I put a lot of time (and money) on my folded headphone amp project, but it didn't turn out as I wanted. The sound is just a bit too polite and unexciting. So I've been pi**ed of with amps for a while.
I thought I could use the board for something else, and since I've read all the raving about the JLH, I wanted to give it a try. With just some jumpers, it fit perfectly on the board, mirror imaged of course since I used N-channel input in the amp mentioned above. I choose JFET input because I wanted it DC-coupled, but since JFETs are so temperature sensitive there was over 100 mV difference in offset going from cold to warm. So I had to use a servo and input capacitors (WIMA MKS). I use an AD823 as servo, because that's what I had lying around. The virtual ground is a discrete rail splitter, and 3300 uF Sanyo WX capacitors.
Everything seems stable. The noise level is very low. The input FET runs at about 2 mA, the second stage at 4 mA and the output stage at 20 mA. It's housed in a Hammond 1455J1201, powered by 8 AAA's with a LM317T used as a trickle charger and a L78S10CV regulator when using wall power.
Now I understand the raving. It has a very warm and forward mid, fat and friendly bass and fairly clean treble. I think it has a slightly grainy sound, "the JFET sound". Maybe I'll try it with a 2SC2240 as input transistor later on, but now I'm just enjoying music. It sounds like my le Classe A headamp clone when I used JFETs as input stage, but warmer, with a golden glow, and more pronounced mid.
Schematic and images can bee seen at this link
http://www.esnips.com/web/KindofJLHwithFETandservo/
Most components are on sockets for tweakability. If you have suggestions for improvement, please let me know.
I thought I could use the board for something else, and since I've read all the raving about the JLH, I wanted to give it a try. With just some jumpers, it fit perfectly on the board, mirror imaged of course since I used N-channel input in the amp mentioned above. I choose JFET input because I wanted it DC-coupled, but since JFETs are so temperature sensitive there was over 100 mV difference in offset going from cold to warm. So I had to use a servo and input capacitors (WIMA MKS). I use an AD823 as servo, because that's what I had lying around. The virtual ground is a discrete rail splitter, and 3300 uF Sanyo WX capacitors.
Everything seems stable. The noise level is very low. The input FET runs at about 2 mA, the second stage at 4 mA and the output stage at 20 mA. It's housed in a Hammond 1455J1201, powered by 8 AAA's with a LM317T used as a trickle charger and a L78S10CV regulator when using wall power.
Now I understand the raving. It has a very warm and forward mid, fat and friendly bass and fairly clean treble. I think it has a slightly grainy sound, "the JFET sound". Maybe I'll try it with a 2SC2240 as input transistor later on, but now I'm just enjoying music. It sounds like my le Classe A headamp clone when I used JFETs as input stage, but warmer, with a golden glow, and more pronounced mid.
Schematic and images can bee seen at this link
http://www.esnips.com/web/KindofJLHwithFETandservo/
Most components are on sockets for tweakability. If you have suggestions for improvement, please let me know.
It's battery powered. How can I improve the supply without losing too much voltage? I added those small resistors (2R7) to isolate the rails going to the first stages from the output transistors. I think it's called capacitance multiplier.Nordic said:I suspect the grainyness is from the inherent low noise rejection...
Spend some effort on the PSU next.
How would you describe the difference between bipolar and JFET transistors?
Or maybe it's the sound of the capacitors sinking the return current. I guess Sanyo WX isn't the best. Maybe some BlackGates?
Hi,
You can't compare. They are fundamentally different devices each with it's own advantages and disadvantages. At it's most simplistic level you can say the Bjt is "current driven" and the FET "voltage driven" but there is much more to it than that. It's how they are used that matters and whether they are used correctly.
nelsonvandal said:
How would you describe the difference between bipolar and JFET transistors?
You can't compare. They are fundamentally different devices each with it's own advantages and disadvantages. At it's most simplistic level you can say the Bjt is "current driven" and the FET "voltage driven" but there is much more to it than that. It's how they are used that matters and whether they are used correctly.
I guess you're right. The folded cascode amp using the same SK170 as diff pair, isn't the slightest grainy. I wonder if the reason for JLH being so easy to like is beacuse of coloration or the lack of it.Mooly said:Hi,
You can't compare. They are fundamentally different devices each with it's own advantages and disadvantages. At it's most simplistic level you can say the Bjt is "current driven" and the FET "voltage driven" but there is much more to it than that. It's how they are used that matters and whether they are used correctly.
I think,
after read so many sound reviews of JFET and Pass MOSFET based amps,
that when we let JFET / MOSFET be the important (input) stage
then we get another more pleasing sound.
At least to many ears.
Is probably one reason Pass amplifiers are so much enjoyed.
Another is of course TRUE CLASS A.
Which also give a very good sound.
after read so many sound reviews of JFET and Pass MOSFET based amps,
that when we let JFET / MOSFET be the important (input) stage
then we get another more pleasing sound.
At least to many ears.
Is probably one reason Pass amplifiers are so much enjoyed.
Another is of course TRUE CLASS A.
Which also give a very good sound.
nelson,
I am personally not surprised, it was predictable.I put a lot of time (and money) on my folded headphone amp project, but it didn't turn out as I wanted.
The sound of heavily abused JFETs, I would say.I think it has a slightly grainy sound, "the JFET sound"
I haven't really given it up yet. The main problem is the ground channel, and I've just tore it out. It sounds very smooth and overly clean with resemblance of AD797, AD829 and ADA4841. There's no sign of any erroneously behaviour in this amp. It's just that its a bit soft, but it's nice to have it when my ears are sore from listening to more lively amps. Is this the "folded cascode sound"? So many posts here, but not a word on how things sound! Don't you trust your ears?Lumba Ogir said:nelson,
I am personally not surprised, it was predictable.
Lumba, what was predictable? What could I have done different, or is it the topology or layout that's bad?
I'm going to try a bipolar input, but I'm out of 2SC2240 at the moment. While I'm waiting for my order, why don't you just tell me how a JFET should be used.The sound of heavily abused JFETs, I would say.
OK, so now I'm abusing a SC2240 as input transistor. I had to change all the resistor values to balance it. The grain is gone, the sound is brighter and maybe more detailed. Gone is also some weight and texture and it's less "euphonic". This is exactly the same differences I've experienced in my other amps when changing from JFET input to bipolar input. I can't really say one is better than the other, just different. It's nice to to have amps with different sound signatures.
Lumba Ogir, please tell me how to implement JFETs in a proper way in this amp.
Lumba Ogir, please tell me how to implement JFETs in a proper way in this amp.
Thank you Lumba, I'll see if I can implement this in my amp.Lumba Ogir said:Nelson,
please read post #51 in the other thread. I put it that way because it`s exactly what proper cascoding gives and JFETs need.
Regarding the JLH clone, I'm back to bashing JFETs. I like it better this way. It's a bit too forward and a little bit grainy, but lively and "musical". With bipolar input it had the same sound signature as my le Classe A clone and the latter is better.
nelsonvandal
one absolute advice:
Lumba Ogir may seem like he knows what is best.
And I ask, who does?
On the other hand I have not often seen such one abstract subjective guessings guy. Which is very good.
We want some reality - not random guess.
nelsonvandal,
go your own way, by your own experience & ideas.
your ideas are TOO GOOD to be ruined by some besserwisser.
Be it he is from my own country of ABBA Sweden or not.
Sorry, Lumba, no offense,
but if you post some circuits / amplifiers of your own + comments,
I might start thinking other of your capacity as one DIY Audio member.
Until you do, you are not much to listen to, I am afraid.
Lineup
Good Audio Regards to nelsonvandal & Lumba Ogir.
My website with the best possible amplifiers I can think of,
schematics and real amplifier constructions:
http://lineupaudio.freehostia.com/forum/index.php
one absolute advice:
Lumba Ogir may seem like he knows what is best.
And I ask, who does?
On the other hand I have not often seen such one abstract subjective guessings guy. Which is very good.
We want some reality - not random guess.
nelsonvandal,
go your own way, by your own experience & ideas.
your ideas are TOO GOOD to be ruined by some besserwisser.
Be it he is from my own country of ABBA Sweden or not.
Sorry, Lumba, no offense,
but if you post some circuits / amplifiers of your own + comments,
I might start thinking other of your capacity as one DIY Audio member.
Until you do, you are not much to listen to, I am afraid.
Lineup
Good Audio Regards to nelsonvandal & Lumba Ogir.
My website with the best possible amplifiers I can think of,
schematics and real amplifier constructions:
http://lineupaudio.freehostia.com/forum/index.php
Lineup,
I am not guessing. Documentation has been available as theoretical explanation and unanimous subjective opinion for many decades. Some people, like you, think it is possible to increase the efficiency and get away unpunished. Not quite, there is a price to pay.
I am not guessing. Documentation has been available as theoretical explanation and unanimous subjective opinion for many decades. Some people, like you, think it is possible to increase the efficiency and get away unpunished. Not quite, there is a price to pay.
No problem, you are very welcome any time...Sorry, Lumba, no offense
Nelson,
consider this output stage topology very thoroughly.
http://www.diyaudio.com/forums/attachment.php?s=&postid=1546505&stamp=1214276684
consider this output stage topology very thoroughly.
http://www.diyaudio.com/forums/attachment.php?s=&postid=1546505&stamp=1214276684
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Amplifiers
- Headphone Systems
- JLH headphone amp with FET's and servo