Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Twilight Tube-MOSFET mu-stage hybrid headphone amplifier
New Posts  All Forums:Forum Nav:

Twilight Tube-MOSFET mu-stage hybrid headphone amplifier - Page 3

post #31 of 35
2/10/09 at 1:01am
Thread Starter 
Progress!

Added 1 uF of capacitance to the 0.1 uF cap connecting the tube plate to the MOSFET gate. Frequency response is -0.5 dB at 20 Hz, unloaded.

However, I double checked at the bias resistor (R4 on the homepage diagram) is indeed 422K ohms. What could be going on?
LL
Reply
post #32 of 35
2/10/09 at 12:38pm
I have no 100% sure idea why, but I have experienced a similar thing with my Melos.

The gain stage is a 6DJ8 in a long tail pair, with the headphone out cap coupled off of the "in phase" leg. Anyways, the input of the headphone out is a single jfet in a source follower feeding a BIG mosfet source follower like this link, there is a DC servo and no output cap though. When I was first moding the amp, I tried to use a 0.1uf cap thinking that it would be acceptable, but bass was audibly rolled off from just below vocals. The stock value of 2uf maintained wide frequency range.

This link MAY be helpful, but im not 100% sure it or my explanation is right.
link
The last post is where I think it may mater. If the gate impedance of a mosfet drops as it becomes "less reverse biased" couldn't this happen as the audio signal hits it?
Reply
post #33 of 35
2/11/09 at 2:03am
Thread Starter 
Quote:
Originally Posted by nikongod View Post
...a source follower feeding a BIG mosfet source follower like this link...
I remember that one. Why did I make the input capacitance 1 uF I wonder? You know I can't remember the details. I wonder if I knew something about the input impedance of a JFET that I don't now, since it seems to me that the value is much higher than I would have needed to overcome just the 500 kohm bias resistor. [I'm getting this weird feeling of deja vu right now, like I've definitely been through this before.]

Anyway, fact of the matter is the power-off, DC resistance of the MOSFET in the Twilight is indeed near-infinite. The measured resistance to ground from the gate is about 470K, as expected for the 422K bias resistor and the resistor network it attaches to. When it's powered up an running though, the MOSFET gate appears to effectively be about 70K from the point of view of the audio signal.

Searching Google didn't give me any useful insights as to why this might be so. But it is so nonetheless.

Note that I actually prefer the sound with the smaller capacitor! To say more precisely: I feel the overall balance at -3dB at 20 Hz is actually a more correct presentation than the more linear response, which is a bit boomy.
Reply
post #34 of 35
2/11/09 at 2:31am
I honestly did not know that was you, COOL!
Your note on the power caps ringing in that article was one of the key things that got me thinking about whether to bypass electrolytic caps as opposed to blindly buying more and more film. Thanks

Holly crap! The op-amp phono stage I built was based on an article your website. WOW, Im in geek mode now. I built the passive EQ one.
Reply
post #35 of 35
3/31/09 at 5:48am
Have you tryed testing with a unbypassed cathode resistor? You may see less gain, but you may prefer the sound to having that cap there. I used to bypass mine, but since have removed them on my past amps.
Reply
New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › Twilight Tube-MOSFET mu-stage hybrid headphone amplifier