This is something that has been bothering me and doesnt make much sense.
The transformers with 4-8-16ohm taps. Ive heard that they only use the whole winding at the 16 ohm tap, and that if you use the 4 ohm tap your only using a 1/4 of the iron and its going to perform like a transformer thats rated much smaller. Literally its a 16ohm coil with taps in the center and the 1/4th mark.
Now in my mind i thought that these transformers just had quad secondaries.
4ohms, all 4 secondaries are wired in parallel.
8ohms, 2 sets are wired in series, then paralleled.
16ohm all 4 sets are wired in series.
In this way the whole iron is used at any given time. Or am i completely wrong?
The transformers with 4-8-16ohm taps. Ive heard that they only use the whole winding at the 16 ohm tap, and that if you use the 4 ohm tap your only using a 1/4 of the iron and its going to perform like a transformer thats rated much smaller. Literally its a 16ohm coil with taps in the center and the 1/4th mark.
Now in my mind i thought that these transformers just had quad secondaries.
4ohms, all 4 secondaries are wired in parallel.
8ohms, 2 sets are wired in series, then paralleled.
16ohm all 4 sets are wired in series.
In this way the whole iron is used at any given time. Or am i completely wrong?
Impedance ratio is not proportional to number of turns, it is a square of number of turns. To double impedance you only need to increase the number of turns by a square root of 2, meaning two sections of a 4R & 8R transformer are not identical (real life figures of one of my OPTs are 100 turns for 4R and additional 41 turns for entire secondary winding of 8R).
So yeah, you are completely wrong
So yeah, you are completely wrong
People say believe half of what'cha see; some or none of what'cha hear...
You're right, it doesn't make sense and shouldn't
It's completely bogus to think that fewer turns on the secondary == less use of the iron or smaller rating.
The thing you heard was "made-up", possibly to explain something or more likely to sell something, or perhaps they just didn't remember the facts, so...
There may be a slight disadvantage in frequency response with a multi-tap winding but there is no derating.
What this world needs is a good factual primer on electromagnetics... If only Michael Faraday was still with us.
Michael
Last edited:
secondary impedance
Hi,
You are right, and you are not quite right.
Let's take an example of a single ended output transformer with a 5k primary impedance, and 4, 8 and 16 ohm secondaries.
This transformer can be wound using different winding techniques.
In our example we have 2500 primary windings.
This means that to get 4, 8 and 16 ohm secondaries we must have 71 secondary windings for 4 ohm (70,7 to be exact), 100 secondary windings for 8 ohm, and 142 secondary windings for 16 ohm (141,4 to be exact).
The "cheapest" way to do this is to make a 16 ohm secondary of 142 windings, and make taps at 71 windings and 100 windings for 4 and 8 ohms.
We could repeat this winding technique using several parallelled secondary layers interleaved with primary layers; we call these "sections".
However using the 4 or 8 ohm tap, part of the winding is "unused". It is not correct to say that part of the core is unused, but the coupling between primary and secondary is impaired because of the unused part of the secondary windings. This will result in high frequency loss, in the worst case already within the audio range.
A better way to make multiple secondary connections in the same output transformer is to wind a number of 1 ohm secondary sections. In our example 35 windings would make a 1 ohm secondary (35,35 to be exact). Then in order to have a 4 ohm connection (70 windings) we must connect two of these 1 ohm secondaries in series; for 8 ohm we need three 1 ohm secondaries in series (105 windings); for 16 ohm we connect four 1 ohm secondaries in series (140 windings). So we need at least four 1 ohm secondaries for this winding technique. Actually the 8 ohm connection is 9 ohm; we could say 4, 9 and 16 ohm (2², 3² and 4²). In case of 4 ohm two secondaries are in series, and parallelled with the two other secondaries which are in series. In case of 8 ohm the fourth secondary is parallelled with one of the three other secondaries which are in series.
Quite a story but I hope it throws a bit light on the matter.
Pieter
Tribute Audio Transformers
Hi,
You are right, and you are not quite right.
Let's take an example of a single ended output transformer with a 5k primary impedance, and 4, 8 and 16 ohm secondaries.
This transformer can be wound using different winding techniques.
In our example we have 2500 primary windings.
This means that to get 4, 8 and 16 ohm secondaries we must have 71 secondary windings for 4 ohm (70,7 to be exact), 100 secondary windings for 8 ohm, and 142 secondary windings for 16 ohm (141,4 to be exact).
The "cheapest" way to do this is to make a 16 ohm secondary of 142 windings, and make taps at 71 windings and 100 windings for 4 and 8 ohms.
We could repeat this winding technique using several parallelled secondary layers interleaved with primary layers; we call these "sections".
However using the 4 or 8 ohm tap, part of the winding is "unused". It is not correct to say that part of the core is unused, but the coupling between primary and secondary is impaired because of the unused part of the secondary windings. This will result in high frequency loss, in the worst case already within the audio range.
A better way to make multiple secondary connections in the same output transformer is to wind a number of 1 ohm secondary sections. In our example 35 windings would make a 1 ohm secondary (35,35 to be exact). Then in order to have a 4 ohm connection (70 windings) we must connect two of these 1 ohm secondaries in series; for 8 ohm we need three 1 ohm secondaries in series (105 windings); for 16 ohm we connect four 1 ohm secondaries in series (140 windings). So we need at least four 1 ohm secondaries for this winding technique. Actually the 8 ohm connection is 9 ohm; we could say 4, 9 and 16 ohm (2², 3² and 4²). In case of 4 ohm two secondaries are in series, and parallelled with the two other secondaries which are in series. In case of 8 ohm the fourth secondary is parallelled with one of the three other secondaries which are in series.
Quite a story but I hope it throws a bit light on the matter.
Pieter
Tribute Audio Transformers
Thanks for the factual explanation. So with cheap winding techniques you can end up with lower frequency response performance when using certain taps.
Many transformer builders recommend a single secondary winding for this reason. Maybe for the ultimate no-compromise performance.
I'm still not aware of any transformers that will "perform like a much smaller transformer" whan using a lower impedance output tap.
Cheers,
Michael
Putting a 4 ohm speaker on the 8 ohm tap of an existing amplifier, specifically a tube amp, reduces the load impedance seen by the output tube(s) by 1/2. With a class A triode that will increase distortion and current draw, reduces damping factor, and might increase power output depending on the idle current and amp topology.
Designing an amp to use a 4 ohms load on the 8 ohm tap and provide the correct anode load will still have 2 problems. The DC resistance of the coil will be excessively high for 4 ohms, limiting the damping factor, and the inductance of the primary will be too low for the primary circuit impedance, resulting in loss of low frequency power and range.
Putting 8 ohms on the 4 ohm tap of an existing amplifier may be OK but might not be good for high-gNFB amps. The result is a flatter load line on the output tube(s) resulting in higher voltage swing (if the circuit allows it) and improved damping factor. The DCR of the output coil will be low relative to the 8 ohms, further improving the damping factor. Maximum output power, all other things being equal, will be 1/4 (1/2 the voltage headroom). In reality you get a little more than 1/4. I have been happy with the results of doing this in the case of a zero-nfb triode amp and a box speaker, even though the power is down by 6db. I suppose you end up with extra iron, but it seems to me there will be higher flux density relative to the signal level to compensate.
Michael
It seems my question went the wrong way. I was asking if putting an 8ohm speaker on a 4 ohm transformer would reduce audio quality.
I have both 4ohm and 8ohm speakers that i like in the house. Some have told me to just get a 6ohm transformer. Others have said stay away from multi tap transformers.
Im trying to find the best solution less having 2 sets of transformers. If the only compromise of running 8ohm speakers on a 4ohm transformer is reduced output, ill do that.
I have both 4ohm and 8ohm speakers that i like in the house. Some have told me to just get a 6ohm transformer. Others have said stay away from multi tap transformers.
Im trying to find the best solution less having 2 sets of transformers. If the only compromise of running 8ohm speakers on a 4ohm transformer is reduced output, ill do that.
Nightanole,
With 8 ohm load on 4 ohm transformer you reduce max power in watts by (a little less than) half, not a quarter.
Presume you have a 2k5 / 4 ohm transformer. Loading with 8 ohm means that, besides the loss of power, the tube will "see" a 5k load.
It depends on your type of loudspeaker how this will change audio quality.
When you have low efficiency loudspeakers you might run out of power.
With higher efficiency loudspeakers you might get away with the power loss and might even benefit from the higher damping factor (important in case of bass reflex loudspeakers). At the end you must try and listen.
Multiple secondary impedance output transformers can be done without loss of quality. My latest 4 - 9 - 16 ohm generation single ended output transformers (also with 2k5 and 3k6 primary impedance to choose) have equal bandwidth for all impedance settings (15Hz - 40kHz minus 1 dB).
Pieter
Tribute Audio Transformers
With 8 ohm load on 4 ohm transformer you reduce max power in watts by (a little less than) half, not a quarter.
Presume you have a 2k5 / 4 ohm transformer. Loading with 8 ohm means that, besides the loss of power, the tube will "see" a 5k load.
It depends on your type of loudspeaker how this will change audio quality.
When you have low efficiency loudspeakers you might run out of power.
With higher efficiency loudspeakers you might get away with the power loss and might even benefit from the higher damping factor (important in case of bass reflex loudspeakers). At the end you must try and listen.
Multiple secondary impedance output transformers can be done without loss of quality. My latest 4 - 9 - 16 ohm generation single ended output transformers (also with 2k5 and 3k6 primary impedance to choose) have equal bandwidth for all impedance settings (15Hz - 40kHz minus 1 dB).
Pieter
Tribute Audio Transformers
I suppose, but this does need to be balanced against damping factor, depending on the speakers. Hiraga worked with what, horns and baffles?
I'll need to look into that. Is it in SP?
Agreed, with a low DCR transformer the relative increase in DCR is less of an issue. Assuming the desired DF is in the range of about 2.5 to 8.
My point about inductance is more in the context of SET, that when you use a transformer built for a given primary circuit impedance at a much higher primary circuit impedance than it was designed for, the inductance as designed may not be sufficient for the same low frequency extension. e.g. the transformer built for lower impedance will have less L due to needing to support more DC current. The smaller L driven by higher impedance will suffer at the low frequency end.
And of course the power output is 1/2 when using an 8 ohm speaker on the 4 ohm tap
Cheers,
Michael
Hello Michael,
Jean Hiraga investigated the influence of primary impedance on second and third order distortion in an article in his magazine "Maison de l'Audiophile" no. 41 (1987) which describes the design of his Lectron JH50 amplifier. He shows K2 and K3 curves for the 300B and EL34. When you can read French I'd be happy to scan the article and send it to your email address.
Regards,
Pieter
Tribute Audio Transformers
Jean Hiraga investigated the influence of primary impedance on second and third order distortion in an article in his magazine "Maison de l'Audiophile" no. 41 (1987) which describes the design of his Lectron JH50 amplifier. He shows K2 and K3 curves for the 300B and EL34. When you can read French I'd be happy to scan the article and send it to your email address.
Regards,
Pieter
Tribute Audio Transformers
HI Pieter nice to meet here , I have ask you a nice amorphous 3.6k multitap opt ,my full range is 5.5ohm is possible 4-5,5 (6)-16 ?
And on primary add a taps for 2.5k have trade off ?
best
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.