Driving Low Impedance Loads

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have some old (vintage) amplifiers which are switchable between Class B (120 W) and Class A (30 W). These amps are not rated for Class A operation into a 4-ohm load.

What should I expect will happen if I drive them like this anyway? What are the likely limiting factors?

Thanks and regards,

Rob
 
Depends. Is there something that expressly implies you shouldn't use 4 ohm speakers or is it just that the power rating is stated in terms of 8 ohms? Sometimes a manufacturer just lists the 8 ohm rating and doesn't bother with the other even though it is perfectly ok.

In any case the modst obviouis result of switching to class A is that it will put out a lot of heat. With 4 ohms it will put out even more.

That's all generalities. Why not name the amp there is a fair chance someone here might know something specific about it.
 
For a class A amplifier, the Thevenin circuit is like an AC current source in series with the output impeadence, in series with the load impeadence(speaker). If the output impeadence of the amp is 8 Ohms, then when you connect a 8 Ohm speaker (assuming nonreactive load for simplicity) half of the voltage is across the load, and the other half is across the output impeadence (output transistors), 50% efficiency not counting realistic losses, equal power dissapation. Since the Thevenin circuit is a series circuit, the current is same, and half the power is dissapated in the load and half is dissapated in the outputs. So when you connect a 4 Ohm load onto a 8 Ohm amp, more voltage is accross the output impeadence than the load. Therefore much more heat is dissapated in the amp then in the load. The overall impeadence is lower so more current flows and this creates a bad scene.:hot: :hot: :smash: :smash: Not to mention the outputs go into saturation and distortion you will get. In order to get maximum power transfer, you need to match the impeadence as close as possible to the output impeadence of the amp. Lowering the output impeadence requires a circuit redesign... or impeadence matching transformer.:xeye: :apathic: :whazzat:
You can always experiment to see if it will drive a 4 Ohm load, but you will want to be careful and monitor the circuit and heat very closely. If the circuit isn't designed for 4Ohms in class A, I wouldn't do it.

Chris
 
"For a class A amplifier, the Thevenin circuit ..."
:whazzat:
This is wrong, big words but wrong.

I do not know this amp so any comments are guesses. If the amp switchs to lower power supply voltages and higher Iq for the class A setting I would expect that it will just go out of class A when driving 4 ohm loads at higher outputs - if the Iq is just sufficient for 30W into 8 then any power higher than 15W into 4 ohms would no longer be class A (simpistic calculations but probably close)
Since it is rated for 140 W into 4ohms it seems that the output section will not be taxed by the load.
Of course as I said THIS IS A GUESS because I do not know this amp.
 
For a class A amplifier, the Thevenin circuit is like an AC current source in series with the output impeadence, in series with the load impeadence(speaker). If the output impeadence of the amp is 8 Ohms, then when you connect a 8 Ohm speaker (assuming nonreactive load for simplicity) half of the voltage is across the load, and the other half is across the output impeadence (output transistors), 50% efficiency not counting realistic losses, equal power dissapation.

The amplifier in question has an output impedance well less than 0.1 ohms, and I don't believe that this is the mechanism resulting in Class A efficiency of 25%. Rather, the output transistors are dissipating heat, whether there is input or not.

I was kind of hoping for a response on how the amplifier driver circuit would run out of current in certain circumstances with low impedance loads, or some other such mechanism. There is a relationship between rated load and quiescent current for Class A which I don't quite understand, yet.

FWIW, this amplifier doesn't sound any different in Class A or Class B, but it does run substantially hotter in Class A. Either I have tin ears, or the Class B mode is actually pretty good.

I have run this amp on my low impedance speakers, and it sounded OK. I'm just wondering if doing so will result in premature failure, or cause it to clip at a low-ish power leavel.

Thanks and regards,

Rob
 
There is a relationship between rated load and quiescent current for Class A which I don't quite understand, yet.

Part of ambiguous response may stem from "rated load ", even when the marketing department keeps their hands off, this is soft figure.

