4 ohm taps or 8 ohm taps?


I am driving a pair of Swan Diva 6.2 speakers with a MingDaMC34AB vacuum tube amplifier (75 wt/channel ultralinear mode). I have liked the sound for the dollars invested.
The speakers are rated at 6 ohm impedence. I have been using the 4 ohm taps on the amp. I was recently advised to use the 8 ohm taps. I did, and swear the speakers sound too bright, "tinny", and even lost some speaker transparency.
Question One: Am I nuts or is this possible?
Ouestion Two: If it sounds better with the 4 ohm taps, is there any harm or danger to the system of using them in this system?
Thanks for your help and opinions.
R Sasso
rsasso
Almarg,

Excellent posts and well written clarification of what happens in this situation.
The 4 ohm tap puts out lower voltage than the 8 ohm tap, which would result in less current flowing into a given load impedance.

I should have added to this statement that you will tend to turn up the volume control a bit to compensate, i.e., to obtain equal volume regardless of which tap you are connected to. However, that does not change the rest of what I said above unless you were to clip the amplifier to a significant degree, which would be plainly evident as distortion.

Regards,
-- Al
Contrary to what you may have read or heard, I would expect (with one irrelevant exception described below) that for any given speaker there would be LESS strain on the amp using the 4 ohm tap than using that same speaker on the 8 ohm tap. Although as I and others have said, it is extremely unlikely that you would have a problem either way.

The 4 ohm tap puts out lower voltage than the 8 ohm tap, which would result in less current flowing into a given load impedance. Also, the load impedance seen in the plate circuits of the output tubes would be higher (again meaning less current flow) if the speakers are connected to the 4 ohm taps rather than the 8 ohm taps (since the load impedance seen on the primary side of the output transformer is equal to the load on the secondary side multiplied by the square of the xfmr turns ratio, the stepdown ratio to the 4 ohm taps being greater than the stepdown ratio to the 8 ohm taps).

The irrelevant exception I alluded to would be if the speaker impedance at some frequencies dropped down to outlandishly low values, say 1 ohm or so. In that case the output impedance of the amplifier would become a significant determinant, and perhaps the major determinant, of how much current would flow from the amplifier. And since that output impedance is lower on the 4 ohm taps than on the 8 ohm taps, greater current might flow from the 4 ohm taps in that situation. However, the only speakers that I am aware of that have impedance curves dropping down that low are certain planar speakers like the 1980's Apogees. I am not aware of any box speakers that do that, and certainly none that are rated at 6 ohms nominally.

I suspect that those who may have asserted the contrary are misled by the fact that a lower impedance speaker, or a speaker that drops down to low impedances at some frequencies, which would typically be connected to the 4 ohm taps, represents a more difficult load than a higher impedance speaker, which would typically be connected to the 8 ohm taps. But here the speaker is not a variable -- we are talking about connecting the same speaker to one tap or the other.

Finally, take a look at the video Pacific Valve has about your amp: http://www.pacificvalve.us/MDMC34AB.html. Fwiw, it is described as an amp that "loves difficult loads."

Not to worry!

Regards,
-- Al
If you are really worried about impedance dips and overstrain then maybe invest in a pair of Paul Speltz's zero autoformers which raises speaker impedances and lightens the load the amp(s) are driving.
I think he has a in home 30 day trial period so if not interested you can return them.
To Stringreen

So, a number of folks have suggested that if the speaker impedance curve drops below 4 ohms at places, and I am using the 4 ohm taps, it risks overstrain on the amp and amplifier damage. Is that an unrealistic concern, even if I am listening in the triode or Class A mode where output power is about 40 wt/channel? (I obviously can't find an impedance curve for these speakers)

Thanks
I agree with Al;it has always been my experience to use all available taps and then use the one that sounds the best to your ears.
Al's response covers almost all of the substance on the impact of different taps, save one issue:

Some amps have "good" sounding taps and "bad" sounding taps, regardless of the speaker load presented them. Variation in output transformer design and manufacture can be material, varying from model to model and even between examples of a given model. From your post ("tinny" treble), I wouldn't be surprised if the issue turned out to be sub par 8 ohm taps on the Ming Da.

Good Luck

Marty

BTW, the issue of manufacturing variability extends beyond taps (though it's less commonly a real problem, IME). If you test enough pre-amps, you'll eventually encounter it. Two different line level inputs on the same pre should be identical sounding, but -sometimes- this isn't the case. IME, the "sometimes" is much more frequent on output taps than on line inputs.
I found an interesting forum response at this url......http://en.allexperts.com/q/Audio-Systems-835/matching-amp-speaker-impedence-1.htm
Actually, I think there is a better consensus (though far from a perfect consensus) on this question than on many other issues in audio. The consensus being to go with what sounds best. And no, you won't hurt anything either way.

Re question one, there are many reasons why it is possible, led by the fact that the speaker probably deviates considerably from 6 ohms at many frequencies. Also, the higher output impedance of the amplifier on the 8 ohm tap will accentuate the effects of the variations in the speaker's impedance vs. frequency curve, and will also result in a lower damping factor and therefore lessened bass control. Also, the plate circuits of the amplifier's output tubes will see differing load impedances depending on which taps the speakers are connected to, with resultant effects that will vary depending on the amplifier design.

The bottom line is that the amplifier-speaker interface is complex, and both components differ from idealized models in many ways, meaning that the only way to reliably predict what is best for specific components is to listen, and perhaps to compare notes with others who have used the same or similar components.

Regards,
-- Al
You hear what you hear, if that's nuts then we're all looney. I would contact the speaker manufacturer on the second question, they can give you the correct answer. You will get some responses to your post here telling you it's okay and others telling you it's not okay and then it usually(not always) turns into a pissing contest between posters. Sad but true.