Is it bad to start an amp without a load?

Perkinsman

"I've never met a fixer that I couldn't break"
I've heard that you should use a speaker or "dummy load" on an amp under repair. Can someone explain why a load is needed and how to determine the proper size. What happens if you don't use a load or use an underpowered load? Do I also need one for testing small radios? If not, what minimum power level do I need one for?
 
Transistor amplifiers can generally be operated without a load.

Do not ever operate a vacuum tube amplifier without a load because the infinite load impedance is reflected back to the tubes via the output transformer as an even higher impedance. This can cause arcing in the tubes and inside the transformers possibly causing their failure.

You do need a load if you are going to take measurements with the amplifier delivering power. The power rating of the load resistors should be equal or greater than the power you are running for the test. However, slightly smaller wattage resistors can be used if the measurement time is limited to a few seconds, just long enough to take the measurement.
 
unloaded amp?

Yes.
There was a time when transistor amps would blow if started without a load. Very early designs.
Years ago starting anything w/o a load meant instant death.

Paul
 
. As Fred said, tube amps need one hooked up for safety. Properly stable designs won't go stupid without one, but its not worth the risk.

SS amps usually are OK, but you won't hurt anything by using one. Sometimes you'll find amps where the service manual specifically tells you to unhook anything from the output and run it straight open. If it says to do that, it should be perfectly happy running without a load.

+1 on using a load bigger than the max output power of what you're testing. An 8 ohm resistor of suitable wattage, or you may have to make a bank of several resistors to come out to the proper wattage and resistance.
 
Tube amps. need to be loaded, SS maybe.

A tube amp. has a large coil of wire tied to the output (your output transformer).

A quick surge of current in a coil generates a high Voltage spike (think of your points and spark coil in older cars).

Excess high Voltages generated in an output stage (with NO load) can jump across circuitry in UN-desired places, tube sockets, the OP tube itself, and the most expensive part, the OP transformer insulation!

I even have seen guitar amps. SHORT the output when no speaker is plugged in, play one of these "muted" amps. hard and all you get is a red plating tube but NO OP transformer damage.

This is not to say every tube amp that is powered up with no load WILL "short out", but why risk it?

Most solid state amp. are fine with no load (I would use caution with any that have a transformer as pert of the OP circuit).

Mark T. :music:
 
I'm curious as to what happens if you blow a speaker while playing through a tube amp. Lets say voice coils blow and no more load...does that lead to instant amp death?
 
Ok, appreciate those answers. Lets say I want to hookup a dummy load to my tube amp that has 2, 4, 8 & 16 ohm options and is rated at 40 wpc. How do I determine the proper ohm rating & wattage of the dummy load? Once I determine that, is there an advantage to using a store bought dummy load over using a speaker? Since most tube amps are 8 ohm output, can I just get one 8 ohm, large wattage (500 or 1000W) dummy load or even an oven element?
 
8 ohm, 40 watts or better total is what you need. I'd probably go for 100w just so you have room to grow with amplifier size. You can build one yourself with several high power resistors. Ideally, you want non-inductive types.


A speaker works, but you're not going to want to do full power testing while listening to that screaming in your ear. I did some work on a 350 watt per channel amp that requires it to be run at full blast while taking some measurements. No way am I listening to a 2khz tone screaming at me for that much time. There is also a chance there is a problem with an amp that would kill a speaker but likely will not kill a dummy load. No point risking your speakers.
 
A quick look on Amazon brings up an aluminum housed wire wound 8ohm-50watt resistor and a 4ohm-100watt resistor. The amp I'm working on is 40watts so I'm just going to go with the 8 ohm resistor and call it a day. But it brought up a thought.

I understand that resistance in serial is additive, 8 ohm +8 ohm in serial is 16 ohm. And in parallel it is total value /2. 8 ohm + 4 ohm in parallel is 6 ohm.

But what of the wattage? Is the wattage max constant? Additive? Does Serial / Parallel matter?
 
And in parallel it is total value /2. 8 ohm + 4 ohm in parallel is 6 ohm.

The formula for resistors in parallel is Rtotal = R1×R2/(R1+R2). An 8 Ohm and 4 Ohm resistor in parallel has an impedance of 2.67 Ohms.
 
.....
But what of the wattage? Is the wattage max constant? Additive? Does Serial / Parallel matter?

The wattage must be calculated individually for each resistor. If resistors are in parallel, the voltage will be the same on all of them. So, if you have a 4 ohm and 8 ohm in parallel and there is 8 volts, the 8 ohm resistor will be drawing 1 amp and will be dissipating 8V x 1A = 8 watts. The 4 ohm resistor will be drawing 2 amps and will be dissipating 8V x 2A = 16 watts.

