Impedance mismatch & blackface amps...

[cerebralpaul]

I have a Super Reverb with a 2x12 baffle, right now the speakers are running 4ohms parallel. I've heard this is OK but that the amp will break up earlier and have less clarity. Is this the consensus? I'm looking for options to fix this, namely the Weber load dump or z-matcher- it looks like the load dump may be the best choice because I should be able to set it for 4ohms, max attenuation and plug it into the ext speaker jack- this will make my tranny happy while leaving the signal path to my speakers clear, no? I'm worried that the z-matcher inbetween speakers and output may adversely affect tone(which is what I'm trying to fix in the first place). Opinions?

[Old Tele man]

...with vacuum tube OT's, one-half Z to twice Z is usually OK, or typically (low (optimum) high): 2-(4)-8 ohms, 4-(8)-16 ohms, etc..

[cerebralpaul]

OK so running the 2ohm tranny in the SR at 4ohms is "safe", but how about the tone question? Does it break up sooner and/or lose clarity this way?

[Old Tele man]

...to side-step that problem, I just use one of Ted Webers (bare) Z-Matcher transformers between the 2-Ohm OT of my SR and the 8-Ohm JBL/D130F in my VibroClone cabinet...works great.

[Wakarusa]

but how about the tone question? Does it break up sooner and/or lose clarity this way?

Running into a higher impedance should reduce headroom and tube life (more of the available power is dissipated as heat in the power tubes). I seem to recall that it also affects frequency response but am drawing a blank on exactly how/why (and would have to drag out very thick books to remind myself).

[cerebralpaul]

Well it sounds like I want to get the impedance right since I'm a fan of headroom and clarity... Is there any thoughts about my idea that running a 4ohm load dump from the ext. speaker jack may be the best solution because I won't be adding anything(i.e. z-matcher) to the signal path?

[cerebralpaul]

Bump?

There isn't much difference I guess?

[John Phillips]

Well it sounds like I want to get the impedance right since I'm a fan of headroom and clarity... Is there any thoughts about my idea that running a 4ohm load dump from the ext. speaker jack may be the best solution because I won't be adding anything(i.e. z-matcher) to the signal path?
If you're after headroom that will be a bit counterproductive because you're then sending half the power to the dummy load. The loss from mismatching is not as much as half - only somewhere between 10% and 30% depending on whether you're measuring maximum power or power at a given input signal.

The best solution is to get a pair of 4-ohm speakers - Weber will do these I think. The second best is to use a Z-Matcher. I don't know what the power or tone loss via one is, but I would think fairly minimal.

[cerebralpaul]

OK cool, thanks for the help!

[Old Tele man]

The second best is to use a Z-Matcher. I don't know what the power or tone loss via one is, but I would think fairly minimal....I measured just under 1dB power loss with 50W input at 1kHz (from Super Reverb chassis)...I couldn't 'hear' any sonic losses, but comparing 4x10" D110F's (2-ohm load) to 1x15" D130F (8-ohm load) is an obvious "apples-to-oranges" comparison.

[John Phillips]

...I measured just under 1dB power loss with 50W input at 1kHz (from Super Reverb chassis)...I couldn't 'hear' any sonic losses, but comparing 4x10" D110F's (2-ohm load) to 1x15" D130F (8-ohm load) is an obvious "apples-to-oranges" comparison.
Interesting... 1dB is equivalent to about a 20% loss of power, which is roughly the same as you'll get with a 2:1 mismatch without the Z-Matcher. (Although your test is with a 4:1 mismatch obviously.)

So I think it's a bit six of one and half a dozen of the other as to whether it's better to use one or not over only a 2:1 mismatch, I would say... other than possibly keeping the amp a bit happier - although maybe not: could having another highly inductive transformer connected to the output in fact increase the risk of flyback voltage damage?

I think you'd have to try it with the same speakers to be really sure about any tone change (comparing mismatched to matched with the Z-Matcher).

But getting the right speakers is definitely the best way to go if you're after maximum headroom, and possibly clarity, since there clearly is a fairly significant loss via the Z-Matcher.

[Old Tele man]

But getting the right speakers is definitely the best way to go if you're after maximum headroom, and possibly clarity, since there clearly is a fairly significant loss via the Z-Matcher....+1, and since the 8 ohm D130F exceeded the nominal Z/2-to-2Z range, I opted for the Z-Matcher as the BEST "lesser-of-evils".

...FWIW, the loss that I actually measured was 0.89 dB, which isn't huge, but also isn't trivial.

