Perhaps common knowledge to anyone that builds speakers and measures impedance or looks at the impedance publication for the driver(s). But it's a very good point to bring up regardless because lots of folk do things without measuring and think "4ohm" is "4ohm" for the whole response which its clearly not when you look at the manufacture publication. Even if you don't or cannot measure, it's already given to you on the spec sheet if you just look at it. But, not all of them give that info either, or are older and don't have publications, etc. So, always a good idea to measure them if you can.

Example, cheap common driver following. Notice the impedance/phase graph? It's showing you that it's only at 4ohm (the published impedance of 4ohm) at 5hz and 70~100hz range bandpasses. Way higher everywhere else.

All of the following with a GRS 8SW4HE driver just for reference.

I recently did this with two drivers on a 4ohm stable amp. I tried series and then parallel. But I only did the parallel wiring to the amp because my impedance measurement confirmed I would never actually be showing a low impedance to the amp to stress its power supply out.

So here's the two drivers, 4ohm nominal impedance (above) wired in series for 8ohm. Notice in reality it's 9ohm at the lowest at 5hz and always higher than that, at all higher frequencies, much higher. Only gets down to 10ohm around 70~100hz. Super stable to use this. But also, lower output potentially for most low power amps.

So, what if I wire them in parallel, for 2ohm? Dangerous for some amps maybe. But, let's look at the actual measurement of impedance:

It's not 2ohm. It's practically 3ohm at 5hz. It's over 3.2~3.3ohm at 10hz, it's over 6.5ohm at 20hz, higher above that. It's only down to 4ohm again by 50hz, and dips to 3.1~3.2ohm from 70hz to 100hz bandpass. So in reality, this is totally fine on a 4ohm stable amp that isn't attempting to blast 5hz content (which, is exceedingly rare in real content anyways, so not worth worrying over, plus most amps just don't have high output at 5hz anyways too). Even the 10hz here is fine for a 4ohm stable amp. So even though these two drivers are wired 2ohm parallel nominal, they're anything but 2ohms. They're still totally fine for a 4ohm stable amp this way and we can see exactly where and what value they are with an impedance measurement.

Very best,

 

Thanks! That's a great example and helps me quite a bit. Still learning here so it all helps.

 

Certainly a worthwhile post. There are plenty of posts on the forum about how to "match" an amp to speakers based on impedance. The posters typically don't understand that the published impedance spec for a speaker is not a constant. It's nice to have a clear explanation as above.

 

The basics jive with what I think I know, and in terms of general application, for the general consumer, I think it is good information. In terms of impedance, I think anybody with a DATS or other method of generating an impedance sweep has probably already figured that out (or should have), and can easily confirm it.

However, that being said, not to misconstrue what you are saying, but if somebody got the impression that impedance doesn't matter, I think that would be incorrect. While it may be outside of general purpose application and mainstream use cases and performance requirements, some amps are impedance sensitive in terms of how well they perform, not just in terms of overheating, shutting down, or blowing up output devices.

Also, I don't know that the suggestion is a bad one, because I don't know how any given AVR is actually implementing the output 'limits', but if they were doing it by switching the output tap on the mains AC power supply transformer, that would hypothetically increase the current delivery capability of that portion of the power supply, which hypothetically could be of some use if the power supply was otherwise dropping voltage sag on the recharge cycle because of that limitation. Again, I don't know, but it would be nice to see detailed data on that, i.e., multiple measurements across variations of impedance (down to maybe 2 ohms) and frequency, and that data might be specifically relevant to a given model, depending on the specifics of the implementation of the limiting method, the power supply capabilities, etc.

The point that content is not a continuous test signal is a good one, and valid, though the degree of difference will depend on THE CONTENT, and the advantages of this as far as lowering prolonged current demand will similarly vary depending on the interaction of the content, the amplifier, and the loudspeaker. I.e., if you had something running a 10 Hz square wave at a high level (like EOT famously), into a loudspeaker impedance that was low at 10 and 30 hz, the larges frequency components of that signal, that could continuously be WORSE than equivalent pink noise peak voltage or continuous single sine wave signal, if it is bother larger in actual RMS value than the peak voltage would imply, compared to a nominal voltage noise signal or continuous sine wave, and with many loudspeakers might be dumping of pile of voltage into an impedance that just sucks up current on the best of days.

 

My electrostat speakers go down to 0.8 ohm at 20khz. This doesn't mean it's hard to drive. On the contrary it is quite easy in term of needing enough power.

But trust me when I say it's incredibly hard to match a high enough quality amp for it to sound as good as it can possibly be. I've tried many. Only 1 succeeded.

To further elaborate on this amp, have a look at it, a boat load of capacitors and a 2,000 watt transformer. Idk what it means in terms of performance but some of y'all might. In terms of power, nothing unusual. 200 watt at 8 ohms.

