Author Topic: POWER DESIGNS PRECISION POWER SUPPLIES  (Read 330215 times)

0 Members and 1 Guest are viewing this topic.

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #125 on: September 05, 2012, 01:14:22 am »
Also what kind of tests can you recommend I do with my pathetic equipment (4000 counts Chinese DMM) and no oscilloscope (yet!) before I send it?

I have 5 of these and I have checked and calibrated them with my In cal 6.5 digit 24ppm meter.
My suggestions after my experience with them:
Don't have it calibrated unless you have a darn good reason , It is probably more accurate as is, than your meter. 
Don't worry about ripple,  its rated at 100 microvolt P-P, unless you have a very good differential probe setup and a scope with 500 microV per div you are not going to be able to measure it. You are going to see all kind of noise instead.
Do invest in some Caig de-oxit and thoroughly clean the rotary switch contacts.  They will give a lot of error if not in good condition.
The calibration is very easy and just requires a accurate meter. 
Save your money for a more accurate meter and then calibrate it yourself.

PS,  Forget about drift after warmup. Unless you have a darn good bench meter your meter is going to drift more than this supply.
« Last Edit: September 05, 2012, 01:19:10 am by robrenz »
 

Offline Aldobrandi

  • Regular Contributor
  • *
  • Posts: 61
  • Country: us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #126 on: September 05, 2012, 02:30:13 am »
Thanks a lot T4P and robrenz for your suggestions. I did run for a few hours and was not able to measure any drift what so ever. I think time has come to complete my lab with a decent meter and a scope. I am so excited about this supply, I had a good hour and a half of looking at it under all possible angles, the construction is superb.

Also on a side note, do you recommend that I purchase a new meter in the $100-$200 range, or should I buy something better on ebay but then take the risk that it won't be calibrated either ? I am sure I can find used and calibrated 87-III but are these worth the $100-$150 they usually go for or should I spend a little more for an 87-V ?

Thanks again and sorry for momentarily hijacking the thread with my DMM question!
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #127 on: September 05, 2012, 03:27:23 am »
This is a do as I say, not as I do moment. :-[  Don't get caught up in high precision until you need it or you can buy it realy cheap. Get a decent scope and start building things.  Then buy things you actually find yourself in need of.

Offline Aldobrandi

  • Regular Contributor
  • *
  • Posts: 61
  • Country: us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #128 on: September 05, 2012, 03:44:34 am »
Got it, thanks for the kind warning  :D I know this is a slippery slope and watching your amazing videos believe me I'm this close to buying a microohm meter or a 6.5 bench meter. Haha but I'll be reasonable. Thanks again!
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3697
  • Country: sg
    • T4P
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #129 on: September 05, 2012, 03:50:46 am »
You can try out a Rigol DM3068, very versatile that thing

As for handhelds i assume you would take a look outside of flukes right?
For a start, the Agilent U1252B and the Sanwa PC7000 ( also known as the Brymen BM867 ) or if you do not want to spend that much well look at a Mastech MS8218 or an UT61E
 

Offline Aldobrandi

  • Regular Contributor
  • *
  • Posts: 61
  • Country: us
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #130 on: September 05, 2012, 06:54:48 pm »
I'm definitely open to any good meter. I guess I have the Fluke reflex like many but I have been eyeing the Agilents for a bit of time now and although I don't think I'm ready to spend the money for the U1252B I think it's gonna be a showdown between the U1272A (love the data logging capacities and the other nice features like the Auto Hold) or more likely the more humble U1242B. Once again I really appreciate the suggestions!
 

Offline linux-works

  • Super Contributor
  • ***
  • Posts: 1993
  • Country: us
    • netstuff
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #131 on: September 19, 2012, 04:13:11 pm »
I can't seem to resist this style of psu.  just bought my 4th one, this week (the 5020):



the others are 2020, 2005 and 2010.  one is oven based and the rest are lm399 (I'm guessing) based.

the 5020 was a deal at $100.  could not say no ;)

a friend asked me why I need so many.  and I don't -need- so many.  I just wanted them.

