Please explain amplifier output impedance

Atmasphere,

Interesting. Perceived loudness clearly can be different from actual loudness. I have experienced this too. My speakers go well over 100 db continuous SPL levels at the listening position and yet the perceived sound is less loud than when my daughter plays iTunes at maximum distorted levels from the mini 10W speakers connected to our PC in the study.

In fact I have to ask her to turn it down even though the SPL levels are miniscule in the kitchen which is about 20 feet from the study; distortion is tiring, distracting and it seems we are very sensitive to low levels of it.

This may also explain why compression and limiting applied to modern pop CD's makes them sound very loud and unpleasant, especially at higher listening levels (when perceived loudness due to distortion and real loudness become most unpleasant). The CD "loudness wars" are a way to intentionally manufacture distortion in order to get a unpleasant & louder sounding music that gets everyones attention.

I read somewhere that IMD distortion occurs in the ear and this is how we perceive loudness, therefore, distortion that is added before the sound reaches our ears is interpreted as loud.

Shadorne- Yes. This is a major reason why tubes continue to be popular- they create less loudness cues.

Tvad: There isn't a specific point where an amp is either considered to be high or low output impedance as it is a relative thing. Having said that, a lower output impedance is typicall considered to be a more desirable trait in most cases. As previously mentioned, this sole criteria should not be used to judge whether one amp is "better" than another, but looked at as part of the total package when considering system synergy / component compatability.

If one wants to avoid potential problems in this area, look for speakers that maintain a relatively higher nominal impedances. Some speakers are specifically designed to stay above 10+ ohms. As you probably know, these are designed to be more compatible with tubed amps that typically demonstrate higher output impedances. Sean
>

10-12-06: Sean
...lower output impedance is typically considered to be a more desirable trait in most cases.

In your view, what measurement is considered to be a lower output impedance?

BTW, I recently did an extensive search for loudspeakers. I'd say the list included 20 loudspeakers from 20 manufacturers. They were all dynamic designs. Only one model had a nominal impedance above 10 ohms (Coincident Super Eclipse III). I don't believe a 10+ nominal impedance is a realistic goal when shopping for loudspeakers...even on esoteric horn loaded loudspeakers.

Perhaps you could offer three loudspeakers that have 10+ ohm nominal impedance?

Tvad: The speakers that came to mind right away were the Coincident's that you mentioned. I don't know of others that i can rattle off though, as i really don't pay attention to / go out of my way looking for such designs. Most all of my gear / installations revolve around lower impedance designs, hence my lack of familiarity. Having said that, my preferences in loudspeakers also tends to dictate my preferences for higher powered SS amps that run in a very rich Class AB or Class A mode.

As to output impedance on an amp, my personal opinion ( and that is all that it is ) would be to "draw the line" at about .2 of an ohm or so*. Obviously, there's a bit of a "fudge factor" involved here and what i would consider to be "acceptable" would somewhat depend on the load that it would be driving. Obviously, i would like to see it lower than this, but with some designs, that would require the use of massive amounts of feedback. At that point, it becomes a balancing act as far as which evil you want to subdue the worst.

With that in mind, most tubed amps are noticeably above this level, hence the looser low frequency response and greater variations in performance from speaker to speaker. I have seen some tubed amps that exhibited output impedances that were quite high i.e. above 8 ohms on certain taps. These amps would tend to be extreme sonic chameleons i.e. drastically changing their bandwidth, transient response and tonal balance as the load that they saw was varied. It would be a situation like this where some speaker cables can DRASTICALLY make very audible differences. Sean
>

* For those that can't do the math, an output impedance of .2 ohm would produce a damping factor of 40 as referenced to an 8 ohm speaker. This would be an acceptable starting point for someone trying to drive a larger woofer with a decent sized motor structure. Smaller diameter woofers with smaller motors and / or limited excursion might get away with a slightly lower DF ( damping factor ) without any really noticeable problems.

In some cases, a lower damping factor might actually be desirable, as it might lend a fuller tonal balance to an otherwise thin sounding system. Only problem is, that loss of DF would typically come at the expense of transient response, so once again, system synergy comes into play.