 

Contrary to popular opinion, not all 200w amps are alike.

 

Maybe the old beast I used for my AR9 LS ( 4 Ohm Nominal, 3.2 Ohm min) 5 driver 4 way towers in the early '80's would work.

The Dynaco 416 was one of the early high current designs ( by brute force, not sophistication) with 8 power transistors per channel, massive heat sinks and a 2 speed fan. On top of that, when you add the optional C100 capacitor pack ( Which I did), you get an extra 100,000 microfarads ( total) of capacitance.

It was rated down to 2 ohms, putting out 450W per channel at that impedance. At 8 ohms, it was 200W, at 4 ohms it was at 300W and was conservatively rated.

Great Amp for big speakers that are difficult to drive and near bullet proof.

 

I love me that sweet vintage stuff!

 

Basically, speaker ohm ratings listed by the manufacturers are meaningless. Also, if your amp has a 4-8 ohm switch, never put it on 4 regardless of your speakers. They only add this to get around certain testing requirements. Doing so could cut the output by half.

Can confirm, have measured this myself.

On Denon models, the 4 ohm mode is hidden and is really only there as a checkbox item. If you want to cut power in half, use eco mode as it does the exact same thing in a more configurable way.

 

Did you try one of the Quad Current dumping amps? (606, 707. 909, Artera)

"Unconditionally stable into any load " was always part of the spec, and of course they were designed to mate to the Quad electrostatic speakers.

Worked very well with my 1.6ohm Gallo speakers too

No I haven't had the pleasure to try it. The Sanders Magtech is famous for driving very hard planars too.

A lot of fantastic speakers are weirdly hard to drive.

 

You are incorrect with regard to a 4 Ohm selection on an amplifier cuts power by half compared to the 8 Ohm selection.

The 4/8 switch directs more current to the part of the circuitry design that causes the output stage to be able to handle a higher load impedance.

Current and power are directly proportional. Current and resistance are inversely proportional. That is Ohms Law. It seems to be where many get confused.

 

Are you talking about tube amplifiers here? On AV receivers it is very much the case that putting the impedance switch to 4 ohm gets you less power, when using 4 ohm speakers (which most speakers are, ignoring the fact that it varies with frequency). Audioholics has a good article explaining this. Tube amps are different beasts.

Edit: by the way, the difference is sometimes much more than half. There are examples of AV receivers delivering less than a quarter of the output when set to 4 ohm.

 

I've been doing a lot of research lately trying to learn more about how speakers and amps work so I can better understand building my own. I am very new to all of this so maybe this info is not eye opening for many here but I thought I would share. If you can. take the time to watch this guys video explaining how speaker and amp manufacturers list their impedance ratings.

Basically, speaker ohm ratings listed by the manufacturers are meaningless. Also, if your amp has a 4-8 ohm switch, never put it on 4 regardless of your speakers. They only add this to get around certain testing requirements. Doing so could cut the output by half.

Hi. I am happy you found our video useful. In this series we try to simplify topics that can be complicated for the average user. That means that there is of course deeper levels to the topic, but I think the main points of this video is important, and it’s important to spread this knowledge to a wider audience.

There are of course edge cases, but for “mainstream” systems, I strongly believe that the best the average user can do about impedance is to ignore it. Impedance is highly relevant to system designers, and highly irrelevant to normal users who connect a typical dynamic speaker to a typical amplifier or AV receiver.

The “cheat code” is: 1: assume all speakers are 4 ohm, and 2. assume all AV receivers should have the impedance switch on the highest setting available - although this seemingly contradicts rule 1.

As soon as you introduce tube amps, ribbon or electrostatic speakers, or multiple speaker pairs per output into the mix, you are no longer what I consider a “typical user”, and now this rule of thumb no longer applies.

 

I think the only caveat is some of the older amplifiers that were designed to run 12 and 16 ohm speakers which were prevalent in the late 60's and 70's.

Putting some nice new 4 ohm or even 8 ohm speakers on such an amplifier has been known to damage them.

As a matter of practice I always check specifications first.

In the last year I've done some accurate evaluation and comparison between 4, 6, 8, 12 and 18 ohm loads. I can definately say sound quality at the same (ie adjusted to match) volume level varies. Sort of works like a loudness attenuator. Low Ohm has a tendency to sound like the loudness attenuator is on and higher Ohm tends to sound like loudness is off but it slides out to more than the base line ie 12 and 16 Ohm puts more airiness and spatialness to the sound.

 

Also, by the way, I am assuming that A and B on your amplifier is two separate output channels. Otherwise it wouldn’t make sense to have separate settings for A and B output. Which amplifier is it?

