I've had a Heathkit ID-5252 Audio Load for many years. It's just 4 non-inductive load resistors in a box.
One must KNOW the load resistance during the test for the power test to be of any use.
If the load resistance changes from 4r261 to some unknown value during the test then you are wasting your time and your electricity.
All (the common) metals have a positive temperature coefficient.
Many alloys are available that have a very low temperature coefficient.
Load resistors for power testing must use a low temperature coefficient material and not exceed the material's temperature range specified for that LOW coefficient.
If the load resistance changes from 4r261 to some unknown value during the test then you are wasting your time and your electricity.
All (the common) metals have a positive temperature coefficient.
Many alloys are available that have a very low temperature coefficient.
Load resistors for power testing must use a low temperature coefficient material and not exceed the material's temperature range specified for that LOW coefficient.
Power resistors that require an added heatsink generally operate at a higher temperature than resistors of equal power rating that operate as stand alone.
This is down to the thermal gradient from load resistance material to the outside air/water/oil, or final cooling medium.
A small resistor with lots of thermal interfaces and lots of thermal resistances will run hotter than a big resistor with a single thin layer of insulation.
Aluminium clad mineral insulated power resistor attached to a separate heatsink will run the metal resistance element very hot !!!
I do not recommend them for repeatable power testing.
This is down to the thermal gradient from load resistance material to the outside air/water/oil, or final cooling medium.
A small resistor with lots of thermal interfaces and lots of thermal resistances will run hotter than a big resistor with a single thin layer of insulation.
Aluminium clad mineral insulated power resistor attached to a separate heatsink will run the metal resistance element very hot !!!
I do not recommend them for repeatable power testing.
I use my load banks to limit current for other operations than just amplifier testing. ( Lately I use them for an anodizing current source where they run for a couple hours at a time)
My theory is to run a larger number of smaller metal clad resistors with an over-sized power handling total spread out on a heatsink that is larger than required with forced air cooling. This way the temperature doesn't rise (mine might move a couple degrees) under normal testing. (This is much the same as is done with feedback resistors in some amplifiers.) The resistors I provided the link to are 1% with a 20ppm/C temperature coefficiency so a couple degrees isn't going to cause much variance in operation. I don't see a measurable difference in resistance between start and finish of testing an amplifier. This likely won't be the case with all metal clad resistors. I've seen some where you can rotate the resistor in it's case. That can't be a good thermal connection.
I found non-metal clad resistors ran very hot as the current increased and force cooling was much harder effectively accomplish . A simple fan wasn't as effective because there just wasn't as much surface area as conveniently available being they weren't on a heat sink. (I didn't take the time to assemble a bank of them to try a tunnel for cooling. Metal clad looked much easier to work with) I'm not a fan of dunking them in water as the water in my area is very hard and conductive. Oil and other such coolants are a sloppy mess to clean up when finished. I prefer to just unplug the fan on mine and put it on the shelf and move on to something else.
An externally hosted image should be here but it was not working when we last tested it.
our test setup....dummy loads are 1500 watts each..
AJT's exposed wire load resistors with a thin insulation will run cooler and maybe much cooler than a metal clad attached to a heatsink.
The heatsink could be cool. The metal cladding will be slightly warmer.
It's what is inside the metal cladding that will be running hotter.
Even my ceramic cased wirewounds will run hotter than an exposed wire larger resistor.
The heatsink could be cool. The metal cladding will be slightly warmer.
It's what is inside the metal cladding that will be running hotter.
Even my ceramic cased wirewounds will run hotter than an exposed wire larger resistor.
It's pretty much a personal preference and also depends on how much you plan to use them and how much accuracy you require. Forced air heatsinks are a much cleaner option than an oil bath. I wouldn't think of having a bucket of oil sitting around for testing purposes myself. That's a huge mess waiting to happen.
