Choosing a dummy load for bench testing

[feds27]

I was reading various threads where dummy loads are mentioned for bench testing but it's not clear to me what load I should use for my amp. It sounds like I should match the dummy load to my amp, but, double the wattage. So if my amp is rated 100W into 8Ohms, it sounds like I should use an 8Ohm 200W dummy load.

I also noticed people recommend bench testing at 8Ohms and someone said he uses 2 - 4Ohm 200W resistors per channel. Am I correct in guessing he probably connected his resistors in parallel for 8Ohm x 200 Watts total? Why wouldn't he just use a single 8Ohm 200W resistor?

Thanks.

 

[skibjr]

I would say the rule of thumb should be that your dummy loads be rated at least double that of your amplifier's output at the specified impedance. It wouldn't be a problem to have 500W 8-Ohm dummies on an amp spec'd to run 100 WPC into an 8-Ohm load, for example.

Regarding two 4-Ohm resistors, they should be in SERIES, and not parallel! Paralleling them would yield a 2-Ohm load which would most likely damage the amplifier being tested. :nono:

Now, some folks want to have the ability to test amplifier output at 4-Ohms, which is usually the minimum safe operating impedance for most "home-use" receivers and amplifiers. In this case, I would recommend an extra set of 8-Ohm dummies, paralleled to yield 4-Ohm loads. If you go this route, make sure your second set of dummies is at least double the expected output of your unit when it's running into a 4-Ohm load. Power handling capability of two resistors of equal resistance in parallel is twice that of the lowest wattage resistor.

 

[ChrisMarantz]

All you need is one 8 ohm 200 watt and one 4 ohm 200 watt non inductive loads.

I bought mine from Parts Express, I actually have 2 of each which comes in handy some times.

 

[feds27]

Thanks guys. I think I'll go with the 8ohm 200watt ones.

skibjr, you're right, I meant 2 x 4ohms in series would double the the impedance. I saw that in Echowars recommended equipment sticky he says he uses 2 x 4ohms for each channel. I was trying to figure out why he would do that.

 

[ConradH]

There are many ways to do this and what you do often depends on what you can easily get. I use a slab of heat sink with ten 40 ohm metal body Dale resistors. They're wired in 2 banks of 5 for 8 ohms and two channels, or I can parallel or series the banks to get one channel of 4 or 16 ohms. They'll do a total of 500 watts, though I'd need a fan to do it continuously. Brief measurements at higher powers have been no problem. Why 40 ohms? It's a convenient value and I got a boatload of 'em cheap at a hamfest.

 

[catrafter]

All you need is one 8 ohm 200 watt and one 4 ohm 200 watt non inductive loads.

I bought mine from Parts Express, I actually have 2 of each which comes in handy some times.

Not to quibble, but I think two loads are needed. General testing procedures require running both left and right channels simultaneously as the specification is something like 'XXX Watts per channel into X Ohms, both channels driven'.

Tom

 

[avionic]

Not to quibble, but I think two loads are needed. General testing procedures require running both left and right channels simultaneously as the specification is something like 'XXX Watts per channel into X Ohms, both channels driven'.

Tom

Run a pair of these Arcols....Fastened to a big ass heatsink.

http://www.mouser.com/ProductDetail...GAEpiMZZMtbXrIkmrvidL5X%2bgDHY9H3BLdCo7WsbnA=

It work's great..
http://www.audiokarma.org/forums/showpost.php?p=3930884&postcount=43

 

[laatsch55]

Or scrounge up an old BPI 2,000watt load bank. Mr. Catrafter is right , if you can't load both channels and drive em it won't be a valid test. You don't need to scope both but you do need to drive both.

 

[skibjr]

Not to quibble, but I think two loads are needed. General testing procedures require running both left and right channels simultaneously as the specification is something like 'XXX Watts per channel into X Ohms, both channels driven'.

Tom

:yes:

 

[sregor]

Power resistors get hot. Do not mount on wood or other non heat resistant surfaces. Provide ventilation and (as avionic mentioned) heat sinking. Almost started a fire once with 250 watt resistors mounted on plywood......Smelled the wood getting hot, and left two brown outlines of the resistor.

 

[gadget73]

If you're really going to be running high power for long periods, you can oil cool them. If built to a proper size, you can use a 1 gallon paint can full of mineral oil as your heat sink. Make sure the can has a vent so it doesn't build pressure as the oil warms.

 

[Robot17]

There are many ways to do this and what you do often depends on what you can easily get. I use a slab of heat sink with ten 40 ohm metal body Dale resistors. They're wired in 2 banks of 5 for 8 ohms and two channels, or I can parallel or series the banks to get one channel of 4 or 16 ohms. They'll do a total of 500 watts, though I'd need a fan to do it continuously. Brief measurements at higher powers have been no problem. Why 40 ohms? It's a convenient value and I got a boatload of 'em cheap at a hamfest.

I really like your dummy loads - where did you get the heat sinks? Are wire wound resistors okay?

Thanks
Bot

 

[Satch]

I built a dummy load capable of handling 2 x 200 Watts into 6.3 Ohms in an old amplifier I picked up cheap in a thriftstore. Gutted it and re-used the heatsinks and speaker terminals as inputs.

 

[datafone]

You can get ones like this 8ohm 300w on that site that can't be linked.

Or a 100w metal shell wire wound type like in the second pic

 

[avionic]

Or scrounge up an old BPI 2,000watt load bank. Mr. Catrafter is right , if you can't load both channels and drive em it won't be a valid test. You don't need to scope both but you do need to drive both.

Oh contrare..If your testing mono blocks , its legit! Or true dual mono blocks.:scratch2:

 

[ConradH]

I really like your dummy loads - where did you get the heat sinks? Are wire wound resistors okay?

Thanks
Bot

Like most everything I've got, the heat sink was a slab I picked up surplus. For low value resistors I don't think inductance is anything to worry about, so wire wounds should be fine. If you can get the non-inductive versions, so much the better, but the Q of an 8 ohm wire wound is near zip.

 

[Robot17]

Like most everything I've got, the heat sink was a slab I picked up surplus. For low value resistors I don't think inductance is anything to worry about, so wire wounds should be fine. If you can get the non-inductive versions, so much the better, but the Q of an 8 ohm wire wound is near zip.

Yep. Thanks.

Bot

 

[Monty_]

You might also consider building an active load which might end up cheaper but will take a bit more work:
http://www.youtube.com/watch?&v=8xX2SVcItOA

Example of a fancy version from http://www.audioscientific.com below:

 

[Monty_]

Here's a simpler active load:

http://endless-sphere.com/forums/download/file.php?id=20030&sid=28f34e0e026147bb8ecbe72cc5bfa7e1

 

[Jon_Logan]

I wouldn't use any of these active load examples for an AC signal.