Power Supply Testing

[Chet T16]

What's the standard run of things I should test on my Atten power supply that may be of interest to people?

Using the Rigol in single shot i captured it powering up (no switchable outputs) Should this be done with a load or without or both? Not that I have anything other than 1/4w resistors here atm.
Voltages? Max/min/middle/shouldn't matter?

What else?

Also, is there a way to get the Rigol to measure risetime in single shot mode?

[Simon]

set you scope to single shot with the trigger voltage low like <1V and set an appropriate time scale and set the trigger on a rising edge, then power up and enjoy the show

[bfritz]

Oscilloscope shot of power up, and power down.

I like to get scope shots of the voltage response for a step load change from minimum load to maximum load, and then from maximum to minimum.  This lets you see a lot regarding stability of the control loop of the supply.  I've seen some inexpensive power supplies ring and oscillate quite badly.  If you have a load that has multiple settings you can change between, that is great.  If not, I usually use a fairly high valued resistor, and put a constant current load in parallel.  I use one channel to monitor the voltage output, and then use one or two channels to monitor the current of the load or loads.  I then usually use the math functions on the scope to convert the voltage across the resistor and load sense R to currents, and add them together.

[BravoV]

Once you've tested on full load to see the regulation and ripple or others like above posts, suggesting to give it hell to see how it's dynamic response by jolting it using this -> http://www.ko4bb.com/Test_Equipment/DynamicLoad/

Built it while ago, used it a lot on psus, and it pretty easy and cheap too, this will quickly give you an idea how good your psu on it's load transient response, over/under shoot recovery/performance, ringing, regulation loop stability and etc.

Usually toy psu will not have a chance under this kind of stress and a good psu will really stand out among the crowd.

[Zero999]

Awhile ago I built a bipolar LM317/LM337 PSU.

I tested the transient response with a MOSFET switching a load resistor with the gate being driven by a signal generator.

[olsenn]

I doubt you'll need a professional DC electronic load, but if it is something you're interested in, it may be worth your while checking out the IT8511 by Itech (it's made by BK Precision).

[Berni]

Yeah step response is where a good PSU will show and you don't really need to use a special electronic load, you can give it a try by taking some large ceramic power resistors and a switch to suddenly put the load on (Could also touch the wires but its not as clean) And you should see a spike in the voltage as you do it, smaller the spike the better.

I saw some PSUs that have there voltage pretty much plumit to near 0 for some time when doing that. I even had a case of where a PLC turned on a 10W light bulb on its own 24V rail and every time it did that the PSU knelled down so much the PLC reset it self only to attempt to turn it on again and again.(Oh and i think it was even rated for 5A so a pretty big supply)

[McMonster]

I saw some PSUs that have there voltage pretty much plumit to near 0 for some time when doing that. I even had a case of where a PLC turned on a 10W light bulb on its own 24V rail and every time it did that the PSU knelled down so much the PLC reset it self only to attempt to turn it on again and again.(Oh and i think it was even rated for 5A so a pretty big supply)

I'm a tech at the university lab with PLCs and I have almost the same thing happening from time to time. In fact I'm sitting there right now. The problem is that we use cheap unmodified ATX power supplies to power students' projects (PLC controlled small models, like a ball sorter machine) and those PSUs are very unreliable. I've been asking to buy proper supplies for some time but with no effect. I've also had some problems with electrical installation in this room that killed 5 PSUs, but they fixed it already.

[PStevenson]

set you scope to single shot with the trigger voltage low like <1V and set an appropriate time scale and set the trigger on a rising edge, then power up and enjoy the show

I'm addicted to doing this, I have the ATTEN ADS1102CAL and I pass hours just plugging PSUs into it to see the rise voltage. haha

[Chet T16]

I've ordered a couple of high power resistors so i'm waiting for those to come

The rigol wasn't showing the rise time when in single shot mode when i tried. Between exams and christmas in a different country there hasn't been much experimentation time!

I'll post up the results of what i get, thanks for the replies

[PStevenson]

I had a few issues at first trying to figure it out. the Chinglish manuals don't help, but if you spend half an hour experimenting with it you'll get there