POWER DESIGNS PRECISION POWER SUPPLIES

[idpromnut]

I'm pretty sure that this thread was to blame, but here they are, and are my two only power supplies that I use every night on the bench:

[robrenz]

Glad we could help

[don.r]

I'm pretty sure that this thread was to blame, but here they are, and are my two only power supplies that I use every night on the bench:

Is that a 2225 above the 6050's?

[uoficowboy]

I have noticed PD supplies going cheaply recently. Probably because this thread hadn't been bumped up for a while. There's a 5020 for $110 shipped right now. I love my 5020, but don't need another! Somebody should grab it!

[saturation]

Its come a long way, but it has become more attractive, I think because of this and similar discussions.  As one of the early starters back when, my 5020 was ~$50 delivered, I have 4 of these types of supplies  .

[idpromnut]

I'm pretty sure that this thread was to blame, but here they are, and are my two only power supplies that I use every night on the bench:

Is that a 2225 above the 6050's?

Yes sir it is!  I hold Dave responsible, but honestly, I am super super glad I started with the 2225... learned a lot about what all the controls are actually doing!

[don.r]

I'm pretty sure that this thread was to blame, but here they are, and are my two only power supplies that I use every night on the bench:

Is that a 2225 above the 6050's?

Yes sir it is!  I hold Dave responsible, but honestly, I am super super glad I started with the 2225... learned a lot about what all the controls are actually doing!

LOL! I love my 2225. 500uV/div is impossible to find in another analog scope.

[robrenz]

LOL! I love my 2225. 500uV/div is impossible to find in another analog scope.

You missed post #142 of this thread here   10µV/div is possible

[don.r]

LOL! I love my 2225. 500uV/div is impossible to find in another analog scope.

You missed post #142 of this thread here   10µV/div is possible

LOL! Yeah... maybe I should say VERY difficult... unless you happen upon a $35 diff amp and the mainframe to plug it into.

[David Hess]

LOL! I love my 2225. 500uV/div is impossible to find in another analog scope.

I often end up recommending or at least mentioning the Tektronix 2225 just because its 500uV per division input sensitivity.

The rarer but available Tektronix 2211 and 2214 combination analog and digital storage oscilloscopes also go down to 500uV per division.

You missed post #142 of this thread here   10µV/div is possible

While I have a couple of 7A22 plug-ins for my 7000 series mainframes, they are tricky to use at their most sensitive setting because their drift and low frequency noise performance is not as good as it could be.  There is an equivalent plug-in for their 5000 series oscilloscope mainframes and also the stand alone AM502 which can be used with any oscilloscope.

You can get better results with a dedicated low noise amplifier but there is an easier way; if you have a DVM with the required sensitivity and noise performance that will operate in sample mode, then the standard deviation of the DC voltage readings is a good way to measure the RMS value of the low frequency noise.

[David Hess]

And I forgot about the 7A15A plug-in for the 7000 series which goes down to 500uV per division and is single ended.  They are more common, less expensive, and easier to maintain than the 7A22.

[robrenz]

What is the relationship of sample rate to what frequency band the SD represents.  Example my 8846A at 6.5 digits would give a 6Hz sample rate. Would that mean the highest frequency component I am measuring is 0.6Hz (nyquist)?

[David Hess]

The low frequency limit depends on the total period of time over which you gather samples.  You will not see noise down to 0.1 Hz if you only collect samples for a period of 1 second.

The high frequency limit depends on the sampling gate duration and not the sampling rate.  In practice for a high resolution DC voltmeter, this will be some multiple of the line frequency resulting in a best case first null at 50 or 60 Hz and the frequency response will follow the sinc function.  The bandwidth is limited by the sampling gate width and not the sampling rate in this case.  Sampling RF voltmeters work this way and while aliasing occurs, the histogram of the signal is the same so the peak to peak, RMS, and average measurements are the same.  Conveniently the standard deviation is equal the AC RMS value.

[robrenz]

Yet another way to measure ripple at specific frequencies in these two posts

[iampoor]

Are there any power designs supplys with a negative rail? I would love a nice high precision +/- supply. 

[PedroDaGr8]

Are there any power designs supplys with a negative rail? I would love a nice high precision +/- supply. 

Two HP Precision 611x series can be wired to run in series mode. According to the manual, simply ground the "center tap"  to create a +/- precision supply.