Large wirewounds like AJT's would work well but they aren't cheap and they aren't as easy to come by. Usually they are a 10% tolerance as well so the accuracy may not be as good (8 ohms may be anywhere from 7-9), although his are adjustable so they can be fine tuned as needed. This is only an issue if you plan to publish numbers. For evaluation of your own products you will be using the same load every time so comparison will be valid as long as the resistance isn't drifting excessively with temperature/load.
Temperature issues are more of a problem if you are running close to rated maximums. If you oversize your test load it will remain cooler under lighter loads and not have issues with heat as badly. Likely AJT's resistors will just get a little warm when he's running a 250 watt amp into them. Mine don't really warm up at all.
I haven't seen any issues with temperature at all with mine. If you look at thermal contact area of a 50W resistor, it's much better than a TO-274 transistor. If temperature was going to be an issue I would likely see some deviation of resistance after running 250 Watts through my resistors for a couple hours while anodizing (I've only done this a couple times so far but measured the resistor bank before and after each time). I haven't measured any difference. The only deviation I've seen is my transformer warming up and the oxidization of the aluminum itself.
It's quit possible there may be some heat buildup during short high power bursts. That would be tough to measure. It might be an interesting experiment some day. I would likely need to dissect a resistor just to figure out where to insert a probe without damaging the resistive element.
Hi Andrew,
Nope.
The thermal resistance is much, much lower with the metallic resistors that mount on a heat sink. I have both the 225 watt tubular and the 250 watt Dale resistors. The 225 watt tubular resistors run much hotter than the ones on the heat sinks at a 100 watt dissipation. It's not a small difference.
With the heat sink mounted resistors, the performance can be enhanced by using fans and / or lower thermal resistance types of heat sink lower than what they list as a required type to get the rated dissipation.
We have blown a couple of these resistors (comical failure mode), but the power delivered to them was greatly over their rating (over 500 watts for the last one). I just grit my teeth and replace them as needed.
Andrew, I would have to say that the 250 watt Dale or Dale equivalent is the best way to go. These are normally a 1% tolerance part and non-inductive. Okay, reduced inductance.
-Chris
Nope.
The thermal resistance is much, much lower with the metallic resistors that mount on a heat sink. I have both the 225 watt tubular and the 250 watt Dale resistors. The 225 watt tubular resistors run much hotter than the ones on the heat sinks at a 100 watt dissipation. It's not a small difference.
With the heat sink mounted resistors, the performance can be enhanced by using fans and / or lower thermal resistance types of heat sink lower than what they list as a required type to get the rated dissipation.
We have blown a couple of these resistors (comical failure mode), but the power delivered to them was greatly over their rating (over 500 watts for the last one). I just grit my teeth and replace them as needed.
Andrew, I would have to say that the 250 watt Dale or Dale equivalent is the best way to go. These are normally a 1% tolerance part and non-inductive. Okay, reduced inductance.
-Chris
we got those resistors from the surplus market and very cheap...
we monitored those resistors during testing and they got up to 140*C,
they were dissipating 1000 watts each...
Small calrod on a N. american electric cooking range is 7R. Did this
once for the Badger prototype. Hole in poly bucket , element terminals sticking out the top. crude , but it worked.
PS - in the water , real close to 7R. PTC usually will not be noticeable until
100's of C (or incandescence on the range).
I did not do the amps at the big radio shack repair center. Those that did ,
just ran speakers at moderate volumes on speakers for small to mid amps ,
They had a resistor load for larger PA type amps .... but rarely used it.
My opinion for a dedicated expensive load would be to test an unknown
prototype. At the OEM repair level (at least radio shack) , it was rarely
used. ??
OS
once for the Badger prototype. Hole in poly bucket , element terminals sticking out the top. crude , but it worked.
PS - in the water , real close to 7R. PTC usually will not be noticeable until
100's of C (or incandescence on the range).
I did not do the amps at the big radio shack repair center. Those that did ,
just ran speakers at moderate volumes on speakers for small to mid amps ,
They had a resistor load for larger PA type amps .... but rarely used it.
