• 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.

One more 4P1L SE

regal how many mA needs the Senns?

Wich one Sowter is more suitable for HD600?

TRANSFORMERS FOR HEADPHONE AMPLIFIERS

For preamp I will need other transformer like the Lundhal LL1660 that I have purchased for #26 preamp, right?

Summarizing what are the necessary best spec for HD600:
-application SE
-ratio
-primary impedance in ohms
-secondary impedance in ohms
-primary inductance in Henry
-power in watts at 50Hz
-DC mA
-core

For sowter and senns you can specify a transformer that could meet both preamp and headphone with one secondary winding. They seem one of the few transformer companies I have corrosponded with who understand headphones. You can fill out their request form and be generic, let them know you want minimal primary loss (small core) and minimum secondary dcr/inductance. I would go with a 7k:300. Then they will work with you from there. Bud at Onetics is also good.




For headphones, you need to figure out how much voltage swing you want and see wether a custom transformer is necessary. If you need less than 12Vrms, then a standard speaker transformer is the way to go. Some have 16 ohm taps.

Part of the tube sound comes from the transformer. It fattens up the waveform slightly, which gives warmth and soundstage expansion (very desireable for headphones), while being much quicker-sounding than a big electrolytic cap. I would describe a transformer as transparent but slightly colored, while capacitors are neutral but not transparent. For preamps and headphone amps, I much prefer transformer-coupled.

A 16 ohm tap with senns (300 ohm) would be a real compromise, remember your load is reflected to the secondary so you end up with a crazy damping factor, and transformer quality is inversely related to winding ratio, so you can do much better than running a 300 ohm headphone on a 16 om tap. Not only that but you get into ringing and all sorts of non-linearity with a mismatched impedance.

Don't forget that the less voltage the output tube swings the less distortion, another argument for a smaller step-down ratio. Headphones really reveal when the tube is swinging into the non-linear portion of the load-line (where the tube curves are compressed.)

I agree fully that a transformer coupled SET sounds better than a 470uf electrolytic cap couped OTL, especially with digital, there is "something" about a transformer that just fits. Good headphones can be incredibly revealing and down-right superior to speakers if you are into listening to soloist midrange like female vocals, strings, and brass, there is nothing lke it, it is very rewarding.

Off topic but I did have good luck with a fuuterman OTL using big 100uf oil can caps for 300 ohm senns but I always missed that SET sound (too much stuff in the single path.) And I'm not I senn guy.

My conudrum with the 4p1l is I basically have two headphones I like: Grados (32 ohms, need only 30 mW of power and require the most absolute quiet noise level) and HifiMan HE5LE's that are 38 ohms but need 1W to 2W of power. The 4P1LE doesn't quite operate at a high enough voltage to deliver 2W peaks into 38 ohms. I am a big fan of NOS russian tubes, a 4p1l-2c4c (or 6c4c) is what I am leaning toward. I regulator like the SSHV is almost a requirement for a black background with the Grados, as is a coleman for the filaments.

One other gotcha that is same for preamps is the power transformers have to be 3 ft away from the output transformers else you get mains hum pickup.

I still haven't pictured how one can do that and have coleman filament regs, almost need three chassis which is sort of a grounding nigtmare. But this is fun stuff:p
 
Administrator
Joined 2004
Paid Member
The 4P1LE doesn't quite operate at a high enough voltage to deliver 2W peaks into 38 ohms.
It does if the transformer has the right ratio, which is (I suppose) what you meant. The voltage out of the 4P1L is going to be stepped down thru the transformer anyway. Might as well step it down to the ratio you need, right?

Is damping factor really that important with most headphones? Maybe with the ear speaker type, but with more normal headphones?
 
It does if the transformer has the right ratio, which is (I suppose) what you meant. The voltage out of the 4P1L is going to be stepped down thru the transformer anyway. Might as well step it down to the ratio you need, right?

Well the Ra of the 4p1L isn't particularly low, so you need quite a bit of inductance, and you need to step down to 38 ohms. So you are looking at a 5k:38 OPT. So 11.5:1 ratio. Two watts into 38 ohms is 288 Vp-p across the primary, no way the 4P1L can swing that without A1/A2 and hitting the ugly portion of the curves, imho anyway you look at it its not a big enough "upgrade" to my 6c4pi/5842 spud amp.

Is damping factor really that important with most headphones? Maybe with the ear speaker type, but with more normal headphones

damping factor effect depends on the headphone driver design. Most modern headphone drivers have been "tuned" to modern SS <2 ohm opamp output impedance.

Simple tests which have been done adding resistors in series have been shown to change the frequency reponse. Not saying its black and white, but damping factor is an issue in how the transducer reponds to power.

The biggest issue as a tube guy is all the signal loss with a high DCR windings, its just that much harder the output tube has to "work", more distortion. And headphones reveal it.
 
Administrator
Joined 2004
Paid Member
So 11.5:1 ratio. Two watts into 38 ohms is 288 Vp-p across the primary
OK, but I see only ~9V needed for 2W peak into 38 ohms. Or 206V P-P on the primary. That just details. Whatever the output impedance of the OPT, 2W is going to be pushing the 4P1L, agreed. Wavebourn seems to indicate it's OK up there, but I've not pushed it that far yet.

Honestly, I don't want headphones that need two watts to function. My speakers hardly need that! :) If I did like them, I guess I'd do like you and go for a more powerful amp. My OPT has a 600 ohm winding, might be interesting to hear what that sounds like on 300 ohm cans.
 
no way the 4P1L can swing that without A1/A2 and hitting the ugly portion of the curves, imho anyway you look at it its not a big enough "upgrade" to my 6c4pi/5842 spud amp.

Huh? :eek: Which ugly portion of the curves? :eek: I'm asking once more: did you ever dare to read the topic from the beginning?

I started this topic to share with you how I got 2.5 W of pristine clean power from one 4P1L tube, and 5W from 2 in parallel, why start again from common musings about common tubes according to common beliefs?

Nobody needs 300 Ohm secondary to drive 300 Ohm headphones that need milliwatts of power. But to drive 50 Ohm 2 W you indeed need 50 Ohm secondary. To drive 32 Ohm 2 W you need exactly 32 Ohm 2W secondary. To drive 5W 50 Ohm headphones you need 2 tubes in parallel and 50 Ohm secondary transformer.

By the way, I found a company in Russia that will make me 32 and 50 Ohm 2 and 5W transformers for headphones. Stay tuned. :cool:
 
I don't see here any "ugly portion of curves". Up to +16V in class AB2 they look nice. As I said, in AB2 driving up to +12V I was getting 2.5V from the tube. In A1 only output power can be obtained around 1.8 W from tube. Why I made the A2 capable, because it's hard not to hit positive voltage on control grid of the output stage on peaks, that means non-linear grid current, rectification of signal, and dynamic change of bias point.

Here again, triode curves that show how nice is the tube when driven in class A2 mode:

4p1l-triode.gif
 
Member
Joined 2009
Paid Member
Huh? :eek: Which ugly portion of the curves? :eek: I'm asking once more: did you ever dare to read the topic from the beginning?

I started this topic to share with you how I got 2.5 W of pristine clean power from one 4P1L tube, and 5W from 2 in parallel, why start again from common musings about common tubes according to common beliefs?

Nobody needs 300 Ohm secondary to drive 300 Ohm headphones that need milliwatts of power. But to drive 50 Ohm 2 W you indeed need 50 Ohm secondary. To drive 32 Ohm 2 W you need exactly 32 Ohm 2W secondary. To drive 5W 50 Ohm headphones you need 2 tubes in parallel and 50 Ohm secondary transformer.

By the way, I found a company in Russia that will make me 32 and 50 Ohm 2 and 5W transformers for headphones. Stay tuned. :cool:

The Russian company will make 300 ohms transformers?
 
Huh? :eek: Which ugly portion of the curves? :eek: I'm asking once more: did you ever dare to read the topic from the beginning?



By the way, I found a company in Russia that will make me 32 and 50 Ohm 2 and 5W transformers for headphones. Stay tuned. :cool:

Ugly is a harsh word, its just that with orthodynamic headphone they have much lower itrinsic THD than speakers so you hear distortion much easier. The 4P1L is extremly linear over all. I just can't seem to explain that its Ra is not low enough to drive an orthodynamic driver 25Vp-p, but thats OK its an unusual situation.

If you find a transformer winder please do share, I too am interested in an all russian tube amp.
 
200:ish would be my choice as I have a pair of DT 931. Heres a little infosite on headphones. Pretty nifty. Meier Audio

Staffan


Great summary of what I was trying to comnumicate, now imagine headphones with specs like 32 ohms 98 dB/mw, or 600 ohms 94 dB/mW or 50 ohms 87 db/mW all tuned for different output impedance, all "liars" on their specs, and all offering lower distortion greater transperancy than the most expensive hand crafted speaker.

Then you see that you really need to "customize" your 4P1LE SE OPT to match your headphone of choice.

With the 4P1LE output it would power 90% of the all the high end headphones with the right output transformer. But its lower voltage rules out some orthos like the HE6 with its super strength rare earth magnet,AKG K1000's, and it rules out electrostatic headphones of course.

Senn and Beyer guys are lucky because those headphones match well with preamp configurations.

But as a positive the 4P1LE's lower plate voltage offers the opportunity to parafeed, imagine a nickle permalloy OPT for headphones, I would love to hear such a thing.

There is also the MGM engineered peerless tl404 autoformer, probably the only output iron really designed for headphones. But if anyone finds a pair I get first dib:D


Amonst all this discussion about impedance matching we mustn't forget that sensitive headphones must have a black background, no 60/120/180hz hum or buzz at all, my rule of thumb for Grados is that stuff has to be -90dB down.