Intentional impedance mismatch questions

[osing]

Hi,

From what I have read, tube amps can usually handle impedance mismatches of up to 100% without much problem. For example, one can unplug 2 speakers from a 4x10in Super Reverb (which is designed for 2 Ohms and has 4x 8 Ohm speakers wired in parallel - parallel for a 2 Ohm load), and get a 4 Ohm effective load which should thus push the amp harder and allow you to get more breakup at lower volumes, and also lowers the volume since two speakers push less air than four. I have not yet done this on my Super Reverb but would like to try it soon. I do have questions that I hope someone can answer, though.

I have a multiple tap OT (2/4/8 Ohm) on my Hoffman AB763 Super Reverb (SR) where I've used the 2 Ohm tap along with a 2 Ohm speaker load (4x 10in 8 Ohm speakers wired in parallel-parallel) as per the standard SR design. If I were to wire in a switch that allows me to select which OT tap to use (similar to what bnwitt did on his Vibroverb project) and also wire in a switch that allows me to disengage two speakers thereby getting a 4 Ohm load, I could effectively do the following with the flick of a couple of switches:

A) Standard 2 Ohm OT tap and 2 Ohm speaker impedance - all speakers active
B) Standard 2 Ohm OT tap and 4 Ohm speaker impedance - only two speakers active
C) 4 Ohm OT tap and 2 Ohm speaker impedance - all speakers active
D) 8 Ohm OT tap and 4 Ohm speaker impedance - only two speakers active

I would not use a mismatch of 2 Ohms to 8 Ohms since that's a 200% mismatch.


1) Could anyone enlighten me as to whether the sounds achieved in C & D above are likely to be desirable?

2) Would C provide cleaner sound and less breakup at a given volume than the stock setting in A?

3) Are C & D essentially equivalent as far as how the tubes and the amp is pushed except that in C you have more volume due to 4 speakers being active versus 2 in D?

4) On the same note, if you were to wire a SR in series-parallel and thus get an 8 Ohm effective load and also used an 8 Ohm OT tap, would the sound be the same as in the stock 2 Ohm setting (2 Ohm OT tap, 2 Ohm effective speaker load), or does it somehow change the behavior of the amp? If the sound is the same, is there any reason why the original designers chose a 2 Ohm OT and wired the speakers in parallel-parallel instead of using an 8 Ohm OT and the speakers wired series-parallel? I realize the power tubes are designed for a certain load, but my impression is that by matching the OT tap and speaker impedance, there can be multiple ways of achieving this load on the power tubes?

Thanks!

[HotBluePlates]

1. Only your ears can decide if the impedance mismatch is desirable.

I would tend towards C instead of D, as most people would play a Super Reverb for its distorted sound, and I'd question the ability of 2 of the stock speakers to handle the power. That said, I don't know the rating of the stock speakers. However, I used to own a '67 Super Reverb, and 1 of the 4 speakers was blown without reducing the number of speakers handling the power.

2. Almost certainly not.

If you set up a pentode or beam power tube with a specific load to produce maximum clean output power, any change from that load (up or down) will result in less clean output power. You might change the balance of distortion products, but you won't get more clean. If we were talking triodes, the answer would be a little different.

One caveat: By increasing the load impedance, you will have a larger voltage present at the speaker terminals. That voltage is sampled by the feedback loop, which is presently optimized for a 2 ohm output impedance. So you will slightly increase the feedback when you switch to a higher load. 30w output and a 2 ohm load implies ~7.75v RMS at the speaker terminals; 30w and 4 ohms implies ~11v RMS. The voltage, and therefore the feedback increases by ~42%. Really, this is not very significant, but I point it out in the interest of thoroughness.

3. Yes.

But see answer 1 above; be careful about the power handling ability of the speakers.

4. The two approaches yield the same basic power and clean output, with the exception of the altered feedback, as noted above. I don't have an answer for why Fender chose not to use series-parallel wiring, but they always paralleled additional speakers when they added them.

[osing]

HotBluePlates,

Thanks for your input! I might just wire in some temporary switches so see (hear) how the sound is affected.

Regarding the speakers, I still have the Jensen P10Rs that came with the amp when I bought it (it' s a reissue that I bought used and since gutted and made into the Hoffman AB763). I believe those are 25 watts. Are you saying that by using only two at 4 Ohms, there is potential risk of damaging them?

A couple of other things I forgot to ask yesterday are:

1) For the speaker switch that would allow me to switch between two and four speakers (4 Ohm and 2 Ohm), is it critical to only operate that switch when the amp is off? If switched when the amp is on, could even the fraction of a second that the amp has no load cause damage?

2) Similarly, if I were to have a switch that allows me to change the OT tap (2/4/8 Ohm), is there any risk involved with switching while the amp is on? There would be a load present as far as the speakers being connected are concerned, so is this perfectly fine?

Thanks again!

[sluckey]

Switching while the amp is on is no problem. Switching while the amp is on and making music/noise is very bad.

[HotBluePlates]

To expand on what sluckey said...

It's not bad for the tubes to idle with no load. But when you send a signal through the output tubes, and there is no load, the inductance of the OT can cause large voltage spikes that may damage the interwinding insulation of the OT.