diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Headphone Systems (http://www.diyaudio.com/forums/headphone-systems/)
-   -   LED biased EL84 headphone amp (http://www.diyaudio.com/forums/headphone-systems/48438-led-biased-el84-headphone-amp.html)

baggystevo82 26th December 2004 05:13 PM

LED biased EL84 headphone amp
 
2 Attachment(s)
Hello all you wise people out there, I hope you can help me.
I've been tempted to buildthis circuit for a while now, but I've been inspired by SY's comments on LED biased output stages. Well this is what I have so far for the output stage, and I have a few questions. (please bear in mind I'm still fairly new to designing my own circuits)
How is the value of the grid input resistor derived, or if it is needed at all?
Same for the resistor between g2 and the anode on Helmut's design.
How is the value of the resistor accross the output terminals derived?

Thanks for any help,
Steve

baggystevo82 26th December 2004 09:03 PM

someone out there must know? Or have I missed something stupidly blatantly obvious?
Steve

G 26th December 2004 09:07 PM

Quote:

Originally posted by baggystevo82
someone out there must know? Or have I missed something stupidly blatantly obvious?
Steve

What bias voltage are you trying for on the cathode?

baggystevo82 26th December 2004 09:10 PM

6.8v
or 4 x 1.7, that's another thing actually, my christmas effected head can't quite get round. As the voltage dropped accross the LED's isn't a product of ohms law, they just have a constant drop anyway (relatively), does that mean that the layout shown in that schematic will have the same drop but with double the current handling of 4 LED's in series on their own?
Cheers,
Steve

audiousername 27th December 2004 10:06 AM

Quote:

Originally posted by baggystevo82
does that mean that the layout shown in that schematic will have the same drop but with double the current handling of 4 LED's in series on their own?
Yep, 'think so. But I've got a 'Christmas-affected brain' too.

baggystevo82 27th December 2004 10:50 AM

Ahh that's good then. I'm about to have my first cup of tea in 4 days :) the recovery starts here, until new years...
So any ideas or clues about the resistor values?
Cheers,
Steve

SY 27th December 2004 11:12 AM

You're right about the LEDs- just think of them as a battery. Two strings in parallel double the current rating and halve the impedance. Figure that you'll end up with about 10 ohms- not bad!

Make sure that the 1K resistor has a nice, high current rating and you'll be good to go. A 470K value for the input resistor is probably OK, though with the bootstrapping inherent in your CF, you could drop that to 100K without harm. The grid stopper should be somewhere between 1K and 10K, with the body of the resistor as close to the tube pin as possible.

baggystevo82 27th December 2004 11:35 AM

Cheers SY. I've just done a bit of searching around and lots of people seem to recommend 1k grid stoppers on EL84's so I think I'll go with that. Any ideas why Helmut's uses only 100ohms here? I'm guessing I should include the 100ohm g2-plate stopper resistor too (is this the right term?), or should I leave it out and only include it if there's oscillation problems?

That resistor accross the output is still a problem for me though. Is it just there to keep the output tied to ground if no headphones are plugged in?

Oh sorry one more, if I drop the value of the grid leak resistor (I think that's its purpose here?) what value of R should I use to calculate the frequency response of the input coupling?

Cheers again!
Steve

SY 27th December 2004 11:49 AM

Quote:

Originally posted by baggystevo82
Any ideas why Helmut's uses only 100ohms here? I'm guessing I should include the 100ohm g2-plate stopper resistor too (is this the right term?), or should I leave it out and only include it if there's oscillation problems?
The resistor's size (when optimized) will vary from application to application. But really, it's pretty non-critical- if you use 4.7K, for example, you'll lose a tiny bit of bandwidth, which is critical only if you're a dolphin. But you'll be nearly guaranteed that the tube won't oscillate due to grid factors. Likewise, the g2 stopper- it may work fine without it, but 100 ohms in series will just eliminate the worry.

Quote:

That resistor across the output is still a problem for me though. Is it just there to keep the output tied to ground if no headphones are plugged in?
Yes.

Quote:

Oh sorry one more, if I drop the value of the grid leak resistor (I think that's its purpose here?) what value of R should I use to calculate the frequency response of the input coupling?

The grid leak is bootstrapped by the cathode resistor. So its effective value is multiplied by the feedback factor. What's the mu of a triode strapped EL84?

baggystevo82 27th December 2004 12:21 PM

Cheers for all the help!!

Ive found some UV LED's that have a drop of 3.2v, not sure what 'resistance' they are though.

I'm just looking for its mu now, but the mu of a 'normal' el84 is 19 I think.
Thanks,
Steve


All times are GMT. The time now is 10:30 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2