• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Speaker mismatch and how it effects the output transformer

Hi everyone, this is from a discussion I had with a friend the other day. I was hoping you could clear something out for me.
So basically what had happened was he had 8 ohm selected from the selector switch of his amp and he connected a 16 ohm speaker. Hearing that something was not right he checked the switch and corrected it. Anyway he was concerned that it might have had a detrimental effect on the output transformer.
My idea was that from what he has done no harm would be caused for the output transformer, but if it was the other way round 16 ohm selected and connected an 8 ohm speaker the output transformer would have been toasted. Then we consulted one of the best tube techs we know and he did mention that from both situations the output transformer could have been fried and that my friend was just lucky that he didn't have his amp fried.
I'll quote what he said "that way around an ohm mismatch is not healthy because tubes hate to be idle.. so it could indeed blow the OT.. but for this amp it does not seem likely because the components and the sheer amount of power available".
Could you please explain why this way round could be a problem for the output transformer.
Thanks in advance.
 
No, no harm done. Even if the amp was UNLOADED, it wouldn't be harmed unless there was an input signal large enough to drive it into clipping. In that case, the voltage developed by the extremely fast current changes will be limited by whatever arcs over first (often the output transformer, sometimes a tube socket). May not catch fire, but will smoke and smell.

With 16 Ohm speaker on 8 Ohm tap, the amp won't deliver full power (perhaps half), but won't be harmed. Reverse that combination and distortion will increase, also less power delivered, but again, no permanent harm.
 
Have your friend get the impedance versus frequency specs for his loudspeaker.
That may be hard to find. But if he finds the impedance versus frequency curve in a magazine review, he will be surprised.

One measurement is incomplete, but easy: take an ohmmeter and measure the DC resistance of the loudspeaker. At least at really low bass frequencies that is its impedance.
I bet it is lower than 16 Ohms.

Loudspeakers are like that. The minimum impedance at some frequencies ismuch lower than the manufacturer's spec. And the impedance is much higher at some frequencies than the manufacturer's spec.

What loudspeaker model?
What amplifier model?
Or if it is a DIY, is it SE, Push Pull, negative feedback or no, output tubes, etc.?
 
Last edited:
Using a DPDT to switch OPT secondaries

Hope I am not hijacking this thread in any way, but I have a query that is in this area, and is not worthy of a thread to itself.

I have recently created a more permanent case for a Tubelab SSE ...
SSE: The saga continues
...
Is it OK to use a DPDT switch to toggle between 4ohm and 8ohm taps on the OPTs? This is instead of having sockets for the loudspeaker for ground, 4ohm and 8ohm.
 
P.S. A stop pin is included to make the 3-way switch into a 2-way switch.

My thinking is that a toggle switch is too easily flipped by accident to the wrong position.
 

Attachments

  • Stop Pin.jpg
    Stop Pin.jpg
    56 KB · Views: 485
Thanks for the insight guys really helpful. My friend didn't mention which amp it was he just said that it was a 100W tube amp. He's got many...
I made a mistake in the first post 4 ohms was selected not 8 ohms. So he had 4 ohms selected from the selector switch and connected a 16 ohm load - would that have been a problem. Was the tube tech right?
I am under the impression that a 2:1 mismatch wouldn't cause much of a disaster, but selecting 4 ohms tap and connecting a 16 ohm load or the other way round would this be a disastrous situation?
Thanks again :)
 
Last edited:
we consulted one of the best tube techs we know and he did mention that from both situations the output transformer could have been fried and that my friend was just lucky that he didn't have his amp fried..

Sounds like another "guru" that has not the slightest clue.
It's quite common for ESL loudspeakers to present an amplifier with a near short circuit at HF, meanwhile the impedance of a typical woofer usually hits about 35-60 ohms anything close to resonance, then we have the influence of crossovers which are designed to roll off one driver while another takes over.

I simply can't understand how people who are supposed to know what they are talking about, end up reciting crap, maybe he was seeing $$$$$ signs? :eek:
 
The danger point is connecting too high an impedance, but 16 ohm speaker on 4 ohm tap is tolerable - it's just not correct for proper power matching. Similarly for 4 ohm speaker on 16 ohm tap.

