EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: sureshot on May 16, 2018, 10:00:45 am
-
I'm looking at putting together a high current resistive load tester for high current power supplys. Typically converted server psu, but also linear power supplys. I've used 12 Volt halogen bulbs up until know. But there very inconvenient to set up, and need a lot of room. The light is a problem as well over long periods of testing. I'm going to be using high power resistors for this unit i planning, wanting a meter or meters. I could use analogue panel meter which are fine really, but i would like a digital read out display.
My question is the source of power to run a digital meter, I've got my eyes on a meter that can read Volts Amps and power in Watts. I really wanted the power supply under test to run the meter whilst under load from up to a maximum of 6 x power resistors. Although they will be switched from 1 x to 6 x depeding on load needed to test.
As i wanted the meter to be powered from the psu under test, the meters power draw I'm not taking into consideration, as this is a basic high current test. And resolution is not critical. Do you think it possible to run the power panel meter off the power supply under test ? It would be handy if i didn't have to plug in any additional items like the resistive tester i plan to put together. I've posted pictures of the meter I'm wanting to use.
Thank you for reading, any help appreciated.
-
Yes, will be no issue (as long as you don’t need to test down to 3.3 volt or 5 volt supplies).
Basic 12V tests? Fine.
-
So this meter will power ok from the psu I'm testing, which would be under load. What i mean is do i need a seperate power source for the meter ? I've read some meters can be fussy using the same power source that powers the load. It would be much easier if the meter powered from the psu under test. So will it measure its own load current is what I'm getting at.
Thank you for your reply.
-
So this meter will power ok from the psu I'm testing, which would be under load. What i mean is do i need a seperate power source for the meter ? I've read some meters can be fussy using the same power source that powers the load. It would be much easier if the meter powered from the psu under test. So will it measure its own load current is what I'm getting at.
I'm making a few assumptions here, but they are pretty damn safe assumptions.
The power to run the meter itself comes in on the red and blue thin wires. The current measurement is on the yellow and green (there is a small voltage drop on the shunt resistor, which is sensed by the yellow and green wires voltage difference). Accordingly, the small amount of power the meter itself uses is not measured.
I'm fairly sure that if you moved the blue connection from the black terminal as shown over to the green terminal, that the meter would still work and it would then measure the total load on the PSU (the actual load and the power used by the meter). The downside here is that the yellow signal would be below ground reference, and so might not actually work.
(Again, a few assumptions underlying the above, but I think the assumptions are all reasonable.)
However, you are looking at a meter that draws a vanishingly small amount of power as compared to your load. I suggest you ignore it (assume it's zero and don't try to measure it).
-
And damn it, while I gave you the complex answer, I didn't repeat the simple answer:
Yes, it will work to be powered off the PSU under test, certainly with you using pure resistive loads to test. (Loads that put a lot of noise back onto the supply might screw up the meter, which you'd want to address with some buffering capacitors, but a pure resistive load [which a lamp or power resistors are] will be no issue.)
-
I use that meter, but with internal shunt. I think they are the same.
One warning is that they may loose calibration if you leave the load connected and remove the power supply. Both of mine are now off by ~50% because I had them connected to solar panel as a source, and the sun went down...
-
Thank you for the replys, yes I get it I think..
The unit for testing would either be all on (varying load 50 - 300 watts) or completely off. It really is just to replace halogen bulbs as a load source. The light and heat of 300 watts of 50 watt lamps is crazy.
I'm using 50 watt power resistors of 3.9 ohms, so a bit of flexibility between 12 Volt to 14 Volt power supply. But the majority will be testing different converted server power supply.
It's not needed to be critically accurate, Just for testing heavy loads.
I've used analouge meters in previous projects, but thought it might be cool to use a digital power meter. It's a large enclosure with two 125 watt rated cpu heatsinks and fans mounted in the top. Resistors 3 x on each base of the heatsink, so 6 resistors total. I wanted a bit more load really, 500 watt would have been great. But the size of the unit would have been huge, not to mention dumping 500 watts of resistive heat.
Thanks again for the help, at the moment it's something in progress, but getting there slowly.