 

I've been into HiFi for a long time and I own a technology company. Thanks for your info in above thread but am up with all of it.

It's a Panasonic SA-HE200 I own a few of them. They have a great sound.

 

What a bizar topic agian!
So the power is cut in half. Who ff-ing cares. You don't wanna go there! If you get within 3db of the max power you're risking your tweeters.
OK i get it this is the 'max dB' aka avs forum. Have at it!

Impedance only matters to heat production / effficency.
Wanna keep the amplifier cool? you migth consider setting the amp to a lower voltage.
If you wanna heat up the room set it to max voltage.

As long as you stay away from the -3B point of the amp maximum power you will not notice any difference.
I'm lying:
You will notice a 3dB volume change at the volume setting.
and you will waist a lot less energy at low voltage setting

 

Simple way to look at is the Impedance rating of a driver is the lowest point across it's operating bandwidth. This info is provided so you can select the proper amplifier and/or wiring config and not overload the amp with too low of resistance. That is what will burn an amp up or drive it into protection. Impedance rise will simply pull less power from the amp but it won't hurt it which is why only the lowest point is listed in the T/S specs.

 

I've always ignored speaker impedance and have told others to do the same. Ever wonder why it's almost always either 4 or 8 ohms? How is it possible that everyone builds speakers to such an exact degree? Pretty sure most manufacturers don't even measure and just state it's 8 ohms.

 

It would be impressive if a manufacturer could design a competent loudspeaker without knowing anything about its impedance.

 

This is the DIY section.. I would hope everyone here knows the exact impedance of their drivers.

 

I don't know anything about their impedance, but their impudence is well documented.

 

I didn’t actually realise it was posted in the DIY section… that is not the intended audience for my video. Impedance certainly is something a speaker designer needs to consider, DIY or not. What I said about impedance specifications being inaccurate relates only to finished speakers, I would certainly expect the specification sheets of a driver to be at least mostly accurate.

My point about leaving the amp at it’s highest impedance setting is still valid though.

 

I think maybe I posted in the wrong forum.. That could potentially have opened up a can of worms.. My apologies!

 

Basically, speaker ohm ratings listed by the manufacturers are meaningless.

Having worked both as loudspeaker engineer and marketing whore, let me testify in shame you are correct. Further, all speaker specs are useless twaddle. Sensitivity is lied about, Thiele-Small parameters rounded off, impedance has gotten lower and lower but nobody wants to say "4 ohms" for fear someone won't buy those speakers to match to their "8 ohm" receiver. And anyway the impedance only matters if you turn it way up to 11 into clipping.

 

Yup. Speaker impedance is transient. It also changes as the voice coil heats up just to throw another variable at you.

 

It also changes as the voice coil heats up just to throw another variable at you.

Lots of things change between hot states of load and cooling states of no-load. Impedance is just part of the system.

Examples I, Examples II

Very best,

 

1. This all comes down to manufacturer recommended load & importantly is the amp stable (no oscillation) into lower impedance loads (resistive + reactance)?
2. How much stress (current, voltage, temperature etc) derating the manufacturer has a applied to all power devices (transformer, rectifier, filter capacitors, output transistors & power resistors etc) when running continuously at 70% & 100% power into the recommended load. In defence designs stress derating varies but 50% is pretty common.
In general if the amp is stable, any condition less than continuous & 70% power will reduce stress & improve reliability/life as long as the amp is well ventilated.

 

This being the DIY section, I'd imagine the users and posters would be aware of several things.

(1) The effects of crossover over load impedance.
(2) The meaning of speaker impedance: resistance and reactance.
(3) The variation of impedance over frequency ( specially when a crossover is introduced ).
(4) The effects of amplifier output impedance (damping being one).
(5) The difference between different types and designs of amplifier: solid state, tubes, Class A, AB, switching, FETs, bipolars, current amps, voltage amps, feedback, etc...
(6) An understanding of amplifier ratings: marketing vs R&D

To boot, being a DIY forum... I'd expect no one is running AVRs, which are know for using marketing gimmicks in their specifications.

Otherwise, like the guy said, go to a consumer forum, buy an "8 ohm rated" speaker and keep the darn switch in your 2000 watt AVR set to 8 ohms and make sure your subwoofers are set to BOOM BOOM all day long. BTW: the existence of such a switch should be a warning ( Warning Will Robinson! ) that something is amiss with your solid state amp and its ratings.

 

Valid points. I will say however (respectfully), that yes, this is a DIY forum but being here does not automatically make all of us experts or experienced in it. I'm here as a noob to all of this so I can learn from others. So even though I am here, I did not automatically know all of (what is to you and others) the "obvious". I shared this only because for me, it was helpful as a noob and maybe it would be helpful to someone else.

 

Man this thread took some turns.. lol. Thanks everyone for chiming in!