Power and impedance

[Mr. Foxen]

[quote name='flyfisher' post='1081690' date='Jan 8 2011, 12:14 AM']I don't think heat sinks are a factor because you can't get more power out of an amplifier just by increasing the size of the heat sink.[/quote]

I meant more in the style of with more heat sinking, you can run it at 2 ohms minimum rather than 4, as the excess heat generated is dealt with.


[quote name='flyfisher' post='1081702' date='Jan 8 2011, 12:31 AM']Fair point about DC theory not being strictly applicable, though it's one way of 'visualising' what's going on - to a very rough approximation anyway.

Given, as you rightly say, all the variability involved with music waveforms and speaker behaviours, it's a wonder we bother worrying as much as we seemingly do with precise 'matching' of speakers and amps.

But I guess describing a speaker as '4 ohms' or '8 ohms' also helps people 'visualise' things - to a very rough approximation. [/quote]

That is why it is 'nominal impedance', and it isn't DC, because it is impedance, rather than resistance. You get a graph and the nominal is determined from that, there is some sort of rule, which made the Barefaced big one come up as 6 ohm rather than 8ohm due to the crossover, but in the end, the impedance of the woofer was the bit that mattered.

[Bill Fitzmaurice]

[quote name='TimR' post='1081637' date='Jan 7 2011, 06:16 PM']As Bill intimated earlier. The power output is not linear. This is why you get power compression at the top end of the amplifiers range.[/quote]
Linearity in audio is rare. It isn't with speakers, it isn't with amps. If I designed amps for a living instead of speakers I'd have the reason for this particular non-linearity at the tip of my tongue, but it lies too far back in the recesses of my memory to retrieve it.

[Phil Starr]

There are amps which more or less give double the power into 4ohms and even ones which nearly double again into 2ohms but the only amps I've seen that pull this trick off are hi fi amps. All of the designs I have seen that achieve this trick have regulated power supplies. The biggest problem is that there are several impedances in the chain, power supply, amplifier and speaker leads. Under high power all the components heat up and their resistance/impedance will rise. You'll burn your fingers on a mains transformer that has been running flat out for any time.

I would think that the main restriction is the size of the mains transformer and of course, its cost. To reduce its impedance you'd have to increase the size of the windings considerably and you'd have to do something about the heat dissipation as the surface area/volume ratio decreases with size.

In the end you reach a point where the cost of the power supply is exceeding the rest of the amp by a considerable amount and the supply starts to get too heavy to lift so as a designer you start to compromise. In any case if you are using the amp to amplify music rather than test signals you are more interested in the amps short term output than its continuous power. so, yes looking to see how an amp handles into 4 or 2ohms is an indication of how well it handles high currents and the quality of its power supply only giving 70% extra into 4 ohms doesn't mean the amplifier is somehow crap.

I can't imagine there are these sorts of discussions on the guitar forums somehow.

[dincz]

I once had a power amp that put out 280W into 4 ohms. At maximum power (onset of clipping), the supply rail voltages dropped significantly (about 15% from memory). I added a second (identical) power transformer in parallel and the amp then managed 318W. Not that significant and hardly worth the effort or the added weight, but it did at least show that the transformer is a major factor.

Edited January 9, 2011 by dincz

[Lozz196]

[quote name='Phil Starr' post='1081816' date='Jan 8 2011, 09:29 AM']I can't imagine there are these sorts of discussions on the guitar forums somehow.[/quote]

Probably cos they just put everything on 11

[thinman]

[quote name='dincz' post='1082921' date='Jan 9 2011, 09:39 AM']I once had a power amp that put out 280W into 4 ohms. At maximum power (onset of clipping), the supply rail voltages dropped significantly (about 15% from memory). I added a second (identical) power transformer in parallel and the amp then managed 318W. Not that significant and hardly worth the effort or the added weight, but it did at least show that the transformer is a major factor.[/quote]

That being the case, is some of the variance in ratings into different loads due to the duration over which output is measured?