Nonetheless, I took a look at D. Self's essay on Class A. Indeed, a 4 ohm load requires almost doubling the quiescent current. His example was for a 20W amp and applioed only to a push-pull type, which yours must be if it can be swithed into blas B. He claims that figuring out the required quiescent current is lengthy and omits the details. However, a little extrapolation suggests to me that ~1.7A for 8 ohm and ~3.2A for 4 ohms are likely figures for quiescent current.

For most but not all "4 ohm" speakers, 4 is as low as the actual impedance actually goes and it is actually higher for much of the audible freq range.



FWIW, this amplifier doesn't sound any different in Class A or Class B, but it does run substantially hotter in Class A. Either I have tin ears, or the Class B mode is actually pretty good.

They both have the same input/small signal circuitry which are very critical for noise and distortion of many kinds. After all, any anomolies that reach the signal there (especially near the input) are going to be amplified. The advantage of class A relates only to crossover distortion so for better or worse the two modes will have the same merits or demerits for most possible sound impairment mechanisms. The couple of charts I've seen for amps that can switch between modes show that the Class A advantage is most pronounced at high frequencies - you wouldn't expect much advantage in the mid range or lows.
 
The one and only
Joined 2001
Paid Member
cunningham said:
For a class A amplifier, the Thevenin circuit is like an AC current source in series with the output impeadence, in series with the load impeadence(speaker

Don't you mean a current source in parallel with the output
impedance, or alternatively, a voltage source in series with the
output impedance?
 
cunningham said:
For a class A amplifier, the Thevenin circuit is like an AC current source in series with the output impeadence, in series with the load impeadence(speaker)... Lowering the output impeadence requires a circuit redesign... or impeadence matching transformer.:xeye: :apathic: :whazzat:

B Cullingford said:
"For a class A amplifier, the Thevenin circuit ..."
:whazzat:
This is wrong, big words but wrong.


What B Cullingford said!!!
Cunningham, this sounds like you are referring to a triode tube amp, and the one in question is as far from that as you can get.
Even then, since tube amps do use matching transformers, the picture you paint is somewhat inaccurate. The inefficiency of class A stems from the need to have idle current equal to full load current. In case of a trode amp, there is the Rp to consider in all this but we are talking solid state here.

As far as running the amp class A into 4 ohm loads, you will not get anything terribly problematic. Your amp will run out of idle current at 1/4 of the class A rated power into 8 ohms, after that it will go into class B.
The difference is that the output current in class B will continue to rise to accomodate the lower impedance load, while there is still theoretically half of the supply voltage on the currently conducting output device(s). This means more power will be dissipated.
This does to a large degree depend on what topology is used, as it would be different in current steering or current limiting output stages, the former would generate less heat. Either way, you will get more heat in 4 ohm mode, the relative increase would be more for the current steering topology. So, the question remains if the cooling is up to par. Either way, higher temperatures mean shorter component life, and there's no way to avoid this.
 
moamps said:


Power ratio is Z1/Z2=1/2 for the same idle current value.

Regards,
Milan


Yes you are right of course, sorry - I meant compared to power into 4 ohms (theorethically 2x at 8 ohms), but that's what you get when you edit part of the sentence without re-reading it first ;)
Nothing like embarrasing oneself in front of the world over the basics ;)
 
Administrator
Joined 2004
Paid Member
Hi Rob,
I am a bit familiar with your amp. I did authorized warranty service on those at the time. I have the SC-9 preamp, one of the best I have heard. The amp is very attractive to look at.

Your SM-6 is not really class A, just really high bias. It does sound just as good in lower bias class AB as it does in "will not last as long" higher bias class AB. It uses a freon cooling tube as I recall. Don't operate it off the level.

One thing I'd do is check the PCBs for bad solder connections. They will look rough and gray instead of smooth and silvery gray. Leave it in the cooler position for bias. (not "class A") Because many parts run hot on those, I purchased a 300DC instead. Most SM-6's I see are bad connections or failed outputs and other transistors due to heat. Be sure to leave lot's of room over the amp for cooling.

-Chris
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.