If the resistors are in series, then the current is the same but each will drop part of the voltage in proportion to its resistance. So, if you have a 4 ohm in series with an 8 ohm for a total of 12 ohms and this time, we use 12 volts to simplify the calculations, the total resistance is 8 + 4 = 12 ohms, the current is 1 amp. The 4 ohm resistor will see 4V x 1A = 4 watts and the 8 ohm resistor will see 8V x 1A = 8 watts.
 
Thanks everybody. And I feel the fool. My high school self is shaming my 45 year old self on rTotal calculation. As soon as I saw it I dope slapped myself. DOH!
 
Ok, so for testing purposes, I want to match impedence of the amp with the dummy load, but I can go at least the same or larger wattage with the dummy load? As far as connections, since there are only 2 wires/connections on the dummy load, where do I connect on the amp? For testing, do I generally want to run the amp at max wattage & is that done by turning the volume all the way up?
 
The dummy load takes the place of the speaker. If it's a stereo amp, you need two dummy loads, if it's a mono amp, (like a guitar amp that has only one speaker) you only need one. Disconnect the two wires that go to the speaker and connect the dummy load in its place.

Yes the dummy load must be equal or greater wattage than the amp's output. For high-powered amps, I run a fan blowing over the dummy load, it will get VERY hot!

Now for actual testing, you need two more pieces of equipment: a signal generator, and an oscilloscope. The signal generator is used to generate a sine wave which is fed into the input of the amp. The scope is used to look at the waveform of the output signal (or at the waveform of the signal as it passes through the amp).

The amp's maximum power will be reached long before the volume knob is turned all the way up (assuming the amp is being given a decent input signal). So DO NOT hook the dummy load up and crank the volume knob all the way up, bad things will likely result since that will likely drive the amp way into distorted, clipped mode. With the scope hooked up, we can look at the output signal and see when maximum power is being reached: the lovely sine wave will start to get a flat spot on the peaks and troughs. This is clipping--the amp can't deliver any more current so the tops and bottoms get clipped off the signal. Driving the amp even harder makes that clipping get really sharp, to a point where the clipping is causing a square wave rather than a rounded sine wave. At that degree of clipping, massive amounts of power are being passed through the amp's output transistors (in the case of a solid-state amp) and you risk burning them out.

For a small transistor radio, usually I would not bother even disconnecting the radio's speaker, depending on the level of service required.
 
" by turning the volume all the way up?"

If one turns the volume all the way up, there is a good likely hood that the amplifier will be driven into clipping, depending on the level of the signal driving the amplifier.

In general this is not something that you want to do. Basic amplifier testing for repair purposes can be accomplished at less than full amplifier power. For amplifiers that require testing at full output power, then additional test equipment is needed.

Depending on the depth of the testing being done, when testing for maximum power output, one usually employes an oscilloscope, low distortion signal generator, a distortion meter and of course suitable dummy loads.

And as gadget73 mentioned, ideally one would want to use non inductive resistors for the output load for the amplifier.

The wattage rating of the dummy load is the maximum power that it can handle, not the amount of power that will be drawn from the amplifier under test by the dummy load. So using a dummy load with a higher power rating is acceptable. The amplifier under test will only produce the power that it is capable of producing and will not try to produce the rated power of the dummy load.

The power rating of the dummy load resistors is determined in part by how efficiently the resistor is able to shed, or radiate heat. This is intern in part determined by the physical size of the resistor. For example a 10 watt resistor is physically larger than a 1 watt resistor.

The point is that some of the cheap Chinese and other cheap power resistors are appreciably smaller than major brand resistors. This results in the resistors operation at a much higher temperature. Especially if the dummy load resistors are operated near their maximum power rating. This might lead to the unexpected failure of the dummy load resistor and this might not be a good thing for the amplifier under test.

Here is a link to a recent AK thread that discusses power resistors.

For me it is a matter of being safe rather than sorry, although I know a number of people that have used the cheaper, physically smaller power resistors without issues.
 
Created for testing amplifiers.

2qs822w.jpg


2mhsb9d.jpg


Dale 3,9Ω/50W 32pcs.

2 x Papst 80x80mm
 
Wow, nice job. I'm appreciating the work on your dummy load! I understand I need other test equipment, I'm just this post to dummy loads...I'll have question on equipment connection protocols in later posts.

Ok, so when testing a 40wpc stereo audio amp, does each 8 ohm speaker input get connected to a separate 8 ohm, >40watt dummy load? do you test o with both speaker inputs connected the dummy loads or one speaker at a time?
 
Back
Top Bottom