Resistive Load for Amp Testing

[Sonnie Parker]

I having a fellow build me a amp testing load for 2/4/8 ohm testing. It will be similar to the smaller tank you see below... which is about 20 gallon. It will be filled with RV antifreeze. I spoke with Pat Brown who built these for testing pro audio and car amps. He has tested amps rated up to 10,000 watts. He says the temps have never really gotten remotely hot and he uses mineral oil.

At any rate... each element is 8 ohms (measured/tested with an ohm meter). I believe I will need 8 of these wired so that I can use jumpers to come up with the appropriate 2/4/8 ohm loads. However, I'm struggling wrapping my mind around the correct wiring from each element to the speaker terminals.

If I wire each element to one set (+/-) of terminals... we have 8 ohms on that terminal, and we will have 8 speaker terminal sets total. I suppose I can daisy chain the terminals from the back to the next terminal. Then use jumpers to get to 4 ohm or 2 ohm.

Can anyone draw me up a diagram/schematic of the how this would all need to be wired, then how the jumpers would need to be wired to accomplish each load?

 

[RichB]

I have never seen anything like that. I think reactive load would be ground breaking.

I am using this device.

Unfortunately, it is intent on destroying the human race, so I can't let it out of the house.

- Rich

 

[ddude003]

@Sonnie Are you sure your not looking for one of these devices...

I hope you have some Good Insurance and a Halon System...

Seriously??? Mineral oil and/or antifreeze as a dielectric... Dude, try some Fluorinert...

 

[Sonnie Parker]

There are others who actually use motor oil in the paint cans and 5 gallon buckets... a combustible material. Although those elements really never get hot like they would with contact power in a hot water heater.

The guy that built those with mineral oil... you can see the temp gauge covering the spout at the top. He said not even needed.

The guy building the tank uses the RV antifreeze for his welder... says it the safest liquid to use, and a much less expensive than mineral oil.

 

[RichB]

Here are resistors sent by Emotiva to Audioholics to review their XPR-1 kilowatt plus amp.

Emotiva XPR-1 Mono Amplifier Power and Distortion Measurements | Audioholics

- Rich

 

[Sonnie Parker]

I'll have a massive load once I get this system setup... I just need to work out the wiring.

 

[Todd Anderson]

Oh my, Sonnie. You're going to have EIGHT of those???

 

[Sonnie Parker]

Oh my, Sonnie. You're going to have EIGHT of those???

There are 8 of the heater elements in the small tank... NOT 8 tanks. I'll have just one tank, but instead of only 4 elements, there will be 8 in it.

 

[Todd Anderson]

Ah.... I see.

I thought you'd lost your mind!

 

[Sonnie Parker]

Ah.... I see.

I thought you'd lost your mind!

I lost my mind long before this.

 

[ddude003]

Scotty, We Need More Power!

She can'nae take any more, captain... If I push these amps too hard in the condition they're in they'll blow apart...

 

[Sonnie Parker]

Here's what I am initially planning, but not sure if it's right or not for all I need. I could connect directly to the first set of terminals for a two-channel 8 ohm test. Then I would need to jump to the next set via parallel wire to get to 4 ohm. I guess I would need to jump all 4 terminals in parallel of one channel to get to 2 ohms. These will need to be spade and banana plug capable terminals... as I'll have to use banana's to jump, and spades to connect to the amp channel(s).

 

[ddude003]

Oh... Those are water heater elements... I guess I would get out my handy dandy digital volt/ohm meter and double check the measurements at each heater element and then at the corresponding terminals... Once each set of heater element and corresponding terminal checks out, I would then do as you say above, measuring each step of the way from 8 ohms, 4 ohms and 2 ohms... Only if everything is within some reasonable spec would I attempt to connect an Amplifier worth more than about $25 to the thing for the first time... Wear some eye protection and keep a fire extinguisher and some thick rubber gloves handy...

Smoke 'em if you got 'em... I love the smell of napalm in the morning...

 

[Sonnie Parker]

Well the guy that is using them now has tested amps up to 10,000 watts and thankfully he is still alive.

 

[JLM1948]

No audio amplifier needs that kind of load. Even those 10-20kW amplifiers take their power from a 32A Powercon or a Nema 30A, which accounts for a little less than 4kVA of input power. With the inevitable losses, the continuous power delivered to the load is less than 3.5kW.
These amps are tested with tone-burst that result in a continuous power being just a fraction of the peak power (typically between 1/8th and 1/16th).
The usual way of wiring dummy loads is via 16A banana sockets compatible with spade lugs. In order to make parallelling loads easily, the connectors must be duplicated, so each resistor has 4 terminals.

