Output Transformerless vs Transformer Coupled Output Tube Amps
Jan 17, 2014 at 5:32 AM Thread Starter Post #1 of 37

PTom

100+ Head-Fier
Joined
Oct 20, 2013
Posts
415
Likes
24
I'm quite confused regarding the effect of output transformerless and transformer coupled tube amps on the sound quality. Maybe someone can explain a few points for me:
 
1) For driving high impedance headphones, I've heard people mention that output transformerless tube amps are better than transformer coupled ones. Is this because the addition of a transformer to the circuit adds to degradation of the sound quality?
 
2) Doesn't a high impedance amplifier result in a lower damping factor? So is it possible to use a transformer coupled tube amp at low impedance settings to drive high impedance headphones (will the voltage be sufficient?). This would then result in a higher damping factor and therefore superior SQ compared to OTLs.
 
3) Or is the whole point of tube amps to decrease the damping factor via high impedance? If this is the case won't a solid state amp with high output impedance give the same effect?
 
Jan 17, 2014 at 11:35 AM Post #2 of 37
Doesn't a higher output impedance result in a lower damping factor? 
 
Damping factor = Load impedance / Output impedance
 
Regardless of that, I am also curious what kind of audible differences the two options have, or why to use tubes at all, and how any of this is beneficial in terms of hi-fi.
 
Jan 17, 2014 at 11:46 AM Post #3 of 37
Yeah I think you're right, higher headphone impedance and lower amplifier impedances result in higher damping factors. So we want a high damping factor. I'll change the opening post accordingly.
 
Jan 18, 2014 at 2:49 PM Post #5 of 37
  I'm quite confused regarding the effect of output transformerless and transformer coupled tube amps on the sound quality. Maybe someone can explain a few points for me:
 
1) For driving high impedance headphones, I've heard people mention that output transformerless tube amps are better than transformer coupled ones. Is this because the addition of a transformer to the circuit adds to degradation of the sound quality?
 
2) Doesn't a high impedance amplifier result in a lower damping factor? So is it possible to use a transformer coupled tube amp at low impedance settings to drive high impedance headphones (will the voltage be sufficient?). This would then result in a higher damping factor and therefore superior SQ compared to OTLs.
 
3) Or is the whole point of tube amps to decrease the damping factor via high impedance? If this is the case won't a solid state amp with high output impedance give the same effect?


1) The only reasoning I can see for this statement is that OTLs tend to put more power into higher impedance loads. It's usually the other way around for transformer-coupled amps. But either way, if an amp has enough power into the headphone's load, that's the important part. 

2) I guess so, but the general guideline is that damping factor should be at least 8. I'm not sure how important it is to go higher than that. So you might get enough damping factor even with a high output impedance amp, if you're using 600 ohm headphones for example. Whether or not voltage is sufficient depends on the particular amp's power specs. 

3) IMO the point of tube amps is that some people like tubes for a variety of reasons. I don't think it reduces entirely to damping factor. 
 
Jan 18, 2014 at 4:45 PM Post #6 of 37
2) I guess so, but the general guideline is that damping factor should be at least 8. I'm not sure how important it is to go higher than that. So you might get enough damping factor even with a high output impedance amp, if you're using 600 ohm headphones for example. Whether or not voltage is sufficient depends on the particular amp's power specs. 

 
I'm reading that, for example, the Bottlehead Crack OTL has an output impedance of around 120 ohm (http://www.innerfidelity.com/content/marvelously-addictive-bottlehead-crack) which is way higher than 1/8th of the impedance of the Sennheiser HD600/650s they're normally paired with.
 
3) IMO the point of tube amps is that some people like tubes for a variety of reasons. I don't think it reduces entirely to damping factor. 

 
This head-fi article on impedance (http://www.head-fi.org/a/headphone-impedance)  says that tube amplifiers result in 
poorer damping factors and therefore "fuller bass":
 
"Some people prefer the sound of vacuum tube headphone amplifiers which usually have higher output impedance relative to solid state (transistor)  amplifiers, and therefore have poorer damping factor and hence, fuller bass"
 
However I don't think this is the full story. I've heard that different tubes "colour" the sound in different ways so it appears to be a good way to change the sound signature without having to change your whole setup.
 
 
1) The only reasoning I can see for this statement is that OTLs tend to put more power into higher impedance loads. It's usually the other way around for transformer-coupled amps. But either way, if an amp has enough power into the headphone's load, that's the important part. 

I thought transformers in general (not sure specifically about those used in amps) had really high efficiencies (i.e. 90%+) so assuming the high impedance and low impedance settings of a transformer coupled tube amp only differs in respect to the number of turns of either the primary or secondary coils, I would have thought the power output would be similar for high and low impedance loads.
 
 Whether or not voltage is sufficient depends on the particular amp's power specs. 
 

I assumed that the low impedance setting of output transformer tube amps corresponds to lower number of turns of the secondary coils of the transformer which would mean that the output is high current and low voltage. If this is the case, I'm wondering if the low impedance settings are suitable to drive high impedance headphones which require high voltage and low current.
 
Jan 18, 2014 at 5:37 PM Post #8 of 37
I don't have the expertise to explain why OTLs put more power into higher loads, I'm afraid. My comment 1) above was based on observation of manufacturers' reported power specs into different loads. For example, consider the power specs of the little dot mk3 OTL vs the transformer coupled woo wa6.

