Simplistic NJFET RIAA

By the way, on another more general FSP topic now, there was discussion by me a couple of times in the past about what actual noise penalty an active CCS would have brought vs the standard R13 ...

The Mosfet source controlled the TP1-TP2 voltage drift best but gave 6-7dB rise on the noise floor.
...

Anyway I would have possibly accepted up to 3dB more hiss

I meant to reply to this before. Thanks for finally performing the experiment, I have wondered about it from time to time but never done anything about it. Too bad about the result, like you said I would have considered 3dB or so maybe OK, but not 6dB. Still, a noble effort, and thanks again for checking!
 
diyAudio Chief Moderator
Joined 2002
Paid Member
I meant to reply to this before. Thanks for finally performing the experiment, I have wondered about it from time to time but never done anything about it. Too bad about the result, like you said I would have considered 3dB or so maybe OK, but not 6dB. Still, a noble effort, and thanks again for checking!

Although we had already discussed back then that current injection would automatically bring a current noise contribution we still wanted to see the practical result in dB and don't be satisfied from just a simulated answer. Inquiring minds must know, and as I had a chance on a handy board... I remembered :)
 
D1x no

Red is usually safer for lower noise. But green and yellow can be good for that also. Blue and white are surely more noisy.

Very interesting. Have any point to use branded LEDs other than the ones they sell in local shops and more likely are bulk chinese or unkown?
If noise matters in the power supply, how about the quartet D1-D4? Are those more critical to the level of noise?
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Very interesting. Have any point to use branded LEDs other than the ones they sell in local shops and more likely are bulk chinese or unkown?
If noise matters in the power supply, how about the quartet D1-D4? Are those more critical to the level of noise?

When not having dedicated noise measurement results there is no way to tell that branded red in general will be quieter than generic.

D1-D4 quartet in the phono circuit are the voltage bias local supply for the common base cascode's top part and its more critical for noise that is why it uses a low noise JFET CCS (Q7) and its also heavily filtered by C8. D1x in the shunt reg can also contribute a little but the whole Vref is heavily filtered again by C1x.
 
Thanks Nick
I 'm going to use 2 pairs of 369s for LOMC (I 'm going to get a Denon DL-301 MKII)
and I have four of them from GB. Two are 10.59 IDSS and two with 10.71 IDSS.
How it's better to install them? one 10.59 paired with a 10.71 to Right channel or both 10.59 to, let say, Right and the 10.71 to Left?
cheers
 
Thanks Nick
I guess with 369s @10.59 & 10.71mA I will go with a 3.9 R2 & R3
What suggestion you have for loading the 301 MKII? I red about 300-1K is optimal
In any case do we need R1 at 47K and parallel it with a i.e 400-700 ohm resistor or I can go straight with a small dedicated R1 for loading? (I'm not going to use the multiswitch)

Thanks again
 
diyAudio Chief Moderator
Joined 2002
Paid Member
Yes 3.9 Ohm

You could put 330, 470, 560, 680 on the four way DIP switch and R1 47k at first because it takes time to set a TT for a certain cart to decide in conjunction with the loading. In the end you can remove the DIP or set all positions to OFF and replace R1 with the resistor you listened longer best with. You can also make combinations for intermediate values when having the switch by putting more than one position to ON. Measuring across the input RCA with the DMM (when the phono is powered down and the TT cable is disconnected) you don't need make calculations of what combination load the cart sees.
 
thanks again Nick
My 117GR are kind of the weak side mostly between 3-4ma so I guess this going with R3x of 9K1 R6x of 22R
I have very few (4-5) of them at 4.5-5mA, I there any need to use those instead of the weak ones and change resistors to 6K2 and 33R respectively?
Do the Q2x Q3x and Q5x have to be close in IDSS?
 
diyAudio Chief Moderator
Joined 2002
Paid Member
No need to have the highest Idss samples, you just have to have the right resistors for any available Idss so to reach a usable voltage range.

Q2x Q3x Q5x if having some about close better pair them per position in each channel's reg. They don't need be close Idss across the board.
 
Has anyone used a Ortofon Cadenza Black with the FSP yet? I will be helping a friend assemble his soon and was wondering if there was any cartridge loading details I could learn. Cart details below:

Output voltage at 1000 Hz, 5cm/sec. - 0,33 mV
Compliance, dynamic, lateral - 16 µm/mN
Tracking force range - 2,0-2,5 g (20-25 mN)
Tracking force, recommended - 2,3 g (23 mN)
Internal impedance, DC resistance - 5 Ohm
Recommended load impedance - > 10 Ohm
Cartridge body material - Stainless steel Aluminium

Thanks,

KC