 

[pinox67]

I had a (small) dummy load built. Simplifying, the circuit is composed of 4 resistors of 8ohm 200W for each channel, which can be paralleled progressively with input, creating loads of 8, 4, 2.6 and 2 ohm. So, on 2 ohm it can reach up to 800W. The input connection from the amplifier is always the same, the resistors are connected with roboust jumps.
Pay attention to the measurement of differential power amplifiers (like D class)! In order to never make a mistake, I connect the measuring instrument in parallel to the load on its balanced (XLR) input: + and - to pins 2 and 3; pin 1 to the amplifier ground.

 

[pickindoug]

The amateur radio industry has been using oil filled dummy loads for years. They're 50 ohm impedance (but quite possibly that's also the resistive spec). See Heath Cantenna. It may be adaptable to audio power measurement. I have a few amps that can exceed my Tenma 100 watt stereo dummy load, but I also found that 100 watts continuous on that little box will generate the magic smoke. Didn't kill it, but apparently it's rated only for short duration at full power.....fine for a brief power at clipping test.

 

[Sonnie Parker]

The usual way of wiring dummy loads is via 16A banana sockets compatible with spade lugs. In order to make parallelling loads easily, the connectors must be duplicated, so each resistor has 4 terminals.

So on the back side of the terminal, parallel another jump to a second terminal, so that I have two pairs of terminals for each resistor.

 

[Sonnie Parker]

I too had a dummy (small) load built. Simplifying, the circuit is composed of 4 resistors of 8ohm 200W for each channel, which can be paralleled progressively with input, creating loads of 8, 4, 2.6 and 2 ohm. The input connection from the amplifier is always the same, the resistors are connected with roboust jumps.
Pay attention to the measurement of differential amplifiers (like class D)! In order to never make a mistake, I connect the measuring instrument in parallel to the load on its balanced (XLR) input: + and - to pins 2 and 3; pin 1 to the amplifier ground.

This is what many major publications are using... typically several of the 250 watt resistors in parallel. Some use a couple of larger resistors for larger power supplies. The problem lately has been sourcing the resistors. To come up with something that would safely test 2000 watts continuous power... we are at around $1800 just for resistors, but many are on back-order of 20 weeks.

 

[Tim Dawson]

@Sonnie Are you sure your not looking for one of these devices...

View attachment 41835

I hope you have some Good Insurance and a Halon System...

Seriously??? Mineral oil and/or antifreeze as a dielectric... Dude, try some Fluorinert...

RF dummy loads and elrctrical transformers have been using oil for centuries. It's a non problem.

 

[SVChucko]

I built a test load out of a plastic trash can with a lid and 8 water heater elements, and filled it with tap water. Since I don't have anything more than a 120V 20A circuit in the house, that's plenty.

The tricky part is finding water heater elements with the right resistance. I used short 3500W @ 240V elements, which are dirt cheap (about $6 each from Rheem/Ruud direct). They should have a nominal resistance of 16.46 Ω. When they're paired up, they measure at 8.2-8.3 Ω with inductance measured in microhenries. Close enough for my purposes, but maybe not for a serious amp tester.

For higher sustained power, you'd want to use low heat density foldback style elements. Obviously these are more expensive.

Of the choices available, the closest to exactly 8 Ω turns out to be some 6000W 220V elements, which you can get from AliExpress. Ohm's Law says those should work out to 8.07 Ω. And even then you're subject to manufacturing tolerances.

RV antifreeze is fine, but few liquids absorb heat like plain old water. Oil is overkill - these are water heater elements, and you don't hear about people getting electrocuted in the shower!

 

[SVChucko]

Here's a list of common water heater elements I came up with when I was researching the dummy load.

Water heater elements ordered by resistance

(Note: tempco of Nichrome resistance is positive, at roughly
.04%/degree C. So the wire temperature would have to increase
significantly to change more than a couple percent.)

3500W 240V nom. 16.46 ohms
2 in parallel should be very close to 8 ohms
VERY common and cheap ($6-10 each): high watt density screw-in -
Camco #02282/02283
Utilitech #362342
Everbilt #15010
Eastman #60122n
Rheem/Ruud #SP10552KH, UV12896 ($6.16 each direct from Ruud!)
Less common, more expensive: Medium/low watt density foldback style -
Camco #02523, #02912/02913
Rheem/Ruud #AP12903, SP10869KL, UV12903

3800W 240V nom. 15.16 ohms ($10 at HD, qty discounts)

4500W 240V nom. 12.8 ohms (very common, $10-11 @ Lowe's, $12 @ HD)

5500W 240V nom. 10.47 ohms ($12 @ HD, $15.48 @ Lowe's) -
4 Lowe's (Utilitech brand) units measured ~9.9 - 10.2 @ room temp

1440W 120V nom. 10 ohms ($12 at HD)

1500W 120V nom. 9.6 ohms ($7.28 @ Lowe's, $9-10 at HD, qty discounts)

6000W 240V nom. 9.6 ohms ($33 @ HD, Camco #02612/02613 $12.11 @ Amazon)

6500W 240V nom. 8.86 ohms (Dernord brand $30 @ Amazon)

1650W 120V nom. 8.73 ohms (e.g. Grainger #2E754, Camco #02482) $17 on eBay

5000W 208V nom. 8.65 ohms (Camco #05652, Ruud #AP10867NL-5)

1750W 120V nom. 8.23 ohms (Camco #02182/02183) ($13 @ Amazon)

6000W 220V nom. 8.07 ohms (Aliexpress, $13-18)

5500W 208V nom. 7.87 ohms (e.g. Camco #02862) $26 + ship @ eBay

6000W 208V nom. 7.21 ohms (Camco #05692)

2000W 120V nom. 7.2 ohms ($10 at HD, qty discounts)

9KW 240V element nominally 6.4 ohms ($41 at HD, for tankless)

3500W 120V nom. 4.11 ohms ($34 at HD, for tankless)

 

[ddude003]

RF dummy loads and elrctrical transformers have been using oil for centuries. It's a non problem.

It is a joke... I'm sure everyone has a Halon System in their Home Theatre and Fluorinert is about the most expensive dielectric there is... After I realized that Load was using water heater elements, its not about being dielectric it is about being a heat sink... Although, I can remember working on a computer that used Fluorinert for both its dielectric and heat exchange properties... Distilled Water with Water Wetter works for me when only heat exchange is necessary... Have fun don't get any of that RV Antifreeze on your shoes...

 

[linearphase]

So on the back side of the terminal, parallel another jump to a second terminal, so that I have two pairs of terminals for each resistor.

View attachment 41885

Yes parallel all four ( horizontally) sets for two ohms. Just two for 4 ohms. You can also put the left two in series, the right two in series and the parallel those two series connections for double the power 4 ohms. Other combinations are of course possible. The formula for resistors in series is Rtotal = R1+R2 for resistors in parallel it is Rtotal=1/ (1/R1+1/R2) you can extend other equation to include any number of resistors. For example three 8 ohm resistors= 1/(1/8+1/8+1/8)=2.67ohmms by using combinations of series and parallel you can arrive at many different impedances.

A simple calculator s here:
parallel-and-series-resistor-calculator.html

 

[Sonnie Parker]

Here's a list of common water heater elements I came up with when I was researching the dummy load.

Water heater elements ordered by resistance

(Note: tempco of Nichrome resistance is positive, at roughly
.04%/degree C. So the wire temperature would have to increase
significantly to change more than a couple percent.)

3500W 240V nom. 16.46 ohms
2 in parallel should be very close to 8 ohms
VERY common and cheap ($6-10 each): high watt density screw-in -
Camco #02282/02283
Utilitech #362342
Everbilt #15010
Eastman #60122n
Rheem/Ruud #SP10552KH, UV12896 ($6.16 each direct from Ruud!)
Less common, more expensive: Medium/low watt density foldback style -
Camco #02523, #02912/02913
Rheem/Ruud #AP12903, SP10869KL, UV12903

3800W 240V nom. 15.16 ohms ($10 at HD, qty discounts)

4500W 240V nom. 12.8 ohms (very common, $10-11 @ Lowe's, $12 @ HD)

5500W 240V nom. 10.47 ohms ($12 @ HD, $15.48 @ Lowe's) -
4 Lowe's (Utilitech brand) units measured ~9.9 - 10.2 @ room temp

1440W 120V nom. 10 ohms ($12 at HD)

1500W 120V nom. 9.6 ohms ($7.28 @ Lowe's, $9-10 at HD, qty discounts)

6000W 240V nom. 9.6 ohms ($33 @ HD, Camco #02612/02613 $12.11 @ Amazon)

6500W 240V nom. 8.86 ohms (Dernord brand $30 @ Amazon)

1650W 120V nom. 8.73 ohms (e.g. Grainger #2E754, Camco #02482) $17 on eBay

5000W 208V nom. 8.65 ohms (Camco #05652, Ruud #AP10867NL-5)

1750W 120V nom. 8.23 ohms (Camco #02182/02183) ($13 @ Amazon)

6000W 220V nom. 8.07 ohms (Aliexpress, $13-18)

5500W 208V nom. 7.87 ohms (e.g. Camco #02862) $26 + ship @ eBay

6000W 208V nom. 7.21 ohms (Camco #05692)

2000W 120V nom. 7.2 ohms ($10 at HD, qty discounts)

9KW 240V element nominally 6.4 ohms ($41 at HD, for tankless)

3500W 120V nom. 4.11 ohms ($34 at HD, for tankless)

Thanks for all the info. Pat Brown at SynAudCon said he found some at Home Depot that actually measured 8 ohms, but he's on a trip right now and can't give me the brand/model.

I found the 6000w/220v on Amazon, but how in the world would you wire them?