...my concern was exactly what you mentioned--"possible fly-back" voltage--especially considering the 2-ohm up to 8-ohm loading, since going HIGHER in load is more dangerous to OT's than is going LOWER in load.

...I must admit that ANY comparison between systems having BOTH differing impedances (Z) and speakers (4x10" vs. 1x15") is rather borderline "funny"...no way they'll sound the same in the first place, plus the grossly differing effective SPL levels due to multiple-vs.-single speakers and surface areas.

[cerebralpaul]

So was the .089db difference between the 4x10 & 1x15, or between using the 1x15 with & without the z-matcher?

[hasserl]

there clearly is a fairly significant loss via the Z-Matcher.

I don't think this is so clear. The difference in db could be a result in different speaker efficiency. You really need to measure power output to get an accurate assessment of any power loss.

[John Phillips]

I don't think this is so clear. The difference in db could be a result in different speaker efficiency. You really need to measure power output to get an accurate assessment of any power loss.

He did :).

...I measured just under 1dB power loss with 50W input at 1kHz (from Super Reverb chassis)...

[cerebralpaul]

Ah I missed that... I didn't know power was measured in db? But either way, was this the difference in speaker configurations, or the difference between z-matcher & not?

[John Phillips]

Ah I missed that... I didn't know power was measured in db? But either way, was this the difference in speaker configurations, or the difference between z-matcher & not?The difference between the input power and output power going through the Z-Matcher.

(Or at least, that's what I understood/would have done myself if I had one to test. OTm please correct me if not!)

Power is not measured in dB. Technically, neither is volume. Decibels (1/10th Bels, named after Alexander Graham Bell, inventor of the telephone and hence indirectly all electronic signal transfer) are a measure of the relative levels of a signal - any signal. They have become associated with volume because a sound can be said to be a certain number of decibels above a base level (0dB) representing the threshold of human hearing (there is a precise definition of that, but I can't remember the exact figures). More confusingly, this has nothing to do with the 0dB defined for Line Level signals!

Power gain or loss can be expressed as a decibel ratio of input to output power. If you divide the output power by the input power, then take the Log(10) of the result, that's the number in Bels. Multiply by 10 to get the number in decibels. You will find that a halving of power is equivalent to -3.01dB (the negative is because the ratio is smaller than one, being a loss). What's useful is that you can also make calculations without measuring the power directly, because you can also measure the voltage ratio (RMS or peak-to-peak) and work out the power ratio from the impedance, which is (I assume) what OTm did here.

If the result was -0.89dB, you can work back from that and find that the power output was equivalent to 81.5% of the power input, which means that the loss due to the Z-Matcher was 18.5%. This is completely acceptable when going from 2 ohms (amp) to 8 ohms (speaker), because the loss from mismatching 4:1 would be much worse than that, even discounting the risk to the amp from the over-high mismatch.

But going from 2 ohms to 4 ohms, it's less clear. Assuming that the Z-Matcher's efficiency would be the same - no particular reason to think that it wouldn't be, as it's a simple auto-transformer as far as I know - the loss is about the same as that from a normal 2:1 mismatch. It's a little less than the difference in maximum power, and a little more than the difference in equivalent clean signal power, usually are (roughly 30% and 10% respectively). Whether you could easily even tell the difference between those three levels, I'm not sure. There might also be a difference in tone both due to the mismatch and the Z-Matcher, but as you've realised you'd have to use the same speakers to find out which was greater... actually you could, with a Super, because you could wire the 4x10" for 2 ohms or 8 ohms.

[/Practical Physics For Musicians, Chapter 8,549 ;).]

[hasserl]

He did :).

Sorry, my mistake. I read about different speakers and db's and I immediately thought SPL, forgetting that power is also sometimes referred to as db, which confuses weak minded individuals like me. ;)

[TubeBy12]

...to side-step that problem, I just use one of Ted Webers (bare) Z-Matcher transformers between the 2-Ohm OT of my SR and the 8-Ohm JBL/D130F in my VibroClone cabinet...works great.

I use one between my 16-ohm-only Classic 30 head and 8-ohm Recto cab, but I have no basis for comparison to running direct at this point (at least until I can get a second 8-ohm cab to run serially, that is...).

[Old Tele man]

(Or at least, that's what I understood/would have done myself if I had one to test. OTm please correct me if not!)...correct.

... actually you could, with a Super, because you could wire the 4x10" for 2 ohms or 8 ohms....for these tests I used non-inductive 2-ohm and 8-ohm resistors, respectively, so what's coming out of the speakers could be different; however, "...by the numbers..." the resistors provide an "...apples to apples" comparison testing of the Z-Matcher itself...which was my intention.