Testing a 25 amp PSU at full load

[Phil G4SPZ]

I inherited a Manson 25-amp linear regulated power supply, which turns out to have faulty front panel meters but otherwise appears to work perfectly.

Before trusting my relatively valuable shack full of transceivers to this new supply, I want to run a decent full load test. I lashed up a couple of 0.3 ohm resistors in series, and was able to measure 20 amps flowing and good regulation, but not for very long as the solder melted on one of the resistors. Not surprising I suppose, as each resistor is rated at 25 watts and it was dissipating 180 watts, albeit briefly.

Can anyone suggest a cheap but rugged DC dummy load that will take 300 watts (12 volts at 25 amps) continuously? Half a dozen car headlamp bulbs would work, but I would have to buy them! Maybe a projector lamp, or a swimming pool lamp? As this is only for a one-off test, I don't want to spend too much money.

Thank you for any suggestions.

[Radio Wrangler]

Take an electric fire bar. There's enough wire there to take a kilowatt, but at 240v.

However, one twentieth of the wire length will take 50W and run at 12.0v. So unwind it, measure the length and you want 6 lengths each 5% of the original length.

However, it's probably a 13.8v supply.

240v 1kW so R=240 squared/1000 =57.6 Ohms

13.8v, 25A = 0.552 Ohms say we do it in 10 parallel wires, that is 5.52 Ohms per wire.

5.52/57.6*100= 9.58% of the total length That doesn't leave much for connections

So lets change to 8 parallel wires. So resistance per wire =.552*8=4.416 Ohms

And therefore each wire is 7.666% of the original length.

Total length used is 61% of the original, so that leaves enough over for bolted connections and the dissipation per unit length is still less than the original.

So one spare electric fire bar will do the trick if rearranged

David

[threeseven]

Frustratingly,I have an electronic load that would take that in it's stride but, it's out of commission at the moment awaiting a replacement ammeter! I do have another electronic load but I've not tried it since acquiring it. I'm willing to fire it up and try it on your PSU if you like?

[Bazz4CQJ]

Small diameter (1/16 or 1/8th inch, etc) stainless steel tube works reasonably well; such material is used widely for chemical instrumentation, so should not be too hard to get hold of cheaply. I built a 12V load which I estimated as taking 50A at 12 volt using a coiled length of about 3ft of 1/8th tube; ran at dull a red, but seemed happy to do so.

B

[glowinganode]

In a similar vein, wire coathangers can be used as a makeshift resistance.

Rob.

[turretslug]

As we're only talking 12 (13.8)V here, don't disregard that cheap, conformal heatsink material with excellent thermal capacity (aka tap water). Don't tell the H + S folk but I've used a baking dish with a couple of inches of water to load PSUs, dynamotors and inverters, greatly increasing the dissipation capability of available load resistors. Are the 25W, 0.3 ohm resistors mentioned the aluminium-clad bolt-down type?- my main concern is that the core-to-aluminium interface would be the limiting power transfer factor here, rather than the aluminium-to-water interface. It wouldn't be "continuously", as in for ever but certainly "continuously" as in several hours -rated. Long enough to give the PSU a good cook-out and prove that it was trustworthy when up to temperature. The resistors won't go beyond 100 degrees C- well within the capability of anything with the label "power", provided you don't forget and let it boil dry!

Just keep the proverbial children and animals well away...

[G6Tanuki]

Originally Posted by glowinganode

In a similar vein, wire coathangers can be used as a makeshift resistance.

Likewise the 14-gauge galvanised steel wire used for fence-strainers. I'd suggest using something like a wine-bottle as a former to make nice coils.

And as noted, when dealing with 12V you can use water as a coolant provided you don't mind a bit of electrolysis of the bare wire. Boiling away a bucketful should take some time.

[ex 2 Base]

Where I once worked we tested large DC generators sets using, what was a steel tank and a cast iron plate, the tank was full of caustic soda.The plate was lowered into the tank to increase the load.Maybe you could hook up some thing on a smaller scale.Plastic bucket and two steel plates.Ted

[Phil G4SPZ]

Some excellent suggestions there, many thanks I'm actually quite taken with David's electric fire element idea, but I do have a coil of galvanised garden wire, so I'll measure its resistance then try it, submersed in a bucket of cooling water.

