Simplistic NJFET RIAA

[RCruz]

Originally Posted by merlin el mago

Ricardo,

Please note in bolt:



Yes these are mine real values, unfortunately I trusted in Pcx because I have ordered & paid matched pairs.

So it's not necessary 68pF to reach the 16nF for C3, I will remove in both channels?

Tomorrow I will let you know exact values of R1.

Not I haven't got more resistors I have to order more, please be sure with values as you know this kind of resistors are expensives, thks.

I am sorry I can not understand what the bold values mean.

If you want I can calculate the "best" values for each channel but I need the 3k3 values.

You can leave the 68p for now because my calc indicate aprox 16.0n for each channel as is

Look at the simulator results based on your previous indications.
The green cells indicate the best value, above are aproximations I gave you before.

[merlin el mago]

Stg

[RCruz]

Originally Posted by merlin el mago

Ricardo,

I have a couple of Shinkohs 999K 2W, also I have a couple of Caddocks MK132 one 1,002Meg & the other 1,001Meg, wich of both will be better for R5?

I would choose the higher valued ones... Let me know your final choice.

[merlin el mago]

Originally Posted by RCruz

I am sorry I can not understand what the bold values mean.

If you want I can calculate the "best" values for each channel but I need the 3k3 values.

You can leave the 68p for now because my calc indicate aprox 16.0n for each channel as is

Look at the simulator results based on your previous indications.
The green cells indicate the best value, above are aproximations I gave you before.

The bold values are mine right values.

Tomorrow I measure mine special R1 trimmers & let you know the exact value.

Ok I leave the 68pF in both channels.

I can't understand the simulator values, only understands green best values but see only Left channel in both simulations?

[RCruz]

Originally Posted by merlin el mago

The bold values are mine right values.

Tomorