jacco vermeulen said:
ah.. good stuff in that thread
But I think I understand the basics.. so, if I lower the resistance from 22R to say 10R the gain will increase and so will the noise. But, will it be too much noise? Or, can I up R19 (499R) to 1K to get more gain... so, looks like I need to see what my voltage is compared to Wayne's (look like 15-16V is optimum here?), so I'll report back with this info. Also, since my JFETs are well-matched, I'd assume I'd get less noise by lowering the degeneration resistors..
Manu:
yes, I have both. I played with the bias due to that little LED (CCS) some and found a value I liked... then just listened and enjoyed!
or another possibility is that Wayne really used GR devices in his measurements, but put BL on the schematic? seems more likely that I'm confused, but anyways ... I remember denoting on my Ono that I got around 90mV across the 22R resistor if I used BL devices and half that if I used GR devices, hence the theory.
ahhh.. so, I figured it out: for the first stage 20(log(4*499R/(1/22mS + 22R)) = 29.4dB.. but the second stage, my numbers aren't quite adding up... should it be 20(log(2*1000/(1/22mS + 10R)) = 31dB? since each FET doesn't have its own resistor, I'm betting this isn't quite right ...Anyways, the above questions still remain, just wanted to share my epiphany
You do realise that a GR, a BL, and a V grade JFET are all and the same device and from the same production line ?
In a figurative manner of speaking, the mediocre monday samples are labelled GR, wednesday's specimens become Les BLeus and the top notch Friday run is nominated for the Venceremos elite title.
Toshiba doesn't measure each and every individual JFET to grade it, so their ranges overlap.
Means that a low performance BLouie is not just the equivalent of a hot rod GRoiner, it is exactly the same.
A JFET that operates in the GRouch zone is no good for a purpose that requires a V-rod, semiconductor manufacturers label them at the exit to make them universal/generic components.
Most of the consumer stuff is designed such that it doesn't matter whether the JFETs used are low or high performers in their class range.
The transconductance value in the datasheet is not a fixed value either, it varies, though with a much smaller deviation window.
To calculate the gain accurate to the last digit you'd have to measure your devices.
The correlation between drain current and source resistor of a JFET is a bit trickier than it is for regular FETs or BJTs.
You'll find plenty of info if you google for JFET biasing, the 102 application note from Siliconix is referenced in numerous articles.
Safe to say that an average 9mA Idss Blue JFET doesn't drop it's drain current to 3.2mA or less on a 22R degeneration resistor.
Your 4.8-5.5mA numbers for both left/right indicate that the batch of BL's in your Perla are High Rollers.
Yo, Ed, i was actually thinking of eating sushi with chopsticks from those platters while cursing in Japanese.
In a figurative manner of speaking, the mediocre monday samples are labelled GR, wednesday's specimens become Les BLeus and the top notch Friday run is nominated for the Venceremos elite title.
Toshiba doesn't measure each and every individual JFET to grade it, so their ranges overlap.
Means that a low performance BLouie is not just the equivalent of a hot rod GRoiner, it is exactly the same.
A JFET that operates in the GRouch zone is no good for a purpose that requires a V-rod, semiconductor manufacturers label them at the exit to make them universal/generic components.
Most of the consumer stuff is designed such that it doesn't matter whether the JFETs used are low or high performers in their class range.
The transconductance value in the datasheet is not a fixed value either, it varies, though with a much smaller deviation window.
To calculate the gain accurate to the last digit you'd have to measure your devices.
The correlation between drain current and source resistor of a JFET is a bit trickier than it is for regular FETs or BJTs.
You'll find plenty of info if you google for JFET biasing, the 102 application note from Siliconix is referenced in numerous articles.
Safe to say that an average 9mA Idss Blue JFET doesn't drop it's drain current to 3.2mA or less on a 22R degeneration resistor.
Your 4.8-5.5mA numbers for both left/right indicate that the batch of BL's in your Perla are High Rollers.
Yo, Ed, i was actually thinking of eating sushi with chopsticks from those platters while cursing in Japanese.
luvdunhill said:
Apres-Lunch jive :
40mV across 22R is ~1.8mA, 70mV makes ~3.2mA
Transconductance curve ;
=> 1.8mA Id reads 18mS aka ~55.5R.
=> 3.2mA Id smells 24.5mS aka ~ 41R.
1.8mA times (55.5 +22)/55.5 izza like 2.5mA, anda 3.2mA * (41+22)/41 is ~5mA.
So what'll be ?
Which parts and source resistor values would you pick for a phono stage if you were thinking of Willie Nyquist ?
(ZM's latest stuff does wonders for my bowl movement)
jacco vermeulen said:
umm.. "small Johnson" means big current, so 70mV w/ 22R is better than what was in the original schematic? so, I shouldn't look here to change gain, but instead the 499R resistor, maybe change to 1k and I should get 4 ish dB increase? I'm not sure if this will effect the RIAA curve, as it's mentioned in the Pearl text that this value is part of the RIAA, but from what I've learned seems like it isn't?
Just an interlude...
I was hearing today Brahms Requiem (Sinopoli/DG) 2. movement.
Not that you think I like Mr. B., Pan-Germania Pooftaah, but 2nd movement is special...
Anyway ... hearing it through Denon 103 (Thanks ZM, LL) and ONO is special experience.
I recall, I was at one of last Karajan (not that...) Concert in Musikverein in Vienna, Brahms (aimez-vous Brahmer?) same Req.
I had a Stehplatz ... for non German-speaking ... cheap ...
Anyway, that is special experience... owh ... believe me...owh, is good description of this .. because you're out of breath...
Ewig thanks Papa & Wayne
Manu
I was hearing today Brahms Requiem (Sinopoli/DG) 2. movement.
Not that you think I like Mr. B., Pan-Germania Pooftaah, but 2nd movement is special...
Anyway ... hearing it through Denon 103 (Thanks ZM, LL) and ONO is special experience.
I recall, I was at one of last Karajan (not that...) Concert in Musikverein in Vienna, Brahms (aimez-vous Brahmer?) same Req.
I had a Stehplatz ... for non German-speaking ... cheap ...
Anyway, that is special experience... owh ... believe me...owh, is good description of this .. because you're out of breath...
Ewig thanks Papa & Wayne
Manu
Manu said:
excellent piece. I really enjoy listening to the non-churchgoer write church music
I also really have enjoyed listening to Dvorak Symphony No.8 of late.. need to find a good vinyl recording though. I've been listening to both Iván Fischer and Jakov Kreizberg conduct.. they're ok I suppose. I tend to search for "Decca" + whatever performer / composer and it's usually good
jacco vermeulen said:
Your left/right channels are running on a little less/more than 20mA.
Mmm, how much voltage drop was 20mA times 1K again ?
When pushing comes to shovelling you'd have a little Johnson
now your loosing me. you think I don't have enough voltage to do this? I think I'd be willing to raise the gain a bit, the thing is dead quiet, just looking for the best way to do it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.