> I suppose there are a dozen ways.
That's it. There is no ONE way to measure an audio transformer.
There are some guides. At least one winding must see a moderately low impedance to give a wideband response. But only to a point: as has been said, a "zero" impedance forces the source to work too hard and is unreasonable.
Perhaps the best bet (though NOT the most popular today) is to terminate in design impedances. Low-Z source, through 5k resistor, to plate winding, and load secondary with the 8 Ohms. This gives bass extension more than a pentode will, less than a triode will, so is a good selection parameter.
That's it. There is no ONE way to measure an audio transformer.
There are some guides. At least one winding must see a moderately low impedance to give a wideband response. But only to a point: as has been said, a "zero" impedance forces the source to work too hard and is unreasonable.
Perhaps the best bet (though NOT the most popular today) is to terminate in design impedances. Low-Z source, through 5k resistor, to plate winding, and load secondary with the 8 Ohms. This gives bass extension more than a pentode will, less than a triode will, so is a good selection parameter.
Not really knowing how to test one of these measurement wise... But I'm good at hearing differences. I might be inclined to replace one channel of an EL84 amp with one of these 8k 10 watt torroids. Leaving the other channel alone, with an EI 8k 10w, to be the reference. Feed both channels with the same mono music source, playing into both channels at once. Rig an A/B switch at the input jacks. Then just flip the switch between the two channels (volumes the same). Then just listen. Left speaker is the EI OPT and the right speaker is the torroid? Fast flip A/B is something I've always felt comfortable with psychoacoustically where I'm not fooling myself because the flip is so instantaneous. One could flip right in the middle of a cymbal decay or in the middle of a vocal phrase or in the middle of a violin solo.
Maybe a simple in room measurement with a microphone and then compare toa similar known-good transformer?
I'm still rather curious for a pair, and would be interested in buying a set, should someone like to have me send them to them first to run some tests, then send them along to me (postage paid by me, of course) to use in a build. Hmmmm.
I'm still rather curious for a pair, and would be interested in buying a set, should someone like to have me send them to them first to run some tests, then send them along to me (postage paid by me, of course) to use in a build. Hmmmm.
Windcrest77,
I like your proposed listening method. E-I, then Toroid, then E-I, then Toroid, etc.
It reminds me of a presentation and listening session I did at VSAC 2008.
The listening comparison was a parallel single ended 300B amp, versus a push pull 300B amp.
The push pull 300B amp drove a 6k a-a transformer, the parallel single ended 300B amp drove a 1500 ohm transformer (in the class A region, all four 300B "saw" a 3k load impedance).
Each of the four 300B plate currents were the same (about 55mA).
The left channel music output from a CD player was applied to both amps; and the 2 speakers faced each other with the speaker's front panels about 1/2 inch away. One of the loudspeakers was purposely connected in opposite phase.
The volume controls were adjusted until the sound was nulled out completely (dead silence). That means the amplifier gains were now identical.
Then the loudspeaker was connected back to in-phase.
After the gain balance procedure, the Left channel music output of a CD player was alternately switched from the push pull amp, to the single ended amp, and back and forth multiple times.
I like your proposed listening method. E-I, then Toroid, then E-I, then Toroid, etc.
It reminds me of a presentation and listening session I did at VSAC 2008.
The listening comparison was a parallel single ended 300B amp, versus a push pull 300B amp.
The push pull 300B amp drove a 6k a-a transformer, the parallel single ended 300B amp drove a 1500 ohm transformer (in the class A region, all four 300B "saw" a 3k load impedance).
Each of the four 300B plate currents were the same (about 55mA).
The left channel music output from a CD player was applied to both amps; and the 2 speakers faced each other with the speaker's front panels about 1/2 inch away. One of the loudspeakers was purposely connected in opposite phase.
The volume controls were adjusted until the sound was nulled out completely (dead silence). That means the amplifier gains were now identical.
Then the loudspeaker was connected back to in-phase.
After the gain balance procedure, the Left channel music output of a CD player was alternately switched from the push pull amp, to the single ended amp, and back and forth multiple times.
Cool. Yes the high end audiophile industry is based on the fact that there is no "perfect reference" that anyone is capable of actually remembering as they move from listening session to session with time in-between to prejudice.
Fast A/B flipping may not be perfect but it allows one to maybe detect how one system might be "different" than another, not better or worse, just in what way they are physically different sounding. And that in itself is valuable.
Now if I listened to your PP amp today then listened to your SE amp tomorrow (same source, same volume) and you asked me that second day how they were "different". I couldn't give you an objectively honest answer. I would be telling you something I've already processed for 24 hours and rationalized and polluted with all my prejudices about other things since yesterday, wires, topology, the weather, let alone just memory.
But if you asked me how these two amps sounded differently with only 5 milliseconds of time in-between. And I got multiple tries. I think I can give you an honest answer. The audiophile industry is full of buyers who purchase what they think sounds "better", but that difference detection is highly prejudiced. But with only a millisecond of thinking between the listening sessions, there is no time to become subjective and psychological, its all physical at that point, you either can hear a difference or not. The timbre of the trumpet changed in this way, the overtones of the stand up bass popped more on amp A, etc. To me those would be more honest observations because 5 milliseconds is not enough time to apply ones human prejudices or have to exercise the memory at all.
And its ok to simply like something too because of what its topology is. At Axpona year prior I consistently liked most $2,000 horn speakers better than the $50,00 Focal setup. Pre-prejudice, maybe, or maybe I just like that live edgy projection of horns. But speakers are a whole other thing, drastically different sounding, not like comparing two amps at all.
nerdorama and Windcrest77,
The speakers I used at VSAC were commercial models, and were very well matched.
After I gave that demonstration, I realized the one factor that made the demo far less good than it could have been:
The room was full of listeners. Some could see one speaker, some could see the other speaker, some could see both speakers, some could not see either speaker.
And some had a good left ear, some had a good right ear, and some had two good ears.
After I did the gain balance, I should have swapped the speakers to the opposite amplifier. Then I needed to make sure the sound was still nulled. If so, then the balance was still proven good, regardless of which speaker was on which amplifier, parallel single ended, or push pull.
Then disconnect one speaker, and put it under the table.
Then, I should have used a switch to switch Both the Inputs And the Outputs to a Single speaker and a single amplifier at a time.
That way, the sound path to anybody's ear would have remained exactly the same ( the same speaker switched back and forth to the parallel single ended, and the push pull amp.
I learned that too late (thought of it myself, after VSAC 2008).
I really like sound/listening demos when they are carried out correctly.
At two more demos at VSAC 2008, I was asked to bring those speakers, and a pair of matched amplifiers (I had made two identical 2A3 amplifiers).
That was easier, I did not have to make an overhead slide show, and did not have to stage anything, just bring the speakers an amplifiers.
One demo was to illustrate the sound of correct versus incorrect stylus vertical angle on a phono cartridge. I was surprised at how easy it was to hear, both too far negative or too far positive versus the correct stylus vertical angle.
The other demo was to illustrate the sound of a highly compressed musical passage, versus the same musical passage that was not compressed. That was easy to hear the difference. I do not like highly compressed music.
Come on Posters!
Come up with another good demo that we all can perform on our own systems.
Keep designing, keep building, keep listening!
The speakers I used at VSAC were commercial models, and were very well matched.
After I gave that demonstration, I realized the one factor that made the demo far less good than it could have been:
The room was full of listeners. Some could see one speaker, some could see the other speaker, some could see both speakers, some could not see either speaker.
And some had a good left ear, some had a good right ear, and some had two good ears.
After I did the gain balance, I should have swapped the speakers to the opposite amplifier. Then I needed to make sure the sound was still nulled. If so, then the balance was still proven good, regardless of which speaker was on which amplifier, parallel single ended, or push pull.
Then disconnect one speaker, and put it under the table.
Then, I should have used a switch to switch Both the Inputs And the Outputs to a Single speaker and a single amplifier at a time.
That way, the sound path to anybody's ear would have remained exactly the same ( the same speaker switched back and forth to the parallel single ended, and the push pull amp.
I learned that too late (thought of it myself, after VSAC 2008).
I really like sound/listening demos when they are carried out correctly.
At two more demos at VSAC 2008, I was asked to bring those speakers, and a pair of matched amplifiers (I had made two identical 2A3 amplifiers).
That was easier, I did not have to make an overhead slide show, and did not have to stage anything, just bring the speakers an amplifiers.
One demo was to illustrate the sound of correct versus incorrect stylus vertical angle on a phono cartridge. I was surprised at how easy it was to hear, both too far negative or too far positive versus the correct stylus vertical angle.
The other demo was to illustrate the sound of a highly compressed musical passage, versus the same musical passage that was not compressed. That was easy to hear the difference. I do not like highly compressed music.
Come on Posters!
Come up with another good demo that we all can perform on our own systems.
Keep designing, keep building, keep listening!
Last edited:
6a3summer. I think the key point all trade shows demos fall down on is the first question they ask the folks in the chairs. It's usually "which do you like better?" Or "which is better?". Those are questions that are soliciting a value judgement, even before they know if they are even capable of making a physical difference judgement. The exercise should be "how do these sound differently?", and leave it all at that. Then as the AB flipping occurs, occasionally do a fake flip that doesn't flip at all. The demo should strive to solicit honest answers about what is simply different between a or b. Then with the confidence of knowing they can do that, just tell a difference, they can go off and make their own value judgement on their own. Maybe the demo leader should attempt to teach the room about what differences to see if they can detect, sort of like ear training.
And yeah compressed music sounds amazing but not after 30 minutes continuous, at some point anyone who has been to concerts realizes that all the instruments are the same volume and on the same plane. Compressed music is like a piece of cheesecake, amazing, a stimulator of taste buds, but after the third piece you regret it all.
And yeah compressed music sounds amazing but not after 30 minutes continuous, at some point anyone who has been to concerts realizes that all the instruments are the same volume and on the same plane. Compressed music is like a piece of cheesecake, amazing, a stimulator of taste buds, but after the third piece you regret it all.
Last edited:
Windcrest77,
1. I do like your idea of ear training. But I think that is for a separate presentation.
2. At VSAC 2008, for both the overhead explanation of the two amplifiers, and for the listening session, I merely said the purpose of the demo was to listen for the differences, not to decide which was better.
I did not want to tell them what they might or might not hear.
That would give them a pre-conception.
What the listeners did, was up to them, and was varied.
For the other presentation, the compressed music was modern music, not Haydn's surprise symphony.
Probably the best application of compressed music (and FM radio compression) is when you are driving down the road in a noisy car (or a convertible with the top down).
I am not sure what my high school band instructor would have thought of 'todays' compressed music (but I have a pretty good idea of how he would respond).
Another experiment, during high school years was to take a shure 'bullet' microphone and a DuMont oscilloscope and look at the waveform of our premier French Horn player.
The second harmonic was very easily seen in the waveform. Then the horn player changed his technique, and the most perfectly formed sine wave was displayed. He also was capable of playing pedal tones (low notes) that most professional horn players could not. He went to Cornell University after graduation.
1. I do like your idea of ear training. But I think that is for a separate presentation.
2. At VSAC 2008, for both the overhead explanation of the two amplifiers, and for the listening session, I merely said the purpose of the demo was to listen for the differences, not to decide which was better.
I did not want to tell them what they might or might not hear.
That would give them a pre-conception.
What the listeners did, was up to them, and was varied.
For the other presentation, the compressed music was modern music, not Haydn's surprise symphony.
Probably the best application of compressed music (and FM radio compression) is when you are driving down the road in a noisy car (or a convertible with the top down).
I am not sure what my high school band instructor would have thought of 'todays' compressed music (but I have a pretty good idea of how he would respond).
Another experiment, during high school years was to take a shure 'bullet' microphone and a DuMont oscilloscope and look at the waveform of our premier French Horn player.
The second harmonic was very easily seen in the waveform. Then the horn player changed his technique, and the most perfectly formed sine wave was displayed. He also was capable of playing pedal tones (low notes) that most professional horn players could not. He went to Cornell University after graduation.
I bought a single MP-10W80 for low-fi experiments.
I am aware that it is difficult to get meaningful iron core inductance measurements that compare to actual operating conditions. I take what I get with an entire box of salt.
That said, and having measured power, output & mic transformers with skepticism, usually only extracting some vague generalizations, I've never seen anything so weird.
It's a given that inductance will usually be drive level and frequency dependent if using a small signal impedance or LRC analyzer.
Best I could provide was 10 Vrms and the analyzer objected at 20 Hz.
I got crazy leakage inductance numbers, with the secondary leakage inductance with the primary shorted producing about 8x the 'true' inductance. I have never encountered that before. Maybe I lead a sheltered life.
I decided the analyzer was unhappy with the transformer, or I was just having a bad day with my choices of measurement conditions.
Last thing I did was switch to impedance mode and look at magnitude & phase angle to see at what frequency phase angle hits either 45 degrees (-6 dB voltage point) or 0 degrees which I interpret as a ballpark self-resonant-frequency. With mic transformers, this produces such wide implied BW I question whether it's realistic (like maybe I need a termination impedance).
I was pretty horrified to find Z=1 Mohm/1.5 degrees at only 375 Hz!
In this scenario, I am probably not even going to bother retrying measurements. I'll just try an amp build. It can't be as bad as it looks!
With a random unknown transformer, bad measurements are usually Nature's Warning Signs you may have a really poor candidate for your project.
Then there is the phenomenon of communication with Antek. Once I got a response to an inquiry for add'l tech data that they didn't have it. All other inquiries were ignored. Even after a purchase.
I just today noticed their logo on the packing slip says 'Everything shipped from the USA'.
I was going to wonder aloud if there was a language barrier, but I have never had one with other countries. Sometimes they'll refuse proprietary data.
Selling to a DIY market, intentionally or accidentally, usually offers opportunity for mutual benefits or mutual revulsion.
Hard to tell what this communication blackout means. Their strength seems to be attractive prices, no minimum order and easy transactions if you don't ask questions.
There is far more empirical experience/evidence in our forums for their toroidal PT's misused as OT's. I think I should just compare one of their 240:7 V PT's to this and gain more answers than asking them.
I am aware that it is difficult to get meaningful iron core inductance measurements that compare to actual operating conditions. I take what I get with an entire box of salt.
That said, and having measured power, output & mic transformers with skepticism, usually only extracting some vague generalizations, I've never seen anything so weird.
It's a given that inductance will usually be drive level and frequency dependent if using a small signal impedance or LRC analyzer.
Best I could provide was 10 Vrms and the analyzer objected at 20 Hz.
I got crazy leakage inductance numbers, with the secondary leakage inductance with the primary shorted producing about 8x the 'true' inductance. I have never encountered that before. Maybe I lead a sheltered life.
I decided the analyzer was unhappy with the transformer, or I was just having a bad day with my choices of measurement conditions.
Last thing I did was switch to impedance mode and look at magnitude & phase angle to see at what frequency phase angle hits either 45 degrees (-6 dB voltage point) or 0 degrees which I interpret as a ballpark self-resonant-frequency. With mic transformers, this produces such wide implied BW I question whether it's realistic (like maybe I need a termination impedance).
I was pretty horrified to find Z=1 Mohm/1.5 degrees at only 375 Hz!
In this scenario, I am probably not even going to bother retrying measurements. I'll just try an amp build. It can't be as bad as it looks!
With a random unknown transformer, bad measurements are usually Nature's Warning Signs you may have a really poor candidate for your project.
Then there is the phenomenon of communication with Antek. Once I got a response to an inquiry for add'l tech data that they didn't have it. All other inquiries were ignored. Even after a purchase.
I just today noticed their logo on the packing slip says 'Everything shipped from the USA'.
I was going to wonder aloud if there was a language barrier, but I have never had one with other countries. Sometimes they'll refuse proprietary data.
Selling to a DIY market, intentionally or accidentally, usually offers opportunity for mutual benefits or mutual revulsion.
Hard to tell what this communication blackout means. Their strength seems to be attractive prices, no minimum order and easy transactions if you don't ask questions.
There is far more empirical experience/evidence in our forums for their toroidal PT's misused as OT's. I think I should just compare one of their 240:7 V PT's to this and gain more answers than asking them.
> their logo on the packing slip says 'Everything shipped from the USA'.
Probably 100% true. We know where Antek is because they had a fire a few years ago. 20 miles from where I used to live. I Googled and it is an old warehouse down an alley in an older city in NJ. Apparently every week or two a shipment comes into that warehouse from Asia.
Probably 100% true. We know where Antek is because they had a fire a few years ago. 20 miles from where I used to live. I Googled and it is an old warehouse down an alley in an older city in NJ. Apparently every week or two a shipment comes into that warehouse from Asia.
). Mine is wearing a protective hat and gloves.
The idea regarding a SS power amp feeding the secondeary isn't that bad at all, but has to be modified somewhat: Put a series resistor with the nominal load value between amp and the OT, put a resistor of the nomal primary impedance value in parallel with the primary, and do the measurements.
Best regards!
Best regards!