Hi jcx,
You know, sorry but I have to disagree with you. It is rare when we disagree.
There have been many helpful suggestions on how to do this on the cheap, as well as personal experiences with those folks who have done this both ways. If there was no way to set up a dummy load but the real expensive route (that I finally had to take), I could see your point. However each suggestion has been accompanied with information on what to expect and what the limitations are. I think the OP got exactly the information needed to make an intelligent decision on this.
How would you feel if he went out and bought those expensive Dale resistors, but ended up breaking them by bolting them down on an uneven surface? But, he was made aware of the need and associated costs of milling the mounting surface to avoid this.
Load variation is a problem when measurements are being made. He isn't doing anything by ear, he is measuring the output power.
The big question is how much power dissipation rating he needs, and that determines the costs involved. He did buy a nice Fluke meter. Why nullify the accuracy of that meter by using inaccurate and variable loads? Besides, his time is the highest cost item. That means take the measurements quickly without messing around with math and guesswork - unless the OP decides to do it that way.
No one is bullying anyone around here to the best of my knowledge.
-Chris
You know, sorry but I have to disagree with you. It is rare when we disagree.
There have been many helpful suggestions on how to do this on the cheap, as well as personal experiences with those folks who have done this both ways. If there was no way to set up a dummy load but the real expensive route (that I finally had to take), I could see your point. However each suggestion has been accompanied with information on what to expect and what the limitations are. I think the OP got exactly the information needed to make an intelligent decision on this.
How would you feel if he went out and bought those expensive Dale resistors, but ended up breaking them by bolting them down on an uneven surface? But, he was made aware of the need and associated costs of milling the mounting surface to avoid this.
Load variation is a problem when measurements are being made. He isn't doing anything by ear, he is measuring the output power.
The big question is how much power dissipation rating he needs, and that determines the costs involved. He did buy a nice Fluke meter. Why nullify the accuracy of that meter by using inaccurate and variable loads? Besides, his time is the highest cost item. That means take the measurements quickly without messing around with math and guesswork - unless the OP decides to do it that way.
No one is bullying anyone around here to the best of my knowledge.
-Chris
Why do we want the load ? this is a rational question.
To measure the amp to the watt , or to see if it can survive the "customer" ?
calrods and cheap Ebay Chinese resistors can accomplish the latter.
Maybe to measure for publication or an audiophile review would require a more
controlled load. Carlos and myself have lit up toaster ni-chrome with amplifier
outputs.(old DX threads).
The Calrod will stay @ 7 R unless you boil off the water.That would take hours
to get a cup of tea.
After you drink your tea , I'm sure the surviving amp would bring years of service.
OS
To measure the amp to the watt , or to see if it can survive the "customer" ?
calrods and cheap Ebay Chinese resistors can accomplish the latter.
Maybe to measure for publication or an audiophile review would require a more
controlled load. Carlos and myself have lit up toaster ni-chrome with amplifier
outputs.(old DX threads).
The Calrod will stay @ 7 R unless you boil off the water.That would take hours
to get a cup of tea.
After you drink your tea , I'm sure the surviving amp would bring years of service.
OS
repeatability only comes from consistent methods and tools.
A variable load/dummy resistor does not fit with repeatability.
The carbon film Calrod is a carbon resistor. Carbon normally has a negative temperature coefficient. The Calrod is different, it has been formulated to have a tempco of +0.2% to -0.08% per C (similar to an NP0 capacitor)
But you do need to read the link. It measured very well using AC impedance measurement method.
It measured extremely badly using DC resistance measurement (down as low as 20r).
The error mechanism is explained in the text.
The method matters. Repeatability requires thought, not guesswork.
A variable load/dummy resistor does not fit with repeatability.
The carbon film Calrod is a carbon resistor. Carbon normally has a negative temperature coefficient. The Calrod is different, it has been formulated to have a tempco of +0.2% to -0.08% per C (similar to an NP0 capacitor)
But you do need to read the link. It measured very well using AC impedance measurement method.
It measured extremely badly using DC resistance measurement (down as low as 20r).
The error mechanism is explained in the text.
The method matters. Repeatability requires thought, not guesswork.
Hi AJT,
All loads will have a positive temperature co-efficient. How much? So the small impedance rise in those Dale (or similar) dummy loads isn't a big problem. However, this could be quantified and using a nomograph, you could quickly make corrections. Dale probably does have such a thing for these resistors.
-Chris
Yes, to be expected. Here we got with that peak to average power level ratio again. Depending directly on the dynamic range of signal, peak to average power is often 10:1 at the least, 15:1 being more common. So if the average power with music was showing 140 watts, you were asking for 1,400 watt peaks, minimum. A little clipping going on maybe?