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #132 on: September 19, 2012, 06:24:37 pm »
NICE!  They are magnificent supplies, when I got the bug I got 3 in a short time span.  Enjoy them.  They photos look like they are in very good shape.  Did you have any problems with the fine adjust knob as myself and robrenz had?
Best Wishes,

 Saturation
 

Offline linux-works

  • Super Contributor
  • ***
  • Posts: 1993
  • Country: us
    • netstuff
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #133 on: September 23, 2012, 11:42:51 pm »
I have not read the whole thread; what problems with the knob are there?  which model?

they all came pretty dirty and I spent a good 1-2 hours on each one, fully breaking it down and cleaning as much of it as I could.  the 2005 is the oldest and I replaced all the electros inside.  the others I have not gotton to, other than to use deoxit on the contacts.

the top one (5020) is still quite dirty as I have not even taken any cleaners to it or even removed the top cover yet.

once I deox'd the contacts, I saw accurate steps on the click knobs.  amazing that that, alone, seems to get 95% of the job done (or more).

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #134 on: September 24, 2012, 12:07:46 am »
Any, but you recognized the disease and have the treatment already, deoxit, it will get 98% of the issues, the next 1% is recalibrating the dials by turning the adjust screws correctly and the last 1% is cleaning the chassis  :D!  We've had dirt on the fine adjust pots, all it needs too is dexoit.  So we have all had the same issues. 

I have not read the whole thread; what problems with the knob are there?  which model?...

once I deox'd the contacts, I saw accurate steps on the click knobs.  amazing that that, alone, seems to get 95% of the job done (or more).

Best Wishes,

 Saturation
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #135 on: September 26, 2012, 05:45:47 pm »
I suggest compressed air first if you have it to remove all dust that will blow off. Then a thourough wash with a general electronic cleaner to get the general gunk off the switch assemblies and the detents.  The detents will probably have caked old grease in the ball pockets. Then apply deoxit D100L to all the contact fingers and thoroughly cycle the switches. Re lube all the places where the switch shaft engages other parts and grease the detent assemblies. I personally like wash the D100L off and then lubricate with Deoxit G100L.

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #136 on: October 12, 2012, 07:00:42 pm »
Just a bump with a quick test for output ripple.  i got a Preamble 1855 differential amplifier from fleabay and thought I would give it a quick try on the output ripple on a PD 2010.  This was more to see if the diff amp would let me actually see the ripple without all the ambient noise. I was testing at 3mV unloaded.  The precision offset voltage generator in the 1855 allows DC coupling while measuring the ripple.

8846A AC reading at 10NPLC and 200Hz filter. Secondary reading is the DC component. Darn, I am going to have to recalibrate the 2010 its 0.4 uV off  ::)  12.7uV RMS = 36uV PP according to the 8846A.


Both pictures below are this setup: 1855 setup with dc coupling, 10X gain, 100kHZ BW limit, 1M input resistance, 3mV differential offset, 1x probes. Tek 2225 setup with DC coupling 500uV/div, 0.1uS/div, PP auto trigger, LF reject.  This gives 50uV/div on the scope display.

This shot is with the two probe hooks connected and the two standard ground leads clipped to each other.  I would consider that a clean 50uV noise floor.


This shot is both probe hooks removed and the probe tips on the output of the PD 2010.  If you visually subtract out the 50uV noise floor I would say it is below the 100uV PP ripple spec.  I know, not very scientific, like I said I wanted to try the diff amp.  ;D
« Last Edit: March 01, 2014, 01:47:35 am by robrenz »
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #137 on: October 12, 2012, 07:57:20 pm »
That's wonderful robrenz, I don't have a probe to get that much resolution, so its good to see it.  They clearly don't make them like they used too!

