I've been using some very underpowered (2 X 5W|| wire wound) resistors as a dummy load when testing my amp, and have been looking to create a proper dummy load that won't burn the carpet, melt plastic it comes in contact with or imit strange smells 
.......
I've searched and people gennerally recommend non-inductive types, but for 100W rated non-inductive resistors I'm looking at $40AUS each (and I'd want four... 2 X 15 Ohm wired in parallel). so that's $160!!! (luckily I have a heat sink).....
I can get ceramic wire wound (Obviously inductive) resistors for about $17Aus each which works out at about $68.........
Wouldn't the fact that these resistors have some inductance mean that they are a closer (marginally) approximation to the load presented by a speaker? Is it ok to save some bucks and get the cheap inductive ones, or should I just grin and bear it and fork out for the non-inductive ones?
Tony.
.......I've searched and people gennerally recommend non-inductive types, but for 100W rated non-inductive resistors I'm looking at $40AUS each (and I'd want four... 2 X 15 Ohm wired in parallel). so that's $160!!! (luckily I have a heat sink).....
I can get ceramic wire wound (Obviously inductive) resistors for about $17Aus each which works out at about $68.........
Wouldn't the fact that these resistors have some inductance mean that they are a closer (marginally) approximation to the load presented by a speaker? Is it ok to save some bucks and get the cheap inductive ones, or should I just grin and bear it and fork out for the non-inductive ones?
Tony.
Connect multiple low power resistors in parallell, I use metal oxid type 5W and have connected totally 20 in series parallell to get 8 ohm capable of 100W continuous power.
9 39ohm + 1 47ohm resistors is connected in parallell giving very close to 4 ohm, 2 of these combinations are connected in series.
BTW, this is how dummy loads for RF was done long time ago by some manufacturers.
Regards Hans
9 39ohm + 1 47ohm resistors is connected in parallell giving very close to 4 ohm, 2 of these combinations are connected in series.
BTW, this is how dummy loads for RF was done long time ago by some manufacturers.
Regards Hans
Hmmmm that's two options Thanks guys.... I was trying to avoid paralleling up too many resistors, but I guess the advantage of using heaps of 5W is that a heat sink wouldn't be neccessary.
I did think about using 3 X 47R and 1 X 22R all in parallel for 7.6 Ohms at 100W but wasn't sure whether the power disipation would be even or whether the 22R would cop more of a beating since the whole thing isn't symetrical....... that'd work out at a tad under $70Aus.... I would like it to be able to handle about 150W though......
Those are some mutha resistors you've got there Joe I'll see if I can find some BTW why the change on the avatar?
edit: Ummm searched on dale, and dale electronics, but I'm not sure I have found the right site... Got a url?
Tony.
I did think about using 3 X 47R and 1 X 22R all in parallel for 7.6 Ohms at 100W but wasn't sure whether the power disipation would be even or whether the 22R would cop more of a beating since the whole thing isn't symetrical....... that'd work out at a tad under $70Aus.... I would like it to be able to handle about 150W though......
Those are some mutha resistors you've got there Joe
I'll see if I can find some BTW why the change on the avatar?edit: Ummm searched on dale, and dale electronics, but I'm not sure I have found the right site... Got a url?
Tony.
I have a few Dale resistors which are ideal for dummy loads.
See at http://www.kbacoustics.com/dale/
Do not reply to the email address on that page - it is obsolete.
See at http://www.kbacoustics.com/dale/
Do not reply to the email address on that page - it is obsolete.
Why?, because of the increased capacitance? Capacitance is not a big problem when building a dummy load for 8 ohm or similar impedance, the problem with using many resistors is when you connect them in series as it will increase the total inductance and that can have an effect even at audiofrequencies.
For a simple calculation using 20 resistors in parallell:
Capacitance of each resistor ~10pF, giving 100pF for 10 in parallell, 100pF + 100pF in series give a total of 50pF in parallell with 8 ohm. The reactance of 50pF at 20kHz is ~159000 ohm so the effect is neglible.
Regards Hans
Hi Hans,
No I wasn't worried about the capacitance, just the logistics of connecting them all up Ideally I just wanted to bolt a couple of resistors to a heat sink that I have and connect the speaker wires...... Then I saw the price!!!!
These resistors were the last thing I was going to add to an order I'm putting together for various bits and peices and it's allready up to about $140 without them....
Bill, the cheaper resistors I mentioned before look very simmilar to the ones you posted (although maybe your's are low-inductive ones.... I discovered you can get low inductance wire-wound).
Millwood, I think I'd be more likely to burn the carpet if I used heater elements But full marks for the diy element
Maybe I'll just get some of the cheaper 25W ones and mount them on the heat sink, will be better than the 10W power rating I have at the moment. I can allways by some fancy non-inductive types later on, if I feel the need. After all my amp isn't exactly audiophile quality, I just need something that'll make it do some work while testing (which most of the time isn't at full power anyway.
BTW Joe, I found something very simmilar to your Dale resistors at RS Australia, but they were $65 each and 4.7 ohms (although that's not bad considering they are rated at 200W)
Regards,
Tony.
No I wasn't worried about the capacitance, just the logistics of connecting them all up
Ideally I just wanted to bolt a couple of resistors to a heat sink that I have and connect the speaker wires...... Then I saw the price!!!! These resistors were the last thing I was going to add to an order I'm putting together for various bits and peices and it's allready up to about $140 without them....
Bill, the cheaper resistors I mentioned before look very simmilar to the ones you posted (although maybe your's are low-inductive ones.... I discovered you can get low inductance wire-wound).
Millwood, I think I'd be more likely to burn the carpet if I used heater elements
But full marks for the diy element Maybe I'll just get some of the cheaper 25W ones and mount them on the heat sink, will be better than the 10W power rating I have at the moment
. I can allways by some fancy non-inductive types later on, if I feel the need. After all my amp isn't exactly audiophile quality, I just need something that'll make it do some work while testing (which most of the time isn't at full power anyway.BTW Joe, I found something very simmilar to your Dale resistors at RS Australia, but they were $65 each and 4.7 ohms
(although that's not bad considering they are rated at 200W)Regards,
Tony.
Tony,
Do you have any compponent surplus companys around? They seem
often to have a lot of power resistors in the 5 - 10W range, at least
here around. You will have to figure which of those values they have
to use and how to hook them up in series and parallel to get what
you want, but it can be a bargain and you won't need any heatsinks,
which save a lot of money. It'll probably end up a worthy candidate
for the ugly prototypes thread, but so what.
Do you have any compponent surplus companys around? They seem
often to have a lot of power resistors in the 5 - 10W range, at least
here around. You will have to figure which of those values they have
to use and how to hook them up in series and parallel to get what
you want, but it can be a bargain and you won't need any heatsinks,
which save a lot of money. It'll probably end up a worthy candidate
for the ugly prototypes thread, but so what.
Just a word of caution.....
Most of the high power resistors you guys mention will have to be mounted on a fairly large heatsink in order to be anywhere near the listed dissipation figures.
If you parallell plain old fashion wirewounds, it will have to be a flat package,--- if they are made up to form block or a "roll". the innermost ones will have to take a lot more heat than the outer ones, and it will most certainly get very hot in a very short time, -- a minute or two only on full load.
The classical way to make a dummy load is to immerse a 20-50 W ( a parallell package will be just fine) resistor in 1 L of oil,- this will hold for 100W full load for several minutes. Even finegrain sand have been used,but doesn't distribute the heat as well as oil.
Most of the high power resistors you guys mention will have to be mounted on a fairly large heatsink in order to be anywhere near the listed dissipation figures.
If you parallell plain old fashion wirewounds, it will have to be a flat package,--- if they are made up to form block or a "roll". the innermost ones will have to take a lot more heat than the outer ones, and it will most certainly get very hot in a very short time, -- a minute or two only on full load.
The classical way to make a dummy load is to immerse a 20-50 W ( a parallell package will be just fine) resistor in 1 L of oil,- this will hold for 100W full load for several minutes. Even finegrain sand have been used,but doesn't distribute the heat as well as oil.
Hi,
Be careful when using wirewound resistors, normal types have enough self inductance to make a considerable influence even at audio-frequencies.
10uH is ~1.25 ohm at 20000 Hz, there are many wirewound resistors with that inductance and more.
Regards Hans
(I want to measure amplifier response as exact as possible up to >1MHz when I measure open loop response aso, therefore I consider it important to use a good dummy load.)
Be careful when using wirewound resistors, normal types have enough self inductance to make a considerable influence even at audio-frequencies.
10uH is ~1.25 ohm at 20000 Hz, there are many wirewound resistors with that inductance and more.
Regards Hans
(I want to measure amplifier response as exact as possible up to >1MHz when I measure open loop response aso, therefore I consider it important to use a good dummy load.)
I think I need to stop thinking about components for a while, I'm going to go insane!!!! I found some wirewound resistors at RS components that actually list their inductance, 5.4uH for 10R (100W) and 22uH at 100R .... shame the resistors at farnell don't have that info. I found what I thought would do me for the time being without breaking the bank too much (still expensive, but not as much as the first lot).
3 X 24R 50W in parallel at $6.38 each so that's a bit under $40 for two channels.
the heatsink I was planning to use is 15cm X 16cm with a base about 8mm thick and fins about 25mm high.
as mentioned by AuroraB the specs for the resistors call for pretty big heat sinks the above mentioned resistor specs a 30cm X 30cm X 1.5mm thick heat sink, now I'm guessing they are not talking finned, so I think I should be able to get away with my heat sink for a short while (few minutes maybe, considering I'll have 6!!!).
The heat sink is out of an old mainframe power-supply. was being used by 5 very chunky 2SD3036 transistors and 3 other power transistors of some sort. pic attached.
Tony.
Attachments
Basically, there is no need to worry about the inductance of the dummy load, unless you specifically want a purely resistive load. After all, loudspeakers presents a rather ractive load, mostly inductive.
The easiest load I think, is made of oldfashioned carbon comp resistors, 2 or better 5 W, 15 parallelles of 120 ohms, 2 or 5 W immersed in mineral oil, --- good for 200 W for several minutes.
I have a 50 ohms in a gallon box, - rated for 1kW, very apropriately named " The Cantenna".
Otherwise, - the mentioned Dale resistors of various values or combinations are very good, provided you have enough surface area to mount them on.
The easiest load I think, is made of oldfashioned carbon comp resistors, 2 or better 5 W, 15 parallelles of 120 ohms, 2 or 5 W immersed in mineral oil, --- good for 200 W for several minutes.
I have a 50 ohms in a gallon box, - rated for 1kW, very apropriately named " The Cantenna".
Otherwise, - the mentioned Dale resistors of various values or combinations are very good, provided you have enough surface area to mount them on.
Jax said:
I was thinking of Caddocks myself, but they are quite expensive and
definitely needs a lot of heatsinking. Further, I think I read somewhere
on this forum that they are very sensitive to overcurrents. The have
a current limit you must adhere to, even if way below the rated
power dissipation. They might still be a good choice, though, if
taking these things into account. Otherwise, if we are talking
expensive solutions, Dale has low-inductance power resistors
with 4 and 8 ohm resistance. I don't quite remember now if they
are 25 or 50W.
Edit: Found the current limit stuff for Caddocks
http://www.diyaudio.com/forums/showthread.php?postid=311223#post311223
Hmmmm, Just checked the max dimensions 5cm X 3cm should fit but it'll be crowded. Maybe I do need a bigger heat sink after all....
The immersed in oil sounds like a good way to go, but I think it is a bit hazardous to the carpet too. Would be ok if I had a workshop.
I guess the other thing with the inductance is that if I'm paralleling 3 then the inductance will be 1/3 that of a single resistor and as you say a speaker has a significant inductance (as does the crossover if it is more than 1st order). I think I'll just get the 3 X 50W 24R stick them on the heatsink and see how it goes. It has to be better than the two 5W resistors I currently have paralleled (which got so hot they burned the carpet (and it's wool)!
Tony.
The immersed in oil sounds like a good way to go, but I think it is a bit hazardous to the carpet too
. Would be ok if I had a workshop. I guess the other thing with the inductance is that if I'm paralleling 3 then the inductance will be 1/3 that of a single resistor and as you say a speaker has a significant inductance (as does the crossover if it is more than 1st order). I think I'll just get the 3 X 50W 24R stick them on the heatsink and see how it goes. It has to be better than the two 5W resistors I currently have paralleled (which got so hot they burned the carpet (and it's wool)!
Tony.
Yeah, come on people! First you dish out on ultra-expensive super-dooper non-inductive resistors, and then you buy ultra-expensive super accurate OFC ribbon-wound hand-made inductors to precisely model the speaker's reactance! Has the world gone mad (rhetorical question)???AuroraB said:
Don't mean to ruin the thread, but in New Zealand those 5W resistors that everyone seems to be avoiding cost less than a dollar. It takes something like a kilowatt-minute to bring a cup of water to the boil, so why not use water for heatsinking for those bargain resistors? Those retro-style glass coke bottles (or whatever other glass container) would be perfect for holding the water, and if you're worried about the water reducing the resistance slightly - just cover the exposed conductors with a thin layer of oil.
For the ultimate dummy speaker load you could even construct a notch filter network to better model the mechanical impedance peaks that speakers have. Or try to measure the inductance of that imperfect resistor and compensate for it by putting a capacitor (of the correct value) in parallel with it.
Maybe I'm getting cynical, but I can see it coming - a new thread entitled: "top-quality polypropylene or teflon capacitors required for dummy speaker load"....
CM
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.