Jump to content
Hamer Fan Club Message Center
  • 0

Tube amp impedance mis-matches??


tomteriffic

Question

So somebody school me a little on the subject.

When I was a kid, I'd just jam as many speakers/cables into the back of a given amp as I could, given the opportunity. Never had much of a problem when doing that.  When I got a little older, I paid much more attention to the subject where hi-fi/stereo was concerned, particularly since a mismatch to the low side could put the amp into thermal runaway and let all of its smoke out.

I read someplace in the past couple of years that tube amps really don't like mismatches to the high side but can tolerate a mismatch to the low side (within reason) fairly well.  It was on teh intarwebz so it was probably true but I thought I'd ask.  I have an impending mismatch that might be unavoidable and, if so, which way to I want to go?

Link to comment
Share on other sites

19 answers to this question

Recommended Posts

12 minutes ago, SSII x 2 said:

It's safe to use a 16 Ω cab with a 8 Ω output from the amp, but not the reverse!

Not true in most cases.

Tom, download just about any Mesa amp manual and it'll have all the info you'd ever want on speaker/ohm mixing and matching. Decent overview from H&K here: http://blog.hughes-and-kettner.com/ohm-cooking-101-understanding-amps-speakers-and-impedance/

 

Link to comment
Share on other sites

Running a lower impedance cab than the tap you are plugged into is going to pull more current through the output transformer and generate more heat.  This concerns me more than the higher voltage that results in the scenario where the cab impedance is higher than the output tap impedance.  H&K may design their output transformers to handle 1/2 (or 1/4, or...) the rated impedance on a given output tap but I wouldn't do it with just any amp.

I personally err on the side of using a speaker 2x the impedance of the amp output rather than the opposite.   Or just used the correct impedance and not worry about it.

Link to comment
Share on other sites

Just as long as the cabinet impedance is equal to or greater than the output impedance of the amp, you'll be OK.

For an 8-ohm amp, an 8 or 16 ohm cabinet is OK, but don't use a 4 ohm cabinet.

Link to comment
Share on other sites

I've always been well aware of the higher = safer/lower = iffy rule of thumb.  And I've also been aware of the "never, ever let an EL-84 amp see less than 8 ohms unless it has specifically marked taps for less" thing.  Two amp designers told me the last one, pretty emphatically.

In this particular situation, a mismatch is inevitable and the less egregious one would be to the low side (5.33 ohms or 24 ohms referenced to 8 ohms, the only tap available) and the high side might actually be worse.  Sounds to me like it'd be fairly OK.  And I wouldn't have to re-wire the extension cab to work with a number of other amps.

Link to comment
Share on other sites

12 hours ago, Armitage said:

The rule of thumb is it's generally ok to use speakers of higher impedance with a transistor amp... and lower impedance with a tube amp. Mainly because of the output transformer.

Nope. The exact opposite. The impedance match/mismatch is rooted in the amp's output impedance facing the speaker's input impedance. Whether a 4-ohm load is safe with a transistor amp depends on the particular transistor amp's design and especially the size of its power supply (output transformer plus filter capacitors), but generally speaking they're going to be more low-impedance friendly than a tube amp.

Tube amps tend to have higher output impedances; SS amps are easy to design with low output impedance.

The reason tube amps have output transformers is to lower the output impedance to provide a good match with a lower impedance speaker load. That's why better tube amps have dual or multiple "taps" to provide both 4-ohm and 8-ohm outputs, and older speaker cabinets were 16-ohm, such as a pair of Altec 9845As I used to have. Their design dates back to the tube era, so their default impedance is 16 ohms, meaning they'd be compatible with even the wimpiest of output transformers.

$_1.JPG

Link to comment
Share on other sites

On 8/11/2017 at 11:16 AM, Armitage said:

The rule of thumb is it's generally ok to use speakers of higher impedance with a transistor amp... and lower impedance with a tube amp. Mainly because of the output transformer.

David Allen's tube amps I have built have the output jack shorted to be safer if nothing is plugged into it.  It would rather see a zero load than see an infinite impedance load.  (Not advisable to run that way long term...)  He's a retired power engineer with patents, so he should know something.  I also worked for another boutique guy wiring up amps, and all of his had shorted output jacks.  So, you are right.