I'd imagine that with sufficient capacitors in the power supply high peaks could be achieved, but not sustained.

I'd like TimR to elaborate on what he was saying about looking at things the other way round, i.e. output is not halved by doubling the speaker impedance - just for education purposes.

[dincz]

[quote name='thinman' post='1087261' date='Jan 12 2011, 10:19 PM']That being the case, is some of the variance in ratings into different loads due to the duration over which output is measured?

I'd imagine that with sufficient capacitors in the power supply high peaks could be achieved, but not sustained.

I'd like TimR to elaborate on what he was saying about looking at things the other way round, i.e. output is not halved by doubling the speaker impedance - just for education purposes.[/quote]

Don't know about peak ratings but I expect it could have managed brief bursts of higher power. The figures I mentioned were for continuous (i.e. several minutes) operation.

Edited January 12, 2011 by dincz

[TimR]

[quote name='thinman' post='1087261' date='Jan 12 2011, 09:19 PM']That being the case, is some of the variance in ratings into different loads due to the duration over which output is measured?

I'd imagine that with sufficient capacitors in the power supply high peaks could be achieved, but not sustained.

I'd like TimR to elaborate on what he was saying about looking at things the other way round, i.e. output is not halved by doubling the speaker impedance - just for education purposes.[/quote]

I'm just trying to revise this after about 15 years. Looking over a circuit diagram of a power amp the outputs all run through parallel transistors. I can't see where the power rails have any direct connection to the speakers. All the current must run through these transistors.

Increasing the transformer efficiency may help in some way ie the designer may have compromised on the size of the transformer. Wouldn't this result in a corresponding increase in power at 8ohms as well? ie the power at 4ohms may well go up, but so will the power at 8ohms so you still won't get your double. Unless I'm overlooking something obvious. Are you saying that the power rails don't sag at all at 8ohms full power? I thought this sag something to do with the high frequency audio as well.

If the amp can do 500w at 8ohms then double that would be 1000w. Let's say you only get 70% or 700W of that at 4ohms as the power is dependent on the current that the transistors will allow to flow. This current isn't directly related to voltage at high power. The amp is marketed as a 700w amp, NOT as a 1000w one that can't get to its max at 4ohms but can get to 500w at 8ohms.

Hence you get sold an amp that does 700w at 4ohms and the bonus is it does MORE than 350w at 8ohms.

You just increase the power by adding more transistors, not by uprating the already heavy transformer.

Anyway it's been a long time and very hazy.

[dincz]

[quote name='TimR' post='1087458' date='Jan 13 2011, 01:12 AM']Are you saying that the power rails don't sag at all at 8ohms full power?[/quote]

The power supply (particularly the transformer) has resistance. The more current you draw from it, the more voltage is dropped in that resistance. If you double the current through say, a 1 ohm resistor, you double the voltage dropped by that resistor.

Changing the load impedance from 8 to 4 ohms would result in double the output current, all things being equal. Yes, the output current increases, but it has to come from somewhere - the power supply. Because of the power supply's resistance, the supply voltage drops and so the output current is determined by the new lower rail voltage and the load impedance. So you don't get double the current and you don't get double the power.

At 8 ohms, less current is drawn from the output stage and therefore from the power supply, so the drop in rail voltage is less at full power than it is with a 4 ohm load.

[TimR]

OK. I've done some more research. The power supply does have a significant impact.

But it's still the current that's causing the problems (in this case excessive volt drop).

The transformer in my amp is 640W but the amp is rated at 500W into 4ohms. There's no 8ohm rating so that's not much help.

Do the manufacturers get the transformers purpose wound or do they get them off the shelf to keep costs down?

I also have a two channel 1500W into 2ohm poweramp. Its ****ing heavy!

I'm guessing that the amp power is not restricted by having too small a transformer. More of a case that they use the smallest transformer they can that still gives the amp the power out that they require. They could probably put a bigger transformer in but I'm guessing that then they would have to redesign the amp to get more gain.