An open circuit (no speaker load) is an infinite impedance and a valve amp should not be played (signal continuously applied) under these circumstances as damage to the output transformer or output valves could result.
 
Thanks everyone for clearing up my doubt. So just to summarize connecting a different rated speaker/load to the corresponding tap would not be disastrous. Even if it was 16 ohms load connected to a 4 ohms tap or 4 ohms load to a 16 ohm tap. It would just be wrong but wouldn't necessarily blow something up.
I would've thought connecting a 4 ohms speaker to a 16 ohm tap would blow up the output transformer, but it made sense when the point about the speaker would have different impedance values at different frequencies was raised.
I wonder what the tube tech was going on about :confused:
 
Last edited:
It's all about ratios and their squares.

If the amplifier was designed for a 10k load (4 ohm on 4 ohm tap) and you put a 4 ohm load on the 16 ohm tap, your load becomes 2k5. If you put a 16 ohm load on the 4 ohm tap, your load becomes 40k.

On the datasheet for many tubes, you'll see a graph that gives you distortion vs power vs load etc. It is indeed a range.

For practical purposes, the only modes that can damage an amplifier are an open circuit or a short circuit. And even this is supposed to only be a problem if the amplifier is amplifying. With no input signal, neither condition should cause damage.
 
You can read more about impedance matching if you scroll down to the Tube-amp section in this link: The do's and don'ts of guitar amp impedance

You should avoid an impedance mismatch if at all possible in order to ensure maximum power transfer and to avoid stress to the amplifier

Short term mismatches such as connecting 16 or 4 ohm speakers to an 8 ohm tap are tolerable - they will not bring about instant destruction of an amp. A more extreme mismatch such as a 16 ohm speaker on a 4 ohm tap is best avoided.

If you need to make a mismatch it's better to have the speaker impedance lower than the amp's impedance.

EDIT: The link seems to default quickly to the home page - just click on blog to find it again!
 
Last edited:
If you need to make a mismatch it's better to have the speaker impedance lower than the amp's impedance.

Why is that?
--
f the amplifier was designed for a 10k load (4 ohm on 4 ohm tap) and you put a 4 ohm load on the 16 ohm tap, your load becomes 2k5. If you put a 16 ohm load on the 4 ohm tap, your load becomes 40k.

That's my understanding.

Let's take an OPT with 5k ohm primary and 4 ohm secondary.
That's a turns ratio (primary:secondary) of about 35.4:1
That turns ratio never changes. There are always 35.4 turns on the primary for every one turn on the secondary.

If we double the load on the secondary to 8 ohms, the reflected impedance of the secondary back to the primary stays in the same ratio. So doubling the secondary impedance will double the primary impedance. That 5k:4 transformer will now be working as 10k:8.

When you connect a triode to the primary, the max power will be transferred to the load when the primary impedance is 2X the triode's rp. Let's say that triode has internal plate resistance (rp) of 2.5k ohms. In that case, a 5k ohm primary impedance will get you the max power out from that triode. If the load is 4 ohms, then you choose an OPT of 5k:4 ohms impedance (35.4:1 turns ratio).

When you increase the load to 8 ohms and connect that to OPT's 4 ohm secondary tap, the primary impedance increases to 10k ohms. Now the Zpri is 4X the triode's rp. That will operate the tube with more damping (a higher impedance load), yielding lower distortion, but also lower power output into that 8 ohm load.

If the output tube being loaded by the OPT primary is a pentode, then things are more complicated. But you can see from the above that there is a *range* of impedance loading choices with any output tube and OPT. It's not like there's one 'magic' primary impedance that will yield perfect results, and deviating from that will cause catastrophic failure.

This graphic shows how power output and distortion vary with primary impedance in a Push-Pull class AB1 pentode amplifier.

attachment.php


You can see that there's a sweet spot where you get the most power output with a 6k primary impedance. But the lowest distortion happens if a 7.6k primary impedance is chosen, but as you can see, you won't get as much power output. You get to choose your compromise between max power out vs. lowest distortion.