[cmjones01]

I've seen a useful video about this recently from the ever-informative mikeselectricstuff on youtube. It's called 'dummy load in a bucket' for reasons that become clear:

http://www.youtube.com/watch?v=WECW8...re=c4-overview

I'm developing a 24V 30A battery charger at the moment and may well use this technique to load it. Note that it doesn't need resistance wire - fairly thin enamelled copper wire is OK because it's well cooled.

Chris

[cmjones01]

Originally Posted by turretslug

Are the 25W, 0.3 ohm resistors mentioned the aluminium-clad bolt-down type?- my main concern is that the core-to-aluminium interface would be the limiting power transfer factor here, rather than the aluminium-to-water interface.

I've tried abusing those aluminium-clad resistors and they have an annoying habit of exploding when overloaded by more than about 50%, even if they're well cooled. I think it's the limit of the core-to-aluminium interface, as you say.

Much more tolerant are the Welwyn W21/22/23 and so on series of green vitreous enamel wirewound resistors. They will still perform pretty much to specification when glowing bright orange with the glass melting!

Chris

[Ed_Dinning]

Try car headlight bulbs.

Ed

[Nickthedentist]

...or 50W low-voltage halogen downlighter bulbs. Plenty of secondhand ones lying around in peoples' cupboards after they've upgraded to LEDs at the moment from what I've seen at friends'/relatives' house.

Nick.

[Trish]

Originally Posted by Phil G4SPZ

I inherited a Manson 25-amp linear regulated power supply, which turns out to have faulty front panel meters but otherwise appears to work perfectly.
Before trusting my relatively valuable shack full of transceivers to this new supply, I want to run a decent full load test..

I have a Manson EP-925 that I have modified following these changes ---> http://pa0fri.home.xs4all.nl/Diverse...5/ep925eng.htm

My initial experiences of this supply before the modifications were not good.. The fan was noisy. With the amount of heat sinking given the input and output voltages and current levels (even at 18 Amps) left a lot to be desired. In my case it didn't survive under that load for more than ten minutes before the driver transistor decided to go short circuit, which lead to a lot of volts on the output.. The driver transistor is mounted on the same heat sink as the series pass transistors. If I had a rig connected to it, you can be sure that it would have smoked..

So if you do nothing else. I would suggest doing the crowbar addition. I did that and also included a 30 Amp fuse after the capacitor bank and before the crowbar SCR... Also changed the bridge recitfier to a 50 Amp one..

Trish

[M0SOE_Bruce]

Originally Posted by Phil G4SPZ

I inherited a Manson 25-amp linear regulated power supply, which turns out to have faulty front panel meters but otherwise appears to work perfectly. Before trusting my relatively valuable shack full of transceivers to this new supply, I want to run a decent full load test.

I highly recommend load testing. My MyDel MP-125 failed on Sunday. It had worked for over a year on the old xcvr but I guess the new rig put it over the top. Rig has peak current draw of 23A on voice peaks. Fortunately the radio appears to have a OVP sensing circuit that inhibits power to the set when high.

The rated 25A output of the psu should be rated in microseconds.

Bruce

[lesmw0sec]

This reminds me of the time I took a tour of a large transformer manufacturer. They showed me a 6 MW device due for testing. I foolishly asked what they used as a dummy load, forgetting that they would probably require a small town!
With sympathy, the engineer explained that they did the voltage testing and then checked the current capability by shorting the secondary and winding up the primary voltage to achieve the rated secondary current.

[cmjones01]

Originally Posted by lesmw0sec

With sympathy, the engineer explained that they did the voltage testing and then checked the current capability by shorting the secondary and winding up the primary voltage to achieve the rated secondary current.

I'd like to see the variac they use for that!

[lesmw0sec]

[I really can't remember what their input source was. Trouble is, it was the GEC factory in Brisbane, which is a rather long drive from here

Les.

[Silicon]

When the dip beam filament of my car headlamp bulbs burns out I keep it. The other main beam filament is still good and can be used as a 12v load.

Garages and Halfords must bin loads of these. Do you know and friendly mechanics?

[kalee20]

Car light bulbs would be excellent (I like things that light up. You can see something is happening).

Steel wire - coathangers / fence wire - also good. Yes - dunk it in water. At 12V, electrolysis will be slow and not too much of a nuisance. As things heat up, it does become more obvious.