|Elliott Sound Products||Project 124|
A dummy load is essential for testing amplifiers, and although there is really very little involved, setting one up properly (and cheaply) can become irksome. This is one of the simplest projects on the ESP site, but it will perform extremely well. All that's involved is a bunch of 3.9 ohm 10W resistors, which can be cooled using a variety of methods.
The circuit is shown in Figure 1, and each 4 ohm load section uses 9 x 3.9 ohm resistors. This combination has a free air power rating of 90W, or 180W when connected as an 8 ohm load. The two 4 ohm sections can be paralleled to give 2 ohms (at 180W). You can expect the resistors for the basic load (one section as shown) to cost well under $40 or so. This is actually very cheap when you consider the power that can be dissipated if the resistors are well cooled.
My load is oil cooled, and has survived every abuse I've ever been able to throw at it for over 30 years. It still works perfectly, so you can see that this is not a "quick-fix" project - it should last a lifetime if well made.
You will see that the impedance presented is actually 3.9 or 7.8 ohms, and this allows for the extra resistance that will be included due to wiring and connectors (typically banana plugs for connection). Even if all wiring and connectors are very low resistance, you will still end up closer to 4 and 8 ohms than you might imagine. It is unrealistic to expect that the load, connectors and leads will always be exactly the desired impedance, and unless you are doing high precision laboratory testing, the small error is of no consequence.
I recommend that you build two of the sections shown, as this enables you to have a stereo load when needed, but also allows you to use the two 2 ohm sections in series for a 4 Ohm 720W load. You also get a 1 ohm load if needed. Remember that the power dissipation quoted here is in free air - if cooled, dissipation can be increased to at least double the free air rating of the resistors. At various times, I've run mine at up to 5 times the resistor ratings, with no ill effects.
There are several approaches that may be applied for cooling. The very best is water, but it is somewhat inconvenient because the water needs to be emptied after use, and/or needs to be topped up regularly. Water will remove the most heat from the resistors, and because the audio signal is AC, there is no issue with electrolysis and subsequent corrosion of resistor leads etc. Direct DC power supply testing must never be performed with a water-cooled load! The container can be plastic, provided it can withstand a continuous temperature of 100°C without losing strength. Do not be tempted to use automotive radiator coolant (glycol). These coolants are highly conductive, and may be corrosive on some metals - especially if you are likely to use the load for DC testing, even at low DC voltages.
Another alternative (and the one that I've used in my own load) is to use a light grade motor oil. It's not as effective as water, but there's no need to empty it out or top it up as it doesn't evaporate. It's also perfectly safe to use for testing DC power supplies. Oil is non-conductive, so corrosion isn't an issue. An over temperature cutout (such as a thermal switch) is highly recommended, and the container must be metal, with a closely fitting lid. Remember to allow an air space above the oil to allow for expansion!
Oil can be heated to insane temperatures, at which it becomes extremely dangerous. Not only is there a risk of fire, but if you were to come into contact with oil at perhaps 150°C or more, you can be assured of extraordinarily nasty burns. The thermal cutout should either disconnect the load (difficult, because of the number of possibilities) or operate a warning lamp and/or buzzer if the oil temperature exceeds 100°C - although a lower temperature is preferable.
Naturally, there is always a risk of an oil filled load being knocked over. The mess is not only unpleasant, but is extremely difficult to clean up. Any oil cooled load unit should be firmly mounted to a non-flammable surface, away from any other flammable materials, and well out of harm's way so it cannot be touched accidentally. I know how hot it can get from personal experience, and my load uses almost 4 litres of oil, which takes a lot of power for quite some time before it gets really hot.
Finally, the resistor bank can be air-cooled, using one or more 12V fans. Since the fans draw only a small current (around 200mA or so is typical), they can be run from a standard 12V plug-pack supply. These are available cheaply from most suppliers. If you wish, the resistors can be clamped to a length of heatsink, or you can use metal-clad resistors bolted to a heatsink (great alternative, but considerably more expensive). A heatsink gives the fan more surface area to blow on, and increases the thermal dissipation. It also turns a relatively cheap project into a rather expensive one, but it will give many years of service if done properly.
No PCB is needed, but I do recommend that you use some kind of frame to mount the resistors (assuming you don't use a heatsink). Make sure that there is clearance between each resistor, so there is plenty of room for water, oil or air to circulate. Wirewound resistors can withstand at least double their rated power for short periods, but by cooling the resistors you can extend that period to as long as needed for most test procedures.
Make sure that all leads are mechanically joined by twisting before soldering. A frame made from blank fibreglass or high temperature plastic (such as acetal) can be used for water or oil cooled loads, but it should ideally be metal if the load is going to be air cooled (with or without a fan). The method you use to insulate the connections is up to you, but insulation must be rated for at least 120°C.
Keep all wiring as short as possible. This is not for any esoteric reason, but simply to minimise stray resistance. The black dots on the diagram below indicate banana sockets, and the lines should be drawn on your panel so you know what is connected to what.
The panel layout shown above allows you to use short, stout leads with banana plugs to configure the load, and is an almost exact copy of the one I've used for many years. For low power 4 ohm tests, you can use any single 4 ohm section (90W each) - take your pick. Otherwise, use the configurations listed in the table below ...
|Impedance||Link ...||Amp to ...||Comments|
|16 Ohms||2 & 3||1 & 4||360W Mono|
|12 Ohms||C1 & C2||1 & 3||270W Mono|
|8 Ohms||N/A||1 & 2 &/or 3 & 4||180W x2 Stereo|
|4 Ohms||1 & 2, 3 & 4, C1 & C2||1 & 4||360W Mono|
|2 Ohms||1 & 2 &/or 3 & 4||1 & C1 &/or 4 & C2||180W x2 Stereo|
|1 Ohm||1 & 2, 2 & 3, 3 & 4, C1 & C2||1 & C1||360W Mono|
Power ratings are for free air, and can be at least doubled if proper cooling is used. C1 and C2 are Comm1 and Comm2 respectively on the panel layout. It will only take a short while before you can simply look at the panel configuration (if laid out as shown) and be able to work out exactly what you need without even thinking about it. Naturally, the same thing could be done with switches or relays, but you'll end up with a lot of extra resistance in series and a far more complex (and expensive) project. The arrangement shown is very flexible, and has served me well over the years.
|Copyright Notice.This article, including but not limited to all text and diagrams, is the intellectual property of Rod Elliott, and is Copyright © 2009. Reproduction or re-publication by any means whatsoever, whether electronic, mechanical or electro-mechanical, is strictly prohibited under International Copyright laws. The author (Rod Elliott) grants the reader the right to use this information for personal use only, and further allows that one (1) copy may be made for reference while constructing the project. Commercial use is prohibited without express written authorisation from Rod Elliott.|