On the flip side, I have a Boss Katana amp now (solid state) that says specifically a MINIMUM 8 ohm load.  So, you are right again.

Yay!

Link to comment
Share on other sites

On 8/12/2017 at 4:46 PM, fractal said:

David Allen's tube amps I have built have the output jack shorted to be safer if nothing is plugged into it.  It would rather see a zero load than see an infinite impedance load.  (Not advisable to run that way long term...)  He's a retired power engineer with patents, so he should know something.  I also worked for another boutique guy wiring up amps, and all of his had shorted output jacks.  So, you are right.

On the flip side, I have a Boss Katana amp now (solid state) that says specifically a MINIMUM 8 ohm load.  So, you are right again.

Yay!

Power tubes have much higher inherent output impedances than output transistors. As stated in this white paper by a designer of electrostatic speakers and amplifiers to drive them:

Quote

Tubes are inherently high impedance devices.  A large power tube like a 6550 or KT-88 has an output impedance of around 2,000 ohms.  By comparison, a large power transistor has an output impedance of less than one ohm.

Tubes cannot drive loudspeakers directly due to their high impedance.  To correct this problem, output transformers are used in most tube amps.  These transformers have a specific turns ratios that will convert the tube's impedance from several thousand ohms to typically 4, 8, or 16 ohms.

The great majority of tube amps have fairly large and heavy output transformers to lower the output impedance to match typical speakers. If it contains taps for 8 and 4 ohm loads, the transformer will be larger. If it's a 16-ohm-only unit like the one in my Trace Elliott Velocette, the transformer will be relatively small. 

To illustrate, here is the latest edition of the Macintosh MC275 stereo amplifier, originally produced in 1961 but still in production with updated features: 

?size=648,460

You can see that it has three large transformers. One is for the power supply to provide current to keep the voltage constant. The other two are the output transformers for the stereo amp's two channels. On the end panel facing you, you can see a large row of gold-plated speaker terminals, because they are wired into various points (or "taps") in the transformers to provide optimum output impedances for 4, 8, or 16-ohm rated speakers. This amp is rated at 75 watts per channel, weighs 67 lbs., much of it from the output transformers. It costs $5500.

By contrast, here is the Parasound 275 v2 power amp:

275v2_high_angle_black.jpg

It's rated at 90, 150, and 120 watts/channel into 8, 4, and 2 ohms, respectively. It weighs 21 lbs. and costs $595. 

With its slim profile, it obviously doesn't need output transformers to convert transistors' output impedance to match speaker loads. It can remain stable into those very low and fluctuating speaker loads (i.e., 4- and 2-ohm loads) via a high current power supply (with large filter capacitors) and an array of heat sinks attached to the output transistors to draw off the heat generated into lower impedances, and then vent the heat through the very perforated top.

The inherent output impedance of tube amps requires expensive and herculean design efforts to provide output impedances that are friendly to 8 and 4 ohms. Solid state amps only need high current capacity and a heat dissipation system. I've even had a MOSFET solid state amp that was stable into ZERO ohms.

 

Link to comment
Share on other sites

On 8/13/2017 at 0:32 PM, JohnnyB said:

Power tubes have much higher inherent output impedances than output transistors. As stated in this white paper by a designer of electrostatic speakers and amplifiers to drive them:

The great majority of tube amps have fairly large and heavy output transformers to lower the output impedance to match typical speakers. If it contains taps for 8 and 4 ohm loads, the transformer will be larger. If it's a 16-ohm-only unit like the one in my Trace Elliott Velocette, the transformer will be relatively small. 

To illustrate, here is the latest edition of the Macintosh MC275 stereo amplifier, originally produced in 1961 but still in production with updated features: 

?size=648,460

You can see that it has three large transformers. One is for the power supply to provide current to keep the voltage constant. The other two are the output transformers for the stereo amp's two channels. On the end panel facing you, you can see a large row of gold-plated speaker terminals, because they are wired into various points (or "taps") in the transformers to provide optimum output impedances for 4, 8, or 16-ohm rated speakers. This amp is rated at 75 watts per channel, weighs 67 lbs., much of it from the output transformers. It costs $5500.

By contrast, here is the Parasound 275 v2 power amp:

275v2_high_angle_black.jpg

It's rated at 90, 150, and 120 watts/channel into 8, 4, and 2 ohms, respectively. It weighs 21 lbs. and costs $595. 

With its slim profile, it obviously doesn't need output transformers to convert transistors' output impedance to match speaker loads. It can remain stable into those very low and fluctuating speaker loads (i.e., 4- and 2-ohm loads) via a high current power supply (with large filter capacitors) and an array of heat sinks attached to the output transistors to draw off the heat generated into lower impedances, and then vent the heat through the very perforated top.

The inherent output impedance of tube amps requires expensive and herculean design efforts to provide output impedances that are friendly to 8 and 4 ohms. Solid state amps only need high current capacity and a heat dissipation system. I've even had a MOSFET solid state amp that was stable into ZERO ohms.

 

Tomterrific's original question not addressing the design of the amp (or the cost?), but was referring to a mismatch of impedance from the intended design.  I just did a quick search to make sure I wasn't crazy, and every single valid reference said that it is better (not ideal) to go with a lower speaker impedance with a tube amp (than stated at the output), and higher speaker impedance with a solid-state amp.

So, I'll stick with my original statements.  Actually, I try to stick with as much of a non-mismatch as possible.  YMMV.

Enjoy your day everyone!

Link to comment
Share on other sites

On 8/15/2017 at 8:37 PM, fractal said:

Tomterrific's original question not addressing the design of the amp (or the cost?), but was referring to a mismatch of impedance from the intended design.  

I went into the design of the amp and the cost to illustrate that the inherent high output impedance of output tube requires big, expensive output transformers to lower the tubes' output impedance to levels that are compatible with typical speaker impedances ranging from 4 to 16 ohms.

I just did a quick search to make sure I wasn't crazy, and every single valid reference said that it is better (not ideal) to go with a lower speaker impedance with a tube amp (than stated at the output), and higher speaker impedance with a solid-state amp.

I don't know where you got your information, but it just doesn't make sense, when output tubes typically have output impedances around 2K ohms. The output impedance of an amplifier has to be quite a bit lower than the impedance of the speaker it drives to work well. I contrasted the size and costs of tube vs. solid state amps to illustrate just how much more elaborate and expensive it is to lower the output impedance of a tube amp to be compatible with the load impedance of a typical speaker. Where a tube amp requires big, heavy output tranformers to lower the impedance to an acceptable level, a solid state amp does it naturally and economically.

In the '50s and early '60s of home audio and pro audio where the electronics were predominantly tube-based, loudspeakers frequently had 16-ohm ratings because they were an easier load for tube amps. The higher speaker impedance didn't require as large of an output transformer.

 

Link to comment
Share on other sites

It seems like we are talking about two different things and possibly speaking on different frequencies (pun intended - that's as good as I get!)

If I have a tube amp that has a designed output impedance of 8 ohms (not the tube's output, but the speaker out jack on the back of the amp).  The ideal speaker load is 8 ohms.  Yes, there is a large transformer to couple the output of the tubes to the speaker connection.  But that's beside the point.  The amp's output on the back was designed to see an 8 ohm impedance speaker.  The original question was whether it would be better to connect a load of less than or greater than 8 ohms.  I built many amps myself (not designed) and worked for a few years as an assembler for an amp maker (who shared lots of information - I learned a lot).  We would always install a shorting jack for the speaker out because it was better than providing an infinite load if a speaker wasn't connected.  It was still bad not to connect a speaker, but it was better if it was shorted than open.  So, for tubes, I'd rather provide a lower speaker load than stated than a higher - knowing that the load intended for the transformer (however large) would be the optimum load that would work best to couple with the tube's output impedance.

I have a Boss Katana amp (fun, relatively cheap, and sounds pretty good) that is solid state.  The Boss states a minimum of 8 ohms in the manual - and I'm assuming that they don't want you to go lower for a reason.  So, it appears that it is the opposite for solid state.

I hate to quote forums, but here is just one page where every single answer supports what I just said: https://www.quora.com/Can-you-use-8-ohm-speakers-on-a-4-ohm-stereo-system

Thanks for making me think!

Have a great day everyone!

Link to comment
Share on other sites

I've seen this question posted many times before and even Googled it myself, and the answers always spark a debate that leaves me more confused than I was before.  No exception here.  Everyone speaks so authoritatively and cites various experts with links to prove their point, and I have next to no knowledge on the subject so I'm left scratching my head.  :wacko::blink::huh:

I tend to believe Johnny B. on these type of things though, even if I don't understand his explanation.

Link to comment
Share on other sites

I have absolutely no idea as I find myself in the same state as Tommy P (and for the same reasons).

Mesa Boogie comes down on the side of greater-than-intended = safe (see PDF here).

Hughes and Kettener state in their blog either is acceptable (within a 4:1 limit), while lower-than-intended is the safer of the two scenarios.

Little Dawg (as well as some others) mention the half-century-long mismatch of Fender amps (nearly ubiquitous extension speaker jacks) as being safe, but warns not to do this with other manufacturers' fare.

Gollihur Music (no idea, they just have a very effective Google configuration that puts their content at the top of every search) sides with Mesa.

The Marshall cabinet handbooks instruct to match the impedances.  Nothing about mismatches.  The same in the DSL handbook PDF, as well. 

In several places "quotes" attributed to Mike Soldano proffer the opposite opinion, though search as I may, I was never able to find a verifiable direct quote.   

If it's any help (or maybe more confusing), the old tweed Bandmaster ran three 8Ω speakers in parallel for total speaker load of  1/((1/8Ω)+(1/8Ω)+(1/8Ω)) = 2.67Ω.  The original output transformer was 4Ω.  Only one example, but that would seem to indicate at least Leo considered running into a lower impedance was safe.

:wacko:

Link to comment
Share on other sites

5 hours ago, tommy p said:

I've seen this question posted many times before and even Googled it myself, and the answers always spark a debate that leaves me more confused than I was before. 

 

I think some of that confusion comes down to the terms. A higher load is a lower impedance, and a lighter load is a higher impedance. Start tossing around "Higher is better for x" or "Lower is better for y" then someone could get confused and it's downhill from there.

Link to comment
Share on other sites

Thinking back, I can't even count how many perfectly good Super Reverbs (2 Ohm speaker load) I saw wrecked when somebody put a single 15 (8 ohm) or pair of 12's wired as 16 ohm in them.  I'm thinking that, for my purposes, a 5.33 ohm load across an 8 ohm output is probably pretty mild, especially since I wouldn't be working the amp *that* hard for hours on end.  I used to do that with a 2.66 ohm load into 8 ohms all night, damn near dimed, all the time.  In all the above scenarios, we're talking blackface Fenders.

Link to comment
Share on other sites

Just trying to work through it... Ohm's law: one amp = one volt.  The voltage drop across a resistance is directly proportional to the size of the resistance.  

∴ it seems logical that as the impedance load increases (wattage dissipated by the tubes through the output transformer remaining the same), voltage would drop causing current to increase, at the extreme, burning wires (or OT coils) not intended to handle the many coulombs (sorry, haven't gotten to use that word in ages and I don't think I've ever had occasion to type it - and likely isn't even appropriate to this discussion as we're talking AC current and not dissipation of a charge, but I wanted to use it, so there :lol:) suddenly passing through. Taking this logic to the extreme would present the scenario of an infinite load and we all know it is strictly verboten to run a tube amp without a speaker attached.

Likewise ∴ it seems logical that as the impedance load decreases (again, wattage dissipated by the tubes through the output transformer remaining the same), voltage would increase causing current to drop.  Taking this logic to the extreme would present the scenario of no load (i.e., shorting across the termination) which we have read even in this thread is often done by more conscientious builders as a safety precaution. 

I will ponder these conclusions and wait for correction from those more learned than myself.  :)

Link to comment
Share on other sites

Once you add coils and transformers to a circuit things get complex fast.

Tube amps blow up their output transformer with no load. The amp is running full blast on only one side of the transformer and it overheats because there is no where for the output to go.

Transistor amps don't have an output  transformer and they can run forever with no load i.e. infinite resistance.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...