I suppose the HD-6x0/ crack pairing just happens to have good "synergy," but some OTLs have much lower output impedance than that. The little dot mk3 is somewhere under 10 ohms, depending on the gain setting. At least that's what the owner of the company told me.

Tubes often add harmonic distortion, which colors the sound in a way that's pleasing for some. I think you are right that low damping factor can also be part of the tube appeal, but not the full story.
 
Jan 18, 2014 at 5:45 PM Post #9 of 37
  So I think my main question from the above is:
 
Why do OTLs tend to put more power into high impedance loads?


I have more knowledge of OTL's with speakers, but the principles are likely the same.  OTL amps generally have limited current output.  They usually have plenty of voltage.  So with a high impedance, the higher voltage is there and current draw isn't more than the OTL can deliver.  At lower impedance the current goes up, possibly more than the OTL can deliver, so even though it has the voltage it starves for current and runs out of juice. 
 
The other effect is the high output impedance of the OTL.  The output impedance of the OTL will mean it gets more of its power to the high impedance headphone and less to a low impedance headphone.
 
You are right about output transformers.  If done well, they more or less will have the same power output with any of the taps, assuming those taps match the load pretty well. 
 
Different tubes do have some different colorations to sound.  But then so does tastefully done digital EQ.  That EQ is lots easier to live with.  You could go very far toward getting tube sound by adding a series resistance to the output of a good solid state headphone amp, and doing some EQ.  Just won't look so cool as those glowing tubes. 
 
Jan 18, 2014 at 6:24 PM Post #10 of 37
Tubes often add harmonic distortion, which colors the sound in a way that's pleasing for some. I think you are right that low damping factor can also be part of the tube appeal, but not the full story.

 
This is a very interesting point. Here's a link to an earlier thread:
 
http://www.head-fi.org/t/595113/differences-between-tube-amps-and-solid-state-amps-how-do-i-know-which-one-to-get#post_8134487
 
In the post linked, it's suggested that so called Total Harmonic Distortions interact to create a fuller sound by compensating for recording losses.
 
I think the general argument is that SS amps of low output impedance are good at faithfully reproducing what was recorded, but what was recorded might not be exactly what the music sounds like.
 
Jan 18, 2014 at 6:39 PM Post #11 of 37

 
I have more knowledge of OTL's with speakers, but the principles are likely the same.  OTL amps generally have limited current output.  They usually have plenty of voltage.  So with a high impedance, the higher voltage is there and current draw isn't more than the OTL can deliver.  At lower impedance the current goes up, possibly more than the OTL can deliver, so even though it has the voltage it starves for current and runs out of juice. 
 

 
Assuming we can represent the amplifier output as a source of a certain voltage (can we?). Imagine the headphone is represented by a load attached to this voltage source. Then the current output across the load (headphone) will be high (maybe even too high?) if the impedance of the load is low and vice versa for a high impedance load.
 
Assuming this analogy is correct and assuming the power output of the amp is sufficient there should be enough voltage for high impedance headphones and current for low impedance headphone.The only issue I can see with connecting a high impedance amplifier to a low impedance headphone is a low damping factor.

 
Jan 18, 2014 at 7:38 PM Post #12 of 37


No it doesn't work like your assumption.  A tube has a max current it will deliver.  Just assume it is 100 ma for illustration purposes.  It may have 10 volts output, but connect a 10 ohm resistor and it will try to put out 1 amp which is what will drop 10 volts across a 10 ohm resistor.  However it will fail to put out more than .1 amps or 100 milliamps.  It will distort badly in the attempt.  These are not real numbers for tubes btw (other than the 100 milliamp being not so far wrong). 
 
A transformer can trade voltage for current.  Put in high voltage at low current, and get out lower voltage at higher current in the case of tubes.  An OTL doesn't have this help, and tubes don't have high current capability. So they run out of current with too low an impedance load.  With a high enough impedance load they have sufficient current to supply the voltage cleanly without distorting too much.  Multiple taps on a transformer coupled tube amp let you adjust the load and current to something workable for the tubes involved.  Transformers used this way also lower the output impedance of tubes.  So you are correct the higher output impedance of OTL's hurts damping factor. 
 
Jan 18, 2014 at 8:00 PM Post #13 of 37
Do you think the following is a more appropriate circuit diagram:
 
http://lh3.ggpht.com/_gnU30vMbtYg/TX-SPN_L9vI/AAAAAAAAAWU/4kp5Yq-rnbM/s1600-h/output%20impedance%5B2%5D.png
 
This agrees with what you're saying, i.e. if the output (amplifier) impedance is high, current through load is low. If load impedance is high, there will be a high voltage across the load.
 
Jan 18, 2014 at 8:03 PM Post #14 of 37
  Do you think the following is a more appropriate circuit diagram:
 
http://lh3.ggpht.com/_gnU30vMbtYg/TX-SPN_L9vI/AAAAAAAAAWU/4kp5Yq-rnbM/s1600-h/output%20impedance%5B2%5D.png
 
This agrees with what you're saying, i.e. if the output (amplifier) impedance is high, current through load is low. If load impedance is high, there will be a high voltage across the load.


Yes that is part of it.  But even then devices have a max current regardless of voltage.  As you can see in the link however, a high output impedance means a significant portion of the output is wasted elsewhere rather than going to the headphones.  With high impedance phones not much of an issue, with low impedance phones it is.
 
Jan 18, 2014 at 8:45 PM Post #15 of 37
I thought the OTL was more of a compromise. They're cheaper to make for a tube amp.
 

Users who are viewing this thread

Back
Top