of regulation, though the current control still works; pulling it out brings it to 0V without heating up the power transistor resistors (which would happen if Q5, Q6 or Q7 were damaged like when I first got the unit). So this really makes me think there’s an issue with the regulation Op-Amp (U2). Somehow that short-circuit killed it maybe? I replaced Q1 with a 2N3773, but no change, still out of regulation. |O

Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #354 on: October 28, 2014, 06:53:14 am »
Ding Ding Ding! Found a bad component! VR3 is completely shorted. Now, I’m not sure if this happened with the short circuit I described above, or was the cause of all the problems to begin with. Hopefully I’ve got a suitable replacement.

« Last Edit: October 28, 2014, 10:21:05 am by timb »
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #355 on: October 28, 2014, 11:56:27 pm »
So, my fourth precision supply just arrived; another eBay steal. This is a mint condition Model 1025P Remote Programming Precision Power Source, 0-100VDC @ 0-250mA.

It was apparently US Government surplus, based on the front panel sticker. It almost looks like the thing was never used; aside from a small crack on the meter and a scuff on the top cover. The neon lamps are also OEM original and still work with ZERO flicker, which leads me to believe this thing was rarely even plugged in! (The oven lamp would always remain on.)

Speaking of the oven, it heats up perfectly and the thermostat works! I haven’t hooked a programming resistance up yet, but by using the Voltage Limit control on the front, I can get it to go through the full voltage and current range without issues. I had originally bought this as a parts unit to fix my 2005A, but since it’s in such good condition I may end up making my own custom front panel using some nice ganged switches and low PPM resistors; or I could make a small board with relays that sits inside for remote control. Hmmmm….

By the way, if anyone has a 1025/1025P manual, let me know!

If anyone wants to know the date: 1967


Overview


Inside the Supply


Back of the Front Panel


Front Panel


Bottom Side
« Last Edit: October 29, 2014, 04:14:39 am by timb »
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline MarkPalmer

  • Regular Contributor
  • *
  • Posts: 101
  • Country: us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #356 on: October 29, 2014, 02:35:01 pm »
Neat  :)  In the mid 60's where just about every big player in the game was well in to mass produced wave soldered PC boards, PD continued to build all their power supplies one by one, by hand.  This continued for the most part right to their very end, and they are always a joy to look at inside.  I have yet to buy a used PD supply that didn't work perfectly right from the get go, and have never done any repairs to the several I have.  The precision models had a little more complexity to them, this is likely why many of them you come across at auction need some repair. 

-Mark-


 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #357 on: October 29, 2014, 06:51:25 pm »
Yeah, of my 4 units, one had a bad thermostat (understandable when you consider the oven ran 24/7 even with the unit off and that it's mechanical), one had a bad heating cool and the last one had a bad power transistor.

It seems like in the early 70's they went from turret boards to hand soldered single sided copper clad boards.

The green mask on the 2005A is sweet looking and I love the orange color of the 2020B!


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline MarkPalmer

  • Regular Contributor
  • *
  • Posts: 101
  • Country: us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #358 on: October 29, 2014, 08:46:11 pm »
Even with the copper clad printed boards in the later PD supplies, they were laid out like fine artwork- you could put them in a frame and hang them on the wall  :D 

-Mark-
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #359 on: October 29, 2014, 09:28:53 pm »
Does anyone have info on what happened to the company? Back in the day, there was three High End power supply companies in NY, including PD, Lambda and one other name I can't remember. Lambda got bought out and the other one is still going independently.


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #360 on: October 30, 2014, 07:31:18 am »
Damn, this 1025P is pretty amazing. I’m measuring about 500uV stability (6.5 digits with an HP3456A) at 100V! I’ve got an external precision 100k resistor hooked through some cheap ass alligator leads that are twisted together. It goes from 99.9996 to 100.0001 when the heater clicks off and vice versa when it comes on. I’ve had it running for 24 hours solid now and measured only 300uV drift, which is better than the stated accuracy specs! Not bad for a piece of kit made of all discrete components produced in 1967…
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline MarkPalmer

  • Regular Contributor
  • *
  • Posts: 101
  • Country: us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #361 on: October 30, 2014, 02:41:28 pm »
It's almost scary how stable and accurate they can be.  The later PD precision supplies have the LM399A reference, which is the same zener reference used in the HP 3455A, 3456A, and 3457A multimeters, so as long as the precision resistors in the supply don't drift, they should remain every bit as stable as those 6.5 digit meters. 

I haven’t figured out the whole story, but the original PD company was more than likely bought out.  It isn’t hard to notice that the quality of Power Designs small precision bench supplies went downhill dramatically with models like the 2040A and 1001A.  These were encased in cheap ABS plastic cabinets and had meters, knobs, and binding posts that were, well, horrible to put it lightly.  They also used a, “dicky,” as Mr. Jones would put it, plastic thumb-button assembly to select the output voltage.  This led me to think there was no way the original PD company would put anything like that on the market, because it was in such stark contrast to anything they made in the past.  Looking at the manuals for these later PD supplies, the company is no longer in NY or CA, but listed at 14 Commerce St, Danbury CT- most likely the address of the new owner, thankfully no longer in business making this sort of garbage. They obviously lacked the superb metal working capabilities or any concern to retain the type of quality of the original PD power supplies.

-Mark-
   

« Last Edit: October 30, 2014, 02:43:17 pm by MarkPalmer »
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #362 on: October 31, 2014, 07:15:47 pm »
What do you guys think about designing a replacement for the internals of these supplies? I am thinking of:

- Replacement PCB using an LM399A and other modern parts. Op-amps instead of transistors except for the output drive stage.
- LEDs instead of lamps for indicators
- Reuse the voltage and other switches, transformer, and the chassis.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #363 on: October 31, 2014, 10:46:29 pm »
It's doable for sure. I've been thinking the same thing. Basically you could replicate the 2020B design. Use the existing TO-3 pass transistors. Maybe even replace them with 2N3773's. Hmmm. Maybe a joint project?

That would be a lot better time spent than me re-designing the 2005A amplifier oven. Don't you think?

Edit: Though I might still do the 2005A Amplifier.


Sent from my Smartphone
« Last Edit: November 01, 2014, 01:46:18 am by timb »
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #364 on: November 01, 2014, 11:18:03 am »
Okay. I'm getting pissed now. I replaced VR3 on my 2020B. Along with EVERY FUCKING TRANSISTOR AND ELECTROLYTIC CAPACITOR ON THE FUCKING FUCKER. PLUS U1 and U2.

Same exact shit as before. (I guess I had fried VR3 when I shorted the body of Q1 to ground.)

Not only that, but despite being pin for pin compatible and the specs being the same, the u741 won't replace the CP02-whatever OpAmp. So I've got to locate some NOS on eBay to see if that's my issue.

I'm at a loss for what to do if that doesn't work. It has to be in the feedback loop there somewhere.


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline motocoder

  • Supporter
  • ****
  • Posts: 769
  • Country: us
  • Electrical Engineer
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #365 on: November 01, 2014, 02:53:00 pm »
There is another 2020B up on eBay right now. He claims it is working, but based on his listing I am skeptical. Seller might accept a lowball offer if you want to use it for parts or to compare voltages at different points in the cct
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #366 on: November 01, 2014, 04:53:36 pm »
What do you guys think about designing a replacement for the internals of these supplies? I am thinking of:

- Replacement PCB using ... Op-amps instead of transistors except for the output drive stage.

This is not difficult to do but only the input stages of the error amplifiers matter.  An alternative to replacing them is to use operational amplifiers to correct their offset and low frequency noise and drift.

That would be a lot better time spent than me re-designing the 2005A amplifier oven. Don't you think?

Edit: Though I might still do the 2005A Amplifier.

The 2005A design shows that Power Designs used a dual transistor as the error amplifier and stuck it into the oven to do these very things.  Improve the error amplifier and change the reference and the oven can be left out.

Not only that, but despite being pin for pin compatible and the specs being the same, the u741 won't replace the CP02-whatever OpAmp. So I've got to locate some NOS on eBay to see if that's my issue.

U2 is an Analog Devices OP-05 which is *not* directly replaceable with a 741 unless the nulling circuit which Power Designs used is removed or changed.  The OP-05 specifications are significantly better than the 741 specifications in all respects.  The way I remember it, the OP-05 is an OP-07 without the offset voltage trimming done during production so the OP-07 is a direct and superior replacement.  The later improved replacements for the OP-07 like the OP-77, OP-177, and OP-1177 should also work without changes but the OP-07 is inexpensive.

Other alternatives include the LT1001 and LT1097 but the OP-07 is probably the least expensive option with the same or better performance as the original OP-05.  If you want to use a 741, then disconnect the nulling circuit.
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16272
  • Country: za
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #367 on: November 01, 2014, 05:24:01 pm »
Okay. I'm getting pissed now. I replaced VR3 on my 2020B. Along with EVERY FUCKING TRANSISTOR AND ELECTROLYTIC CAPACITOR ON THE FUCKING FUCKER. PLUS U1 and U2.

Same exact shit as before. (I guess I had fried VR3 when I shorted the body of Q1 to ground.)

Not only that, but despite being pin for pin compatible and the specs being the same, the u741 won't replace the CP02-whatever OpAmp. So I've got to locate some NOS on eBay to see if that's my issue.

I'm at a loss for what to do if that doesn't work. It has to be in the feedback loop there somewhere.


Sent from my Smartphone

Careful on replacing opamps that use the nulling pins. There are 2 ways to use them, and the difference is in which power rail it references to. Use the wrong one and it will instantly blow up the input stage of the opamp.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #368 on: November 01, 2014, 11:43:13 pm »
Damn, yeah you guys are right. For some reason I always thought the 741 was a direct replacement, but turns out it's only pin compatible. *Sighs*

I've got a LT1097 on the way (not very expensive) that should do the job.


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #369 on: November 01, 2014, 11:48:53 pm »
What do you guys think about designing a replacement for the internals of these supplies? I am thinking of:

- Replacement PCB using ... Op-amps instead of transistors except for the output drive stage.

This is not difficult to do but only the input stages of the error amplifiers matter.  An alternative to replacing them is to use operational amplifiers to correct their offset and low frequency noise and drift.

That would be a lot better time spent than me re-designing the 2005A amplifier oven. Don't you think?

Edit: Though I might still do the 2005A Amplifier.

The 2005A design shows that Power Designs used a dual transistor as the error amplifier and stuck it into the oven to do these very things.  Improve the error amplifier and change the reference and the oven can be left out.

Sorry, I should have been more specific. I didn't mean use the 2005A error amp as part of a new control board. In moto's thread I was talking about designing a replacement for the 2005A's entire oven unit using modern SMD components, complete with a PCB based heater, to stick in a small 3D printed case.

The idea being it will fit in the hole left by the removal of the current oven on 2005A units.

An alternative to dealing with broken heaters and thermostats, etc. 


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #370 on: November 02, 2014, 03:22:37 pm »
Sorry, I should have been more specific. I didn't mean use the 2005A error amp as part of a new control board. In moto's thread I was talking about designing a replacement for the 2005A's entire oven unit using modern SMD components, complete with a PCB based heater, to stick in a small 3D printed case.

The idea being it will fit in the hole left by the removal of the current oven on 2005A units.

An alternative to dealing with broken heaters and thermostats, etc.

I think you could get away with a modern precision operational amplifier replacing the differential pair without temperature stabilizing that part.  Power Designs needed the oven not only for their reference but also because the differential pair they used while better than discrete transistors is poor compared to an integrated transistor pair.

Looking at the schematic however, I think the easiest replacement using an operational amplifier to replace Q3 would include bypassing Q8 and Q9 because they add a lot of unnecessary gain making frequency compensation difficult but it should still be possible to leave them in with careful design.

Another alternative I would consider is using almost the same oven board schematic without the oven but replacing Q3 with a modern integrated transistor pair like an Analog Devices MAT01 or SSM2212 or Linear Systems monolithic dual transistor and using the LM399A or reference of your choice.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #371 on: November 03, 2014, 03:40:33 am »

Sorry, I should have been more specific. I didn't mean use the 2005A error amp as part of a new control board. In moto's thread I was talking about designing a replacement for the 2005A's entire oven unit using modern SMD components, complete with a PCB based heater, to stick in a small 3D printed case.

The idea being it will fit in the hole left by the removal of the current oven on 2005A units.

An alternative to dealing with broken heaters and thermostats, etc.

Another alternative I would consider is using almost the same oven board schematic without the oven but replacing Q3 with a modern integrated transistor pair like an Analog Devices MAT01 or SSM2212 or Linear Systems monolithic dual transistor and using the LM399A or reference of your choice.

Yeah, I had already settled on Q3 being a MAT01.

So, you don't think there's any benefit to using PCB traces to keep the entire thing at a constant 50c to minimize drift?

Though I imagine with an LM399 and MAT01, you're going to have drift in other places like the selection resistors etc. so it might not matter.


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #372 on: November 03, 2014, 06:16:26 am »
Another alternative I would consider is using almost the same oven board schematic without the oven but replacing Q3 with a modern integrated transistor pair like an Analog Devices MAT01 or SSM2212 or Linear Systems monolithic dual transistor and using the LM399A or reference of your choice.

Yeah, I had already settled on Q3 being a MAT01.

So, you don't think there's any benefit to using PCB traces to keep the entire thing at a constant 50c to minimize drift?

I doubt there will be any benefit in using an oven if modern low drift parts are used.

Quote
Though I imagine with an LM399 and MAT01, you're going to have drift in other places like the selection resistors etc. so it might not matter.

Exactly.  Calculate the worst case error contributed by the LM399 and MAT-01 and see if it stays within the original specifications.
 

Offline timb

  • Super Contributor
  • ***
  • Posts: 2536
  • Country: us
  • Pretentiously Posting Polysyllabic Prose
    • timb.us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #373 on: November 03, 2014, 11:34:49 am »
The LM399 has built-in heater control, right? The data sheet doesn't mention, is it static off/on control, or simply a series of transistors that operate linearly in response to temperature? (I know current can vary on them, so I assume the later. This isn't in reference to anything specific, just a curiosity I've always had with the 399.)


Sent from my Smartphone
Any sufficiently advanced technology is indistinguishable from magic; e.g., Cheez Whiz, Hot Dogs and RF.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 16545
  • Country: us
  • DavidH
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #374 on: November 03, 2014, 01:20:42 pm »
The LM399 has built-in heater control, right? The data sheet doesn't mention, is it static off/on control, or simply a series of transistors that operate linearly in response to temperature? (I know current can vary on them, so I assume the later. This isn't in reference to anything specific, just a curiosity I've always had with the 399.)

It has a pass transistor and heater control circuit which operate linearly.  The power drawn by the heater circuit is inversely proportional to external temperature as expected and drops to zero at about 85C.  At 25C, it draws about 300 milliwatts.  Operating voltage is from 9 to 40 volts.

The heater and zener allow for roughly an improvement of 5 to 20 times over a precision bandgap reference.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf