I have five different flea powered ( 1 to 5 Watts) tube amps I’d like to try with some groovy new 27 ohm headphones I’ve bought - can I make an intermediary device to allow me to listen to headphone off the speaker terminals of these flea powered tube amps?
Yes, its easy. Just two resistors.
The value is up to you to optimize.
I have an old magnavox 2watts el84 amp restored, used as headphone amp. I use 8 ohm plus 1 ohm reduction resistors. Amp will see 9 ohms load all the time, 1/9th goes to headphones.
The value is up to you to optimize.
I have an old magnavox 2watts el84 amp restored, used as headphone amp. I use 8 ohm plus 1 ohm reduction resistors. Amp will see 9 ohms load all the time, 1/9th goes to headphones.
As you already know, headphones are generally very sensitive.
The attenuation of 8 and 1 is 19dB (-19dB).
Hopefully the amplifier hum will not be heard; but you will hear some hum on some amplifiers that were made for loudspeakers.
Real headphone amplifiers should not have any hum.
If they do have hum, shame on the designer.
Just my opinions
The attenuation of 8 and 1 is 19dB (-19dB).
Hopefully the amplifier hum will not be heard; but you will hear some hum on some amplifiers that were made for loudspeakers.
Real headphone amplifiers should not have any hum.
If they do have hum, shame on the designer.
Just my opinions
ElArte,
Most of my vacuum tube amplifiers that I designed and built, have less than 100uVrms hum into an 8 Ohm non-inductive load resistor.
I only use the amplifiers to drive loudspeakers, not headphones.
And . . .
I do not use regulated filaments, I do not use regulated B+, and I do not use global negative feedback.
The types of negative feedback I sometimes use, are only local negative feedback, and are limited to:
Ultra Linear; Plate to Screen of Triode wired Pentodes/ Beam Power tubes; cathode to cathode Long Tailed Pair phase splitters; and push pull output stage's plate to plate.
Some of you never thought that a push pull output stage has plate to plate negative feedback, it does.
But, even my low hum amplifiers might have enough hum to be heard in many headphones, unless you use some attenuation from the output to the headphones.
Enjoy the Music
Merry Christmas!
Most of my vacuum tube amplifiers that I designed and built, have less than 100uVrms hum into an 8 Ohm non-inductive load resistor.
I only use the amplifiers to drive loudspeakers, not headphones.
And . . .
I do not use regulated filaments, I do not use regulated B+, and I do not use global negative feedback.
The types of negative feedback I sometimes use, are only local negative feedback, and are limited to:
Ultra Linear; Plate to Screen of Triode wired Pentodes/ Beam Power tubes; cathode to cathode Long Tailed Pair phase splitters; and push pull output stage's plate to plate.
Some of you never thought that a push pull output stage has plate to plate negative feedback, it does.
But, even my low hum amplifiers might have enough hum to be heard in many headphones, unless you use some attenuation from the output to the headphones.
Enjoy the Music
Merry Christmas!
Last edited:
For a few years I used a 6V6 single end amp for headphones. I loved it. The only problem was a bit of buzz and hiss in the background. To be fair, it was not made for headphones and it as pleasantly quiet on speakers.
Headphones are an easy load to drive, but super revealing of noise.
Headphones are an easy load to drive, but super revealing of noise.
That’s exactly why I was asking! What a tough business! I got into headphones during COVID (turns out that indulgent speaker use is problematic when everyone is home 24/7, who knew?), they can reproduce levels of detail that are sacrificed to room acoustics with speakers.
BUT, it’s a double-edged sword! Unruly electrons are fully exposed. My tinkering with attenuation schemes has given mixed results.
Check out this LPad calculator for just this purpose. The LPad loads the transformer with the parallel resistor and attenuates with the series one (Im guessing). So your volume control has a realistic range, and the transformer sees the right load.
https://robrobinette.com/HeadphoneResistorNetworkCalculator.htm
https://robrobinette.com/HeadphoneResistorNetworkCalculator.htm
Last edited:
Please keep in mind that any "output" impedance placed between the amp and the headphones will cause variations in their frequency response.
Those changes can be minimal or severe depending on the impedance of the headphones and the values of resistors used.
It is better not to add any resistance, but if it is necessary the value of the resistors used for the attenuation should be the lowest possible the amp can handle.
Those changes can be minimal or severe depending on the impedance of the headphones and the values of resistors used.
It is better not to add any resistance, but if it is necessary the value of the resistors used for the attenuation should be the lowest possible the amp can handle.
All the headphones I have ever owned had a very constant impedance across the audio band.
I doubt that very many headphones that are driven by a high impedance need much damping.
They generally are not like a woofer in a closed box, or in a ported enclosure.
And, a 32 Ohm phone that has almost 32 Ohm DCR, does not need 3.2 Ohms to drive it. The 32 Ohm DCR is generally the major factor of electrical damping.
I doubt that very many headphones that are driven by a high impedance need much damping.
They generally are not like a woofer in a closed box, or in a ported enclosure.
And, a 32 Ohm phone that has almost 32 Ohm DCR, does not need 3.2 Ohms to drive it. The 32 Ohm DCR is generally the major factor of electrical damping.
The Sennheiser "300 ohm" large diaphragm phones can have a big bump.
https://www.audiosciencereview.com/forum/index.php?threads/sennheiser-hd650-review-headphone.18774/
It's a bit all over the place depending on the technology and model...planars usually have constant impedance across the frequency range. Dynamics have oddities.
Here are a small selection of well-regarded headphones:
There is a bit of a pattern in these previous 3.
But then...
So, it varies, both in quantity and flatness...
I wish I understood better the implications, but I have not gotten there yet on my journey.
Here are a small selection of well-regarded headphones:
There is a bit of a pattern in these previous 3.
But then...
So, it varies, both in quantity and flatness...
I wish I understood better the implications, but I have not gotten there yet on my journey.
If the tube amp works as current source, there will be increased signal in low frequencies around fs. This can be very beneficial, sort of like loudness pot. But no tube amp is pure current source, its some combination of voltage source (indifferent to impedance changes) and current source.
Its all good.
Its all good.
PRR, ElArte, and adason,
I was wrong about the flat impedance curves of headphones.
So, I took a look at your Sennheiser 300 Ohm model.
The Sennheiser varies from about 310 to 480 Ohms.
About 395 +/- 85 Ohms; (395 +/- about 22%).
That is like an "8 Ohm loudspeaker" varying from about 6.3 Ohms to 9.7 Ohms.
Wow, pretty good!
I wish all loudspeakers had impedance curves that flat.
The worst headphone curves posted in this thread were about:
One at 4:1 impedance, and two at 2:1 Impedance.
Lots of loudspeakers easily exceed 4:1 impedance.
Perhaps we need to worry more about vacuum tube amplifiers that drive loudspeakers;
and worry a little less about amplifiers that drive many models of headphones.
Perhaps a fixed T attenuator, or a fixed Pi attenuator could be used to work with at least one model of headphone that has a wide impedance variation (A low impedance amplifier output, and a low impedance fixed T or Pi attenuator should flatten things quite a bit).
Now, back to driving those pesky loudspeakers!
And, I enjoy driving and listening to many different loudspeaker models.
Impedance . . . What would you like it to be?
Amplifiers can be voltage sources, current sources, or resistive signal sources.
Thevenin or Norton, Amplifiers always have some resistance, no matter how small or how big.
I think of many vacuum tube amplifiers with moderate damping factor as being a resistive signal source.
Just my opinions
I was wrong about the flat impedance curves of headphones.
So, I took a look at your Sennheiser 300 Ohm model.
The Sennheiser varies from about 310 to 480 Ohms.
About 395 +/- 85 Ohms; (395 +/- about 22%).
That is like an "8 Ohm loudspeaker" varying from about 6.3 Ohms to 9.7 Ohms.
Wow, pretty good!
I wish all loudspeakers had impedance curves that flat.
The worst headphone curves posted in this thread were about:
One at 4:1 impedance, and two at 2:1 Impedance.
Lots of loudspeakers easily exceed 4:1 impedance.
Perhaps we need to worry more about vacuum tube amplifiers that drive loudspeakers;
and worry a little less about amplifiers that drive many models of headphones.
Perhaps a fixed T attenuator, or a fixed Pi attenuator could be used to work with at least one model of headphone that has a wide impedance variation (A low impedance amplifier output, and a low impedance fixed T or Pi attenuator should flatten things quite a bit).
Now, back to driving those pesky loudspeakers!
And, I enjoy driving and listening to many different loudspeaker models.
Impedance . . . What would you like it to be?
Amplifiers can be voltage sources, current sources, or resistive signal sources.
Thevenin or Norton, Amplifiers always have some resistance, no matter how small or how big.
I think of many vacuum tube amplifiers with moderate damping factor as being a resistive signal source.
Just my opinions
Last edited:
The reason for the lower source impedance I mentioned is to minimize the variations of the driving voltage, not damping of the headphones.
Years ago I did some headphones measurements for the company I was working and I found wide variations in impedance, our peers gave some examples above showing it.
At the risk of causing much consternation, I wonder if a simple opamp-based unity gain buffer board would be the most practical approach to opening most amps to any headphones.
But, I do not know what I am talking about here (not jesting either; I really don’t know), so it may not be as simple as that, or it may not be desired or necessary at all.
But, I do not know what I am talking about here (not jesting either; I really don’t know), so it may not be as simple as that, or it may not be desired or necessary at all.
You may find a suitable opamp for a particular high impedance headphone; but in general the maximum output current available from opamps will be a limiter factor.
The standard solution is adding a pair of "buffer" output transistor to the output of the opamp and you are done.
Then comes the details of your design: will you bias this stage in class A?, you can even force the opamp to work in class A and the distortion will be very low.
A Google search will show you it, look for "biasing opamps into class A" and also "buffering opamps".
The standard solution is adding a pair of "buffer" output transistor to the output of the opamp and you are done.
Then comes the details of your design: will you bias this stage in class A?, you can even force the opamp to work in class A and the distortion will be very low.
A Google search will show you it, look for "biasing opamps into class A" and also "buffering opamps".
I use my 45 SET (AC filament) to drive HD650 from time to time. No audible noise from headphone at all when no music is playing. I use a 10ohm(as I don’t have 8ohm resistors on hand) in parallel with each driver.
I just measured a couple of my amps... They average 5mV of "noise" driven by the preamp on mute. If I find my shorting plugs, I'll see if it's lower with no input.
I have done the same thing with my 2A3 TSE-II. I wired a 10 ohm resistor across the speaker terminals and connected the Sennheiser's across that resistor. With about 3 or 4 watts per channel on tap you do have to be really careful with inputs and the volume knob. There is no noticeable hum or hiss, but the TSE-II runs regulated DC on the heaters.