All loads will have a positive temperature co-efficient. How much? So the small impedance rise in those Dale (or similar) dummy loads isn't a big problem. However, this could be quantified and using a nomograph, you could quickly make corrections. Dale probably does have such a thing for these resistors.
-Chris
Well surprisingly- I know this is not correct- but when I play my amps at normal volume for my setup (2 12 inch 2 ohm speakers- one per channel) the amp only pushes out about 5v on an analogue meter... yes I know a speaker is connected not a dummy load, but technically that would be only 12.5W...
Anyways IF a load has a constant resistance at 0 degrees to 150 degrees, at least within 5%, would it be a good load to use??
Anyways IF a load has a constant resistance at 0 degrees to 150 degrees, at least within 5%, would it be a good load to use??
yes Chris, and if more precision is needed, we take the actual ohm reading of dummies right after testing, but in this case we did not...
btw, the test equipment i posted in not mine, they belong to a friend and i just showed them how FTC testing is done...
Well surprisingly- I know this is not correct- but when I play my amps at normal volume for my setup (2 12 inch 2 ohm speakers- one per channel) the amp only pushes out about 5v on an analogue meter... yes I know a speaker is connected not a dummy load, but technically that would be only 12.5W...
Anyways IF a load has a constant resistance at 0 degrees to 150 degrees, at least within 5%, would it be a good load to use??
not surprising, since music as compared to sine waves have very low energy content...
Hi Gamerandds,
In truth, speakers have a widely varying impedance with frequency. They are useless for anything other than listening to the amplifier.
-Chris
Nope. With a phase angle like that, you could really upset the amplifier! 150 ° has gone through a purely reactive load, and amplifiers really don't like that!! Besides, that isn't a possible load from any normal speaker system.
In truth, speakers have a widely varying impedance with frequency. They are useless for anything other than listening to the amplifier.
-Chris
I've tried a number of different kinds of resistors, including resistors sold by retailers catering to the technician's trade for the purpose of amplifier load testing and found that their hot resistance can be almost twice their cold resistance.
OTOH NI resistors with specified low temperature coefficients such as those from Dale, Caddock, and Arco can survive heating to 200 degrees F or beyond and even remain within their room temperature spec.
If a resistor's resistance remains stable within a percent or two up to 150 degrees F, then it will probably be good enough for general use.
There is a reasonable question about the purpose of the test.
If the goal is proof of performance of the gear under test or after maintenance, then the stability of the load resistor is of the essence.
If the purpose is to simulate operation while driving loudspeaker voice coils, then relatively large changes in resistance with similar changes in temperature is OK.
Since music has a high crest factor the relevance of sine wave testing to actual use is relevant, but a separate issue. Obviously, this will significantly reduce the thermal load on the resistor and its heating. IME there will still be signficant changes in the resistance of a load resistor that has the sort of temperature coefficients commonly seen in standard resistors.
I have tested a number of water heater elements, and the ones I tested had about close to zero tempco as anything else I've tested.
There are a series of power amp tests on the AVS forum that are based on their use, and contain actual measurements of heating element resistance before and after the test.
Measuring Amplifiers - AVS | Home Theater Discussions And Reviews
More specifically:
http://www.avsforum.com/forum/155-diy-speakers-subs/855865-measuring-amplifiers.html#post10747784
I mentioned stove heater elements earlier, I suspect they're about the same. I've put in a few hundred watts DC (to a 1500 watts or so element), and the resistance measurement afterward was within maybe 2 percent of the before resistance. But I doubt this is truly accurate - see my description below.
Measuring Amplifiers - AVS | Home Theater Discussions And Reviews
More specifically:
Measuring Amplifiers - AVS | Home Theater Discussions And Reviews[/QUOTE]
I recognize the pictures in the thread, I recall a video of some guy with that same setup testing an amp with those water heater elements, full power at 20kHz for a minute or two, and at the end the amp's rather large heat sink was a bit too hot to touch. More relevantly, he measured the load resistance before and after the test, but I question the accuracy of that. During the test the wire inside the heating element could heat up to a much higher temperature than the water, but by the time he connects his meter the wire would have cooled down to the (barely warmed) water temperature, and the resistance changed accordingly. I'd really want a low-value shunt resistor to measure the current through the load as well as the voltage across it DURING the test.
The elements are cheap enough that one can buy many more than are needed for the power rating, and connect them all so that each one only runs at a small fraction of its power rating and so its resistance changes that much less.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.