I hope I explained that reasonably well. Trying to explain these things is a really good test of one's understanding...
--
 
Last edited:
Minor misloading of a tube amp should not hurt it. 16 ohms on the 4 ohm tap of a HiFi amp that is operated in a normal HiFi environment should also be fine.

16 ohms on the 4 ohm tap of a guitar amp that is driven deep into clipping CAN cause damage if the right conditions are present. I have seen a tube socket arc in a Fender Bandmaster.

As mentioned previously a speaker is only at is nominal impedance in the sales literature. Even in the published specs my "8 ohm" Yamaha NS10M's vary from about 6 ohms to over 30 ohms. I have seen them go much higher when hit hard with bass.

In the real world it's impedance varies with frequency, volume level, and the dynamic of the music being played. It can dip to near zero when a big bass drum transient tries to instantaneously reverse the woofer cone's travel. It can reach 5 to 10 times the rated impedance when the woofer is driven at resonance with medium volume. A guitar speaker often has its resonant frequency within the guitar's normal frequency range.

It is possible for an 16 ohm guitar speaker to be 80+ ohms at resonance. Connecting it to the 4 ohm tap and driving the amp into hard clipping can produce voltages over 2 KV on the output tubes even though the amplifier is only running 400 volts of B+.....these are real measured voltages on one of my guitar amps.

If the amp still works as it did before, it is fine.

Is it OK to use a DPDT switch to toggle between 4ohm and 8ohm taps on the OPTs?

This is fine as long as the switch can handle the amplifier's full audio power, a couple Amps in the case of the SSE. The switch must NEVER be operated while music is playing since it will be open circuit for a brief instant. A "make before break" switch can be used in this case to eliminate the "never operated live" requirement, but these are harder to find.
 
Hi rongon! Re 'Why is that?

I was working on the assumption that a 16 ohm speaker connected to a 4 ohm tap could produce a reflected impedance high enough to pose a problem similar to that of working into an open circuit (infinite impedance).

I now see that Tubelab says that 'flashback' is, indeed, possible under these circumstances.
 
Speaker impedance ratings are somewhat confusing. A 8 ohm speaker for example is a nominal reading in that it measures 8 ohms in a region of the audio spectrum. An 8 ohm speaker is liable to dip in low frequencies and raise in high frequencies if my memory is correct. Therefore when listening to a soundtrack the impedance is going to be all over the place so why get worried about it. Simply select the impedance that most closely matches your particular speaker.
 
No harm should be done with a mismatch in either direction. You will lose about half the power and distortion will rise.
I also think the OP needs to find a new tube tech.

You don't need a degree in electronics to fix valve amps.
In the mid 1900's there were "valve technicians" that went around swapping out valves until faults were fixed. Both audio amps and TV's.
I fixed our black and white TV in 1965 at 12 year old by just swapping valves from a duff set with the faulty one. I just looked at the number on the valves to find the right one.
 
  • Like
Reactions: 1 user
Minor misloading of a tube amp should not hurt it.
This is fine as long as the switch can handle the amplifier's full audio power, a couple Amps in the case of the SSE. The switch must NEVER be operated while music is playing since it will be open circuit for a brief instant.
A dangerous thing to do to a tube amp is disconnect the speaker while operating it. "No load" can push the input turns of the transformer over the 600 v the wire insulation is rated at. Easy to do: for example tripping over the speaker wire & yanking it out of the phone jack or screw terminal. I've done it to my ST70. Or switching an ordinary speaker impedance switch while music is being generated.
To prevent future damage I've put a 470 ohm 3 watt (1000 v rated) resistor across the back of my speaker terminal boards, inside the chassis. In this modern day, a 300 v rated TVS diode or metal oxide transient supressor in that position could be even less audible.
BTW I got away with it, no damage and the ST70 still works. A classic amp for a reason, durability + decent sound.
As an aside, I started checking tubes (valves in britspeak) of our fuzzy TV at the convenience store tester at age 8. Worked, TV back to original fuzzovision (512 lines on sets better than our Philco), no serviceman required. 1080 line TV is wonderful; I quit watching fuzzovision about 1965.
 
Last edited: