As long physics tells me that i am right i prefer HiB above EI materials, even when it's nikkel EI.....
But you can give me some datasheets to proof that i am wrong of course
I think that it is a waste of time because I have already tried to tell what is the relevance of leakage flux going from EI to C. The EI transformer can easily have square cross-sectional area which guarantees minimum leakage inductance where the typical 2C core does not. If you use a 2C transformer the typical cores have cross-sectional areas far from being square. So almost all the advantage of the C shape is lost. Only if you use 4C's you can get close. Then you can use a mu-metal can for shielding.... The difference is negligible.
You could not believe what I am saying and we could suppose that you know physics and I don't. Then you could tell Sowter how to make a better transformer because they don't know the physics very well.
I think many people are very optimistic about their knowledge of physics. Oh, by the way, I have got a PhD in physics and you?
Last edited:
I completely agree with what Bas/Pieter is saying, we want a high perm (amorphous), low loss "series" Feed ~60 mA C-core with 5k primary and 4 x 8 ohm secondary's.
To me this is a very exciting to have somewhat of a "spec" for the DHT headphone amplifiers. A quality transformer for headphones has been a long time coming.
But we need to somehow keep costs and the commercial side of a this out of this thread . I'm pretty sure we don't want to be discussing pricing here.
I am reaching out to the mods to give direction on the best way to proceed , possibly Bas open a thread under group buy? Or we can just email for pricing separately (but it would be nice to somehow get a volume price break.)
If cost is not interessting: nanocrstaline materials are the best or for mc high permeable cobalt based amorphous with a flatloop(even better).
Anyway nanacristaline c-core are much worse to get. Manufactures still trying to improve there skills.
I had some of this c-core at home but i was not impressed so far. Hope it will be better now.
VAC (Vacuumschmelze) is now making a new batch nanokristaline c-cores so now can be your moment......
I do. Well enough to tell you that you don't know enough about real transformers! Otherwise you wouldn't ask to demonstrate anything, you would have done it yourself to make your point.
Once you have high grade thin laminations for EI cores there is no going from good to super good. Their loss will be very low as well and a low leakage inductance will influence your transformer performance way above the audio band.
The round perimetrical shape of C cores is not a clear advantage because their cross-sectional shapes are not ideal, typically! The best cross-sectional areas are circular (R-core) first and square second. None of the two is typical of C cores. Then you can have a mu-metal can. Do you have an idea of what this does to the outer magnetic field? A low leakage inductance depends mostly on the geometry, balance, sectioning of the coil(s) and optimization which is not a property of a specific magnetic core. If the coil is not good the transformer will be rubbish. If the core is REAL high grade EI going to HiB C will give negligible improvements except the headroom as the HiB saturates at higher flux (but this is not a rule and will always depend on the actual design!). I gave you a practical example with the Tango's so that you could try by yourself. The U-808 can withstand less DC current but it is actually slightly better at low frequency than the XE-20S because (when used within specs) it has a slightly higher inductance and slightly lower distortion, all the rest equal. At higher frequency they show perfectly matching frequency responses up to 25 KHz. At 50 KHz the U-808 is 1 dB lower. But despite of this the square wave response at 20 KHz (twenty!) of the U-808 is just as good and a lot better than most C-core transformers!! Then is time to listen to them and I challenge anyone to tell me the difference without knowing which one is on! And now I think it's enough.
Last edited:
In my opinion the u808 and xe20s are very mediocre transformers. I can make much better
Btw, the U808 was tested in 'L Audiophile long time ago and the measurements was just avarage ( sound too)
Yes sure, l' Audiophile is just the right source for you. Let's see these transformers of yours how reproduce square waves at 20 KHz. This is the U-808:
Attachments
http://www.tubebooks.org/images/Tango_U808.PDF
High loss, very weak bass, mediocre high. Today a top transformer can do much better.
http://www.monolithmagnetics.com/images/stories/datasheet_S-9.pdf
Last edited:
I am afraid I will not discuss with someone who has no real arguments. An average data-sheet against a real measurement of the same thing is not what I would call a fair response. As I told you can believe that donkeys fly. Bye.
In real world these transformers do exactly the same as in de datasheet. I think you are not fare to these manufactures. They don't lie about this.
I also did measurements myself on a Plitron PP transformer recently and it did actual even measured better (frequency respons and loss) then the datasheet they publish on internet.
Guys, it's obvious you have different views; perhaps it's better to agree to disagree.
Please, let's not turn the thread into a "I can prove you wrong" debate.
Thanks for understanding!
Regarding the shipping of 307A to Ale. I am willing to cover a bit of the shipping cost for whoever in EU sends them, in the interest of science.
Please, let's not turn the thread into a "I can prove you wrong" debate.
Thanks for understanding!
Regarding the shipping of 307A to Ale. I am willing to cover a bit of the shipping cost for whoever in EU sends them, in the interest of science.
Guys, it's obvious you have different views; perhaps it's better to agree to disagree.
Please, let's not turn the thread into a "I can prove you wrong" debate.
Thanks for understanding!
Thanks Iko.
Els & 45 understand the current state of affairs. I do greatly appreciate the contributions from both of you, but my custom series feed 5k:32 EI has a DCR of 12 ohms! And my custom 1W nickel core parafeed 5k:32 had a gap you could drive a truck thru
So as far as nickel series feed and other exotic, Tango, Tamura or Hashimoto will not make a pair of custom headphone output transformers. And as much as Lundahls specs look great they just don't have the off the shelf config we need (their 8 ohm configs are not symmetric.). The current bar is pretty low for custom headphone transformers, as in junior high school physics even in commercial headphone amps that cost thousands, PhD level is probably years away as there aren't many true SET headphone enthusiasts. Note this is a practical engineering argument
Little engineering jab at the scientist to lighten the atmosphere
So I/we need someone who is willing to work with us, I'm willing to try a nano crystalline C-Core. The concept looks sound, lower DCR bit higher perm should help I think.
Thanks Iko.
Els & 45 understand the current state of affairs. I do greatly appreciate the contributions from both of you, but my custom series feed 5k:32 EI has a DCR of 12 ohms! And my custom 1W nickel core parafeed 5k:32 had a gap you could drive a truck thru
The current bar is pretty low for custom headphone transformers, as in junior high school physics even in commercial headphone amps that cost thousands, PhD level is probably years away as there aren't many true SET headphone enthusiasts. Note this is a practical engineering argument
So I/we need someone who is willing to work with us, I'm willing to try a nano crystalline C-Core. The concept looks sound, lower DCR bit higher perm should help I think.
Her a adres where you can buy nanocrystalline c-cores:
SBK nanokristallin VITROPERM*: Sekels GmbH
Btw, i have a windingmachine in case off.......
Site says: Advantages of nanocrystalline c-cores are low losses, high induction and low noise when operating at accoustic frequencies due to the low magnetostriction.
Already looks better than EI M19 to me. But as far as winding my own, isn't going to happen, there is so much skill involved in that practice.
Back to the tubes I found that KR sells a PX25 that matches the original british specs fairly well. The tube has an amplification factor of 9.5 which along with a stepup input transformer would allow enough gain for an all DHT with a lot of power. The nice thing is it uses the same socket as the 300B so it would be easy to switch the coleman reg to allow there use if one finds they need more gain.
Already looks better than EI M19 to me. But as far as winding my own, isn't going to happen, there is so much skill involved in that practice.
Back to the tubes I found that KR sells a PX25 that matches the original british specs fairly well. The tube has an amplification factor of 9.5 which along with a stepup input transformer would allow enough gain for an all DHT with a lot of power. The nice thing is it uses the same socket as the 300B so it would be easy to switch the coleman reg to allow there use if one finds they need more gain.
In real world one should look a bit more carefully. If they do what they say then the square wave response will not look nice. Those Monolithmagnetics are not well sorted out because you will see a nice poorly damped ringing even at 10KHz! Above 100KHz their response is artificial because that is just undamped resonance occurring at 102 KHz. This can also be understood just noticing that the response starts to roll-off and then goes up again, without any calculation. Plitron's 300B OPT is a lot better than yours and the high frequency behavior is perfectly in line with the Tango with extended response.
The U-808 has a desirable balance, its self-resonance is lower at about 60 KHz because of higher capacitances (measured leakage inductance is only 5.7 mmH against 5.3 of your "super" transformer) but there is a super-smooth roll-off just below the resonance with the right damping. It's so good that you can use it with many other valves, not just the 300B, with or without feedback with minimum troubles about stability. In fact is a truly general purpose transformer! Let's not forget that it has 3 different primary taps on a small core.
So, the story is that you where praising the lower leakage flux of the HiB but this is a clear demonstration that it doesn't translate in lower leakage inductance even in comparison to M6 grain oriented E+I core! Exactly what I have been telling from the beginning.
Regarding the primary inductance there are different methods to measure it (i.e. different frequencies and input signals, mainly) and they can give quite different numbers.....
Last edited:
I don't think I am missing anything. EI core thin laminations have low losses, nickel EI laminations have very low losses. You really don't know what you are talking about.
You can only dream my reference system, from the source to the room.
You can see the speaker in my avatar. If you have 200000 euros you can buy it!
I keep coming back to this thread, but it just gets worse.
Now it's up to 200 thousand euros.
When you were talking about the Frank Cooter design in the first post, I was interested. Not that I'm not still interested, but no way am I, or many of the other people who might have been interested in building one, going to build some monstrous 300B design with enough power to drive speakers, just to accommodate a pair of outrageous headphones that I'm never going to own. It's not as if the HE-6s do that wonderfully well in reviews, and the Senn HD800s are much easier to drive.
Nearly 640 posts later and Pano is the only one with a working amplifier.
I really think you've lost your way. Chuck out the requirement to drive every headphone pair under the sun, and do something with a bit broader appeal. If you want a challenge, make it the challenge of producing a design that is readily adaptable to the large majority of phones, uses off-the-shelf or at least catalog parts, and doesn't cost the earth. Something a bit closer to the spirit of what Frank did.
Now you're down to half-a-dozen contributors, some of them giving it 'I'm-so-smart' or 'I'm-so-rich.' It's not very edifying.
Anyway, please don't take offence, it's your thread, you can do what you want. If you want to be a bunch of audiophiles in the worst possible sense of the word, I can't prevent you. I just think you need a bit of a reality check.
Now it's up to 200 thousand euros.
When you were talking about the Frank Cooter design in the first post, I was interested. Not that I'm not still interested, but no way am I, or many of the other people who might have been interested in building one, going to build some monstrous 300B design with enough power to drive speakers, just to accommodate a pair of outrageous headphones that I'm never going to own. It's not as if the HE-6s do that wonderfully well in reviews, and the Senn HD800s are much easier to drive.
Nearly 640 posts later and Pano is the only one with a working amplifier.
I really think you've lost your way. Chuck out the requirement to drive every headphone pair under the sun, and do something with a bit broader appeal. If you want a challenge, make it the challenge of producing a design that is readily adaptable to the large majority of phones, uses off-the-shelf or at least catalog parts, and doesn't cost the earth. Something a bit closer to the spirit of what Frank did.
Now you're down to half-a-dozen contributors, some of them giving it 'I'm-so-smart' or 'I'm-so-rich.' It's not very edifying.
Anyway, please don't take offence, it's your thread, you can do what you want. If you want to be a bunch of audiophiles in the worst possible sense of the word, I can't prevent you. I just think you need a bit of a reality check.
Vegard will kindly send a sample for tracing. Will post curves and spice model then
Cheers
Ale