For readers interested, quite a bit of these precision PSU pop up on eBay weekly. 
Best Wishes,

 Saturation
 

Offline miciotta

  • Newbie
  • Posts: 5
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #138 on: October 15, 2012, 05:48:01 pm »

hi to all....

sorry but you that are a expert, i am beginner...

what power supply regolable at nice price i can buy ?

i see the  PeakTech 6140  0-30 volt  0-5 ampere is good ?

or for 80 - 120 euro what i can buy  ? LINEAR not switching ?


thank...... kiss from mery
 

Offline MBY

  • Regular Contributor
  • *
  • Posts: 184
  • Country: se
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #139 on: October 22, 2012, 08:37:59 pm »
I'm unsure about which thread to post in, but my replacement oven project is coming around (see this thread about that project) and I'm back to benching the PD2005. Therefore, I post in this thread now.

Anyhow, when the temperature is regulated well within one degree C, I'm a little bit disappointed about the stability. With a setting of 9.986 (uncalibrated) I get as close to 10.000 V as I can, I logged the stability for a few hours and get a drift (up and down, up and down, haven't established the cycle period yet) of 140 µV. While not terrible, I think its almost an order of magnitude more than I hoped for. Coincidentally, I think my thermal regulation is about one order of magnitude better than the old thermostat control (I experimented and found it to have a off-transition at 70 degrees C and an on-transitions at 65 degrees), about 0.5 degrees C instead of 5 degrees. (Actually, my external thermometer with its K-sensor bonded together with the NTCs I'm using to control things indicates even better control, better than the resolution of the thermometer of 0.1 degrees, but I know that the PD2005's electronics itself generating some non-negligible heat, creating hot spots).

Interestingly, the stability seems to be improved by a higher set point. Thats not terribly surprising though, but I did not expect a steep improvement. More to come, I'm have to do some more exhaustive tests and take notes. Now its much "gut-feeling", and that is not very scientific... :)
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #140 on: October 25, 2012, 03:38:40 pm »
Found this just looking around and they shows many power designs parts as available.  Looks like the kind of place that will charge 10X the reasonable price.  I don't need any parts at the moment, just thought it might help someone.

http://www.justnsnparts.com/manufacturer/power-designs-inc.aspx

Offline MBY

  • Regular Contributor
  • *
  • Posts: 184
  • Country: se
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #141 on: December 02, 2012, 06:34:20 pm »
Now I have done further tests. The oscillations of the output in CC mode is not due to the oven thermostat as I still see the oscillations of the same duration and amplitude as before. One other thing I have noticed is that the CC circuit is really slow. I haven't noticed it before since the DMMs have a certain update speed. But checking on a scope it becomes clear that the rise time from CC mode to CV mode is very slow, in the order of hundreds of ms. But not always, its hard to reproduce. Easiest seen on high current outputs, like 0.4-0.5 A.

I think this is connected to the oscillations but I don't know how. It may be the input fluctuating somehow and the CC circuits slow response causing it.

Quite annoying. The CC stability is only a few mA (well, thats quite good, but I expected better), the CV stability is at my current oven settings about 10 µV.

Edit: Some data:
Setting 10.000V, CC=500mA
Rise time (disconnect of a 6.8R resistor): > 250 mS to 90%, >500 mS to 10.000V
Fall time (reconnect 6.8R):  25 mS to stable output.

Those figures are actually piss-poor! Well, the unit is not designed as a fast CC limiter, but this slowness affects the total stability of the CC mode, I'm sure!
« Last Edit: December 02, 2012, 06:44:14 pm by MBY »
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #142 on: December 06, 2012, 07:47:55 pm »
This post/thread gave a good opportunity to try out my $35.00 high performance differential amplifier on power supply ripple. For continuity I am posting it here. Yes it is a little bulky and takes up quite a bit of space but the performance is eye watering. OK I had to buy a $150.00 scope to plug it into also so it really cost $185.00 But these scopes have a signal out on the back so this whole setup can really be a high performance differential preamp that feeds a digital scope. :-+

Tek 7603 scope with 7A22 differential amp in the middle hooked to a PD 2005A with 2 bnc to banana cables. The banana cables were less noisy than 1X probes even with a direct probe tip to probe tip adapter that also couples the probe grounds right at the tip.


$35.00 Tek 7A22 differential amp 10µV/div 100000:1 CMRR. Separate upper and lower cutoff filters DC offset capability 1MHz max.BW AC coupled in this case.


PS hookup with 10 ohm resistor and set at 1V.  Cable grounds connected to each other and the PS ground. Red leads are the differential signal leads


Noise floor @ 10µV/div, inputs grounded internally, maybe 5µVpp.    Analog beauty :P


Noise floor @ 10µV/div, cables connected as shown in PS picture above except the two red signal leads are plugged into each other only to see noise floor of setup.


Lower filter is dc upper filter is 3 kHz I would call that a clean 40µVpp ripple. PD 2005A spec is <100µVpp  This is one complete cycle of a repeating pattern not just what I happened to trigger on. It is also line frequency based since ac triggering on line gave a solid trigger. Also the period is .0166 seconds = 60Hz


Lower filter is dc upper filter is 300 kHz.  With higher frequencies added I would still call this a easy 60µVpp ripple without even subtracting out the baseline setup noise/ripple of 10µVpp
« Last Edit: March 01, 2014, 01:28:08 am by robrenz »
 
The following users thanked this post: KK6IL

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #143 on: December 09, 2012, 04:28:44 pm »
10uV/div?  :-+ :-+ :-+ :-+ :-+ :scared:

I guess I know what my next purchase will be.  This a great amp for DC measurements, with some AC on it.  I need to get a differential amp like that!  Low cost differential probes from Testec, or Tie Pie [ BTW they seem to be the OEM for many folks like Yokogawa or old Lecroy] have noise rated in the 100s of uV and can't begin to measure the PD noise floor, and yes the probes alone used cost more than your whole setup!

The most I've done is connect it to my 3456a, and estimate DC variation vs when the meter is shorted, to remove the LSD drift.  For AC all I can muster is using the 3456a AC mode to measure the ripple as RMS. :palm:

That is one kick ass microvolt input scope!  Few can match it today, most top scopes do 500uV/div as minimum, if at all.  Not to mention the dynamic range, 10uV to 10V /div, it took a while to look at that range before I realized it was true!
Best Wishes,

 Saturation
 

alm

  • Guest
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #144 on: December 09, 2012, 06:14:35 pm »
This has come up before, for example when Dave was so impressed by the Tek 2225 that went down to 500 uV/div.

The modern equivalent would be the Tek ADA400(A), which requires a TekScope interface (power pins around the BNC connectors) or $$$ external power supply. Ever wonder why people complain about limited probe choice for the cheap scope brands? ;)

Another alternative would be the Tek AM502 differential amplifier, it is essentially the same circuit as the 7A22, but as stand-alone amplifier with BNC output. Gain is 100-100k with a switchable /100 attenuator. Bandwidth is 1 MHz with switchable high and low pass filter and DC offset, just like the 7A22. It merely requires a TM500 series power supply (eg. TM501), which is much smaller and lighter than a 7000 series scope. In both cases, you can either use 1x probes or special 10x probes as input (eg. Tek P6055). Normal 10x probes have a fair amount of tolerance in attenuation factor. Just 1% difference in attenuation will reduce the CMRR from 10,000:1 to 100:1.

There is currently one on eBay in as-is/dont-know-how-to-test condition for $100, but it looks modified. There appears to be a custom plate around the gain knob and the /100 attenuation button is missing.
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #145 on: December 09, 2012, 07:06:09 pm »
Thanks a bunch for this alm  :-+, your deep insight and concise sum saves me a bunch of legwork for future eBay hunts.  Checking just now, some cost more than a mainframe +7A22 combined, but with patience, does one bag elusive prey. 

This has come up before, for example when Dave was so impressed by the Tek 2225 that went down to 500 uV/div....... snip .......
There is currently one on eBay in as-is/dont-know-how-to-test condition for $100, but it looks modified. There appears to be a custom plate around the gain knob and the /100 attenuation button is missing.
« Last Edit: December 10, 2012, 01:15:26 am by saturation »
Best Wishes,

 Saturation
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16272
  • Country: za
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #146 on: December 09, 2012, 07:11:20 pm »
I have the 5 series version of that scope. The 2MHz differential plugin is great as well.
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #147 on: December 09, 2012, 07:35:02 pm »
I will make up a set of leads per the HP power supply handbook for the noise spike measurement and post results also. I have a 7A13 diff amp to do the 20MHz BW testing. That one was expensive $50.00  ::) it has a interesting infinite impedance mode for 50mV/div and below.

I will also make a more ideal short set of bnc/coax leads specifically for the PARD measurement.  I should have had the red leads in the PS with the PS off to measure the noise floor of the connections instead of just connected to each other. Any part of the de-energized PS that is picking up external noise that the CMR is not rejecting would show up.  We will see how much the test setup can be improved.

alm

  • Guest
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #148 on: December 09, 2012, 08:08:05 pm »
Yep, there is also one for the 5000 series scope (5A22N?), and a slightly more limited version with a higher bandwidth. The 5000 series is much less popular, but sometimes also quite cheap because of their modest bandwidth. Not sure if they have signal out. The plugins look similar to the TM500 plugins, but they are not compatible (different pinout and keying).

The AM502 is not necessarily cheaper than the 7A22, but the mainframe is much more compact. It also has a full quality output, better than the signal out from 7000 series mainframes. I think I got a nice looking one for $70 or so. For old test equipment, there's almost always a test equipment vendor offering (usually not selling) one for a BiN price about 5-10x of what is reasonable in as-is condition.

The 7A13 is the other very interesting 7000 series plugin for which Tek did not make a stand-alone version. It has a large and accurate DC offset range that is very useful for signals superimposed on DC voltages. It also has an excellent overdrive recovery. Sensitivity is much lower than the 7A22/AM502, but bandwidth is much larger. The early mechanical version often had problems with cracked gears, but I'm sure that would not be a problem for someone with the mechanical knowledge of robrenz. I believe Preamble labs hired the 7A13 designer and developed a similar product (DA185x?). Don't expect to find it for $50.
 

Offline robrenz

  • Super Contributor
  • ***
  • Posts: 3035
  • Country: us
  • Real Machinist, Wannabe EE
Re: POWER DESIGNS PRECISION POWER SUPPLIES
« Reply #149 on: December 09, 2012, 08:41:21 pm »
The 7A13 is the other very interesting 7000 series plugin for which Tek did not make a stand-alone version. It has a large and accurate DC offset range that is very useful for signals superimposed on DC voltages. It also has an excellent overdrive recovery. Sensitivity is much lower than the 7A22/AM502, but bandwidth is much larger. The early mechanical version often had problems with cracked gears, but I'm sure that would not be a problem for someone with the mechanical knowledge of robrenz. I believe Preamble labs hired the 7A13 designer and developed a similar product (DA185x?). Don't expect to find it for $50.

I have the late version 7A13 that has a mini LED display of offset voltage instead of the mechanical. I like that it has the offset voltage available on the front to monitor with a more accurate meter.

I also have a the Preamble 1855 $350.00 and the DXC-100  10X-100X differential probe box for it. It was $99.00 but had no probes so I got a set of replacement probes (service parts, they have to be soldered in)from LeCroy complete with all accessories for $70.00  This is good for 100MHz. When you read the calibration procedure for the DXC-100 you understand why maintaining CMMR with attenuating probes is difficult. There are 8 different pots/tuning caps per channel (16 total) in the probe box that need to be tweaked.   Those prices are way lower than what these items typically go for. (It pays to make stupid low offers)

The Preamble has a 100M ohm input mode. Do you guys think the 7A13 "infinite impedance" mode is higher impedance than the 100M mode of the Preamble?


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf