Dummy Load calculations

[Gordo]

I hate to play the "stupid" card here but I think it points to my fundamental lack of understanding of ohms law 

I need to dummy down high power amps to dissipate output of, say, 800 watts.  Till recently I've only been concerned with 100 watt loads which is easy.  I'd assume I can then tap a line out to monitor the amp's output if necessary.  Typically I've only used a dummy for biasing till now.

Can anyone point me in the right direction?

[PRR]

> point me in the right direction?

Uh... there's a bathroom on the right?

You need a resistor rated over 800 watts at whatever impedance the amplifier is supposed to be loaded with.

Since 800W resistors are rare, you might want to know that four right-impedance resistors in series-parallel makes a right-impedance of 4 times the safe dissipation.

So four 200W parts. But those are rare.

So make each "200W" part with four 50W parts. Need 16 resistors total.

An air-cooled 50W resistor is 4" long and an inch around. Those small metal-case "50W" resistors can't take 50W for long unless mounted on a heatshink. For estimation, perhaps as big as the heatsink (and fans) in your 800W amplifer.

All of which is gaspingly costly for casual use. Get a reel of thin iron wire from the concrete and rebar section of the building supply. Measure its ohms. See if you can come close to a good value. Spread it out to shed heat, perhaps around a 2x12.

[Johan]

why would you need a 800w dummy load?   are you repairing a solid state poweramp and need to test it before shiping back to a paying customer?..
800watts sounds like a solid state amp and I fail to  see why you would run it on full and attenuate the output..just turn it down...
...so, ok that wasnt really a proper reply...
three of these in parrallell takes 900watt and would make a cup of tea in seconds..


good luck
J

[therecordingart]

http://www.parts-express.com/wizards/searchResults.cfm?FTR=dummy+load&search_type=main&WebPage_ID=3&searchFilter=dummy+load&x=0&y=0

4ohm and 8ohm 200w resistors.

[Pablo1234]

I agree with Johan its cheaper and easier to use heating elements, be aware that NCa wire increases in resistance with Temp. Personally I would make several small L pads using NCa wire each dumping about 50 watts of power, then heat isn't much an issue with the resistance, one big resistor will dump allot of heat in a small area and will have the greatest resistance change with temp. I made a Calculator in excel to calculate resistance at rated output for this wire I can post for ya. Thier is also a good web site that sells the wire and has all the data you need to calculate it. http://www.resistancewire.com/mainpage.php/

Also if thats the way you would go with this idea I recommend the largest gauge wire you can find, this will spread the heat out over a larger area of wire, also use ceramic bobbins http://www.mangalamceramics.com/ceramic_industrial_components_cordierite_refractory_tubes_bobbin_set.html to wrap the wire, this makes what is called a fire stick, for good reason. watch the inductance also, it can be handy but also could introduce several other problems.

[Gordo]

Hmmm, hadn't thought about making a coil.  Thanks folks I think this gives me a couple of options.

This doesn't happen very often but from time to time I'll see bass or power amps that need to be stressed a bit to locate a problem.  Even if I had speaker cabinets capable of tolerating that much power I hate to put additional "loads" on my sanity (WHAT?) or marriage (TURN THAT THING DOWN!!!...WHAT???) 

[PRR]

> three of these in parrallell takes 900watt


But what impedance?

Pretend the teacup coil is a resistor. (As Pablo says, it isn't a real ideal resistor.)

By the plug, I assume 230V.

300W/230V is 1.3 Amps.
230V at 1.3A is 176 ohms.
Three of those in parallel is 59 ohms.

Few stage-music amps are designed to put big power in 59 ohms; more often 8 4 or 2 ohms.

You could get a 4 ohm load with 45 teacup coils. And with 45 cups of tea the short-term power handling is 13,400 Watts, so a mere 800 Watts may not hot-up the wire enough to temperature-drift too much. However the resulting 17 Watts per cup will never get hot enough for proper tea.

In the USA, a 1500 Watt 120V skillet is near 12 ohms hot, somewhat more when cold. Three such e-skillets HOT seems to be an answer. However a 800W 4 ohm amplifier is only 56 volts, not 120V, so the skillet will be neither hot nor cold but some uncertain value.

FWIW: A 60W 120V lamp is 16 ohms cold but rises a LOT when warm. Over 40 ohms at dim-red. And a recent test of a 12V fuel-pump on a cordless drill battery with a 60W lamp gave very high resistance even though the glow was quite invisible. The pump would start and then stop in a second.

50W parts are common, therecordingart points to 200W parts with heatsinks, DigiKey shows 85 parts in-stock with values 1-16 ohm and power 200W-2,000W.

16 OHM 300W wire-wound $20, 1.125" Dia x 8.500" L (28.58mm x 215.90mm)-- four of these makes a simple no-heatsink 4 ohm 1,200W load.

Get the brackets too.

[DavenPaget]

> three of these in parrallell takes 900watt


But what impedance?

Pretend the teacup coil is a resistor. (As Pablo says, it isn't a real ideal resistor.)

By the plug, I assume 230V.

300W/230V is 1.3 Amps.
230V at 1.3A is 176 ohms.
Three of those in parallel is 59 ohms.

Few stage-music amps are designed to put big power in 59 ohms; more often 8 4 or 2 ohms.

You could get a 4 ohm load with 45 teacup coils. And with 45 cups of tea the short-term power handling is 13,400 Watts, so a mere 800 Watts may not hot-up the wire enough to temperature-drift too much. However the resulting 17 Watts per cup will never get hot enough for proper tea.

In the USA, a 1500 Watt 120V skillet is near 12 ohms hot, somewhat more when cold. Three such e-skillets HOT seems to be an answer. However a 800W 4 ohm amplifier is only 56 volts, not 120V, so the skillet will be neither hot nor cold but some uncertain value.

FWIW: A 60W 120V lamp is 16 ohms cold but rises a LOT when warm. Over 40 ohms at dim-red. And a recent test of a 12V fuel-pump on a cordless drill battery with a 60W lamp gave very high resistance even though the glow was quite invisible. The pump would start and then stop in a second.

50W parts are common, therecordingart points to 200W parts with heatsinks, DigiKey shows 85 parts in-stock with values 1-16 ohm and power 200W-2,000W.

16 OHM 300W wire-wound $20, 1.125" Dia x 8.500" L (28.58mm x 215.90mm)-- four of these makes a simple no-heatsink 4 ohm 1,200W load.

Get the brackets too.

EDIT : OOPS , they won't have enough power .

[PRR]

> recommend the largest gauge wire you can find

Larger gauge means longer length to get the desired resistance.

Fatter and longer, cost rises quickly.

Too thin will melt.

The "optimum" balances cost versus melting (and slow oxidation).

There's tables for bare copper wire melting. Steel wire resistance is less predictable (steel is not a 99% pure metal and is alloyed with different things not specified in "basic tie wire").

Long term, hot steel oxidizes and eventually burns thin.

NiChrome not only has higher resistance (less length to wrestle), it will sit at very high temperature for years because it hardly oxidizes and has a high soften/melting point.

But steel is a lot cheaper, and sold everywhere.

The ass-umed 800W 4 ohms or 14 Amps is "probably" not instant melt-down in rebar-tie steel wire IF well spread out. (And it has to be securely spread-out because it is not insulated!) I could be wrong. Gordo could bring the amp up carefully: 200W for a few minutes, then 400W, etc.

BTW, for VERY! HIGH!-power work, cast iron is a good heater. Pour it thick enough, the oxidation is not a real problem for a long time; and when it does eventually rot-through a replacement is still cheaper than NiChrome or other fancy metal.

Hmmmmm.... all common metals increase resistance when hot, and copper can double when not much above skin-burn heat. So maybe the rise of resistance will fail to fully-load the amp. Which is the whole point.

For $65-$84, the ready-made resistors are probably a much better plan.

[Pablo1234]

I agree with PRR, its the easyest rout with the $20 Fire sticks, but if you bought a spool of NiCa wire you could customize the load much more precisely and at temp. Just do L pads rated at 50 watts each and you don't need a heat sink can spread the heat out over a larger area and won't have to wory about thermal drift as much. Also you can control the amount of inductance buy the way you wind them, more or less inductance or no inductance loads.

800 W L pad 50 Watt drop for an 8 ohm load = ~ .5db

Pi pad would be 278 Rp (41.2ft each)| .46 Rs

713 Watts left
L pad -0.6dB = 112 Rp (16.6ft) | 0.53 Rs

621 Watts left
L pad -.7dB = 95 Rp (14ft)| .62 Rs

528 Watts Left
L pad -0.85dB = 78 Rp (11.6ft)| .75 Rs

434 Watts Left
L Pad -1dB = 65.5 Rp (9.63ft)| 0.87 Rs

345 Watts Left
L Pad -1.3dB = 50 Rp (7.41ft)| 1.1 Rs

256 Watts left
Lpad -1.9dB = 32.7 Rp (4.84ft)| 1.5 Rs

166 Watts left
l Pad - 3dB = 19.4 Rp (2.87ft)| 2.3 Rs

83 Watts Left
L Pad - 8 dB = 5.2 Rp (0.77ft)| 4.8 Rs



Leaves 13 Watts from the 800 total to a speaker total of 17.5 dB drop

This sounds complicated but it really isn't Rs total would be a 13 Ohm wire taped at the Rs values like a potentiometer just crimp the wire at these value points with the Rp wires to the ground. If the output is balanced then you could cut Rs total in half and add one to the negative side tapping at half the Rs values also. this will give you 50 watts per tap drop to distribute the heat over a larger area
http://www.omega.com/pptst/NI60.html Rs would require 4.8 feet of 26 gauge for a 13 ohm wire
NI60-015-50 $10
Total wire needed for Rp would be 150 ft at 30 gauge
NI60-010-200 is a 60 m (200') spool of 30 gauge bare 60% nickel/16% chromium alloy heating wire, $30.