My opinion for a dedicated expensive load would be to test an unknown
prototype. At the OEM repair level (at least radio shack) , it was rarely
used. ??
OS
Hi OS,
We used the Dale resistors for warranty repairs, and out of warranty repairs. I had some 50 watt 4 R tubular resistors for car radios and small car amps. That is one business I don't miss too much. Car audio.
We also serviced the recording studio and MI markets. You meet all kinds of people in that line of work, and it can be fun. The big Carver amps put a beating on those dummy loads. AC power at the bench was an issue then as well.
-Chris
We used the Dale resistors for warranty repairs, and out of warranty repairs. I had some 50 watt 4 R tubular resistors for car radios and small car amps. That is one business I don't miss too much. Car audio.
We also serviced the recording studio and MI markets. You meet all kinds of people in that line of work, and it can be fun. The big Carver amps put a beating on those dummy loads. AC power at the bench was an issue then as well.
-Chris
Radio shack must of bought the same rig for all their service centers. It was similar
to Jwilhelm's fan cooled high power unit.
But , it hung on the wall ... mostly.
We could tell if the amp man was having a productive day by the ratio of loud rock
to curse words.
Bad day = we were greeted by the smell of semi's ....
to Jwilhelm's fan cooled high power unit.
But , it hung on the wall ... mostly.
We could tell if the amp man was having a productive day by the ratio of loud rock
to curse words.
Bad day = we were greeted by the smell of semi's ....
The Heathkit Cantenna had a single 90W resistor in a bucket of mineral (or transformer) oil.
With transformer oil it was rated for 1KW for ten minutes, and 200W indefinitely.
http://www.orcadxcc.org/content/cantenna_va7jw.pdf
I use three paralleled 27Ω/55W enameled wirewound resistors. If the bench starts to smoke (Carver M1.5T) I can dip into a bucker of water.
Heater elements have too much of a temperature coefficient (IMO).
With transformer oil it was rated for 1KW for ten minutes, and 200W indefinitely.
http://www.orcadxcc.org/content/cantenna_va7jw.pdf
I use three paralleled 27Ω/55W enameled wirewound resistors. If the bench starts to smoke (Carver M1.5T) I can dip into a bucker of water.
Heater elements have too much of a temperature coefficient (IMO).
Some of the wirewound resistors sold without temperature coefficient specs for amp testing by DIY-oriented retailers increase in value by 100% if you use them near their rated power.
Just about any carbon composition resistor will perform in the same range, but happily they are now very rare.
I was given a set of 8 ohm ca. 250 watt large heatsink mounted Dale NI resistors and mounted them in a ducted air flow. I also have some Arcol 300 watt NI resistors of the same style and they seem to be as good.
I previously built some dummy loads in gallon paint cans filled with oil and with heatsink stock epoxied to the outside with heat resistant epoxy. The oil really liked to seep out when hot.
Last edited:
Many wire wound resistors are inductive and spoil high frequency testing. One could use some of these power metal film resistors mounted to a heat sink. Use multiple ones mounted to a heat sink, and fix into a breadboard so as could be connected in different ways to produce different values. Also it would be easy to add inductance and/or capacitance for complex load simulation.
this is diyAudio - not a NIST traceable test lab or "legal for commerce" standards compliant discussion
the load variation is not a problem for all test purposes - log hearing sensitivity makes 10-20% error in actual power actually quite small in perceptual consequence
for measured V distortion small tempco load variations don't change the numbers meaningfully for typical uOhm to milliohm output Z of SS feedback power amps
bullying the OP into buying what he doesn't need to tell most of what is useful to know about a diy audio power amp is not helpful
the load variation is not a problem for all test purposes - log hearing sensitivity makes 10-20% error in actual power actually quite small in perceptual consequence
for measured V distortion small tempco load variations don't change the numbers meaningfully for typical uOhm to milliohm output Z of SS feedback power amps
bullying the OP into buying what he doesn't need to tell most of what is useful to know about a diy audio power amp is not helpful
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.