Sent from my LGLS990 using Tapatalk

[David Hess]

Are there any power designs supplys with a negative rail? I would love a nice high precision +/- supply. 

Power supplies are generally not precision devices but many include two symmetrical outputs which may be connected in series or parallel.  More common are bipolar power supplies which may include tracking like the HP 6236, Power Designs TP325, TP340, and TP343, and the Tektronix PS503.

[albert22]

Hi. I have been trying to understand how the decade switch works on the PD 2005 schematics. It is common to use 1,2,2,5 values for a decade but I cannot figure out what is the function of the 4.7, 6.8, 27  resistors.  Also, it seems that in some cases they are paralleling .1% ww resistors with 10% carbon ones.
Please. Can someone explain this?
Thanks and Regards

[timb]

Anyone have a 2010 or 2020 they're looking to get rid of? eBay has been pretty dry lately. (I've got an offer in on a 2005.)


Sent from my Smartphone

[timb]

Just after posting the above I found a 2020B buried away on eBay! Snagged it for $75, shipped. The seller says it goes into overload as soon as it’s switched on; I’m thinking it could be as simple as one of the sense or remote straps not being connected on the back, but we’ll see. Either way, it’ll be fun to fix!

I also grabbed a 2005A for $65+Shipping. Seller says it works fine. Turns out he had two available and listed both as 2005P; in actuality one is a 2005A and the other (with non-working lamps) is a 2005P. The P model is the Remote Programmable version, which appears to only have controls on the front for setting a voltage and current limit plus zeroing. You can see the 2005P in the first picture but the 2005A in the second and third. So if anyone wants a 2005P at a decent price, here you go: http://r.ebay.com/tD7RwT

(The 2005A only allows remote voltage setting, the 2005P can set current and voltage remotely, like the 2020Bs.)

By the way, I’m working on a simple USB adapter for the 2020B/2005P to allow remote control of voltage/current via SCPI commands over serial. If anyone is interested or has ideas, let me know!

[idpromnut]

By the way, I’m working on a simple USB adapter for the 2020B/2005P to allow remote control of voltage/current via SCPI commands over serial. If anyone is interested or has ideas, let me know!

Neat!  I have a project in the back of the draw to do something similar for the 6050, 6050C and 6050D.

[timb]

By the way, I’m working on a simple USB adapter for the 2020B/2005P to allow remote control of voltage/current via SCPI commands over serial. If anyone is interested or has ideas, let me know!

Neat!  I have a project in the back of the draw to do something similar for the 6050, 6050C and 6050D.

Do those use the same 1k/V and 1k/A formula as the 20xx Series?

[robrenz]

You may want to consider making it also work on the very popular HP611X series supplies.
manual link here - remote programming starts at section 3-23

[idpromnut]

By the way, I’m working on a simple USB adapter for the 2020B/2005P to allow remote control of voltage/current via SCPI commands over serial. If anyone is interested or has ideas, let me know!

Neat!  I have a project in the back of the draw to do something similar for the 6050, 6050C and 6050D.

Do those use the same 1k/V and 1k/A formula as the 20xx Series?

I wish!  The principle is the same, a resistance-controlled feedback loop, but the ratios are different. In fact, for the 6050 (which has two voltage ranges, 0-6 and 0-60) the same remote programming loop has two different ratios depending on the selected voltage range.

Originally I was going to use some sort of e-Pot to control the voltage, but I realized quickly that you need an e-pot that can handle 0-60V (for the supplies I have) so that was out due to cost. The other option (which is sorta where I left it) was to use a MOSFET in its linear range as a voltage controlled resistor. But that still requires me learning more on how to do that

[linux-works]

here's an idea, maybe.

one of my projects that I designed was a relay attenuator (for audio use).  its a little like a pot but uses relays and resistors.  put voltage in and get a percent of it, out.  on my board, its 8 relays at half db steps from -128db to 0 db (again, this was meant for audio use as a cpu based stepped atten).  the resistors could be changed to be more linear instead of log and this could be a nice voltage divider that is arduino controlled.

then, you just need a clean stable voltage source and put that on the input of the atten board.

the pdi boxes, iirc, are voltage controlled so having a programmable voltage send to the rear panel for control would do the trick.

search for 'lcduino' and 'delta1' (amb.org is the site) and you'll see the project I'm referring to.

see if that might be a way forward for this.  you can change the step size and range just via the series and parallel resistors that you pick.

then, you simply talk to the arduino over serial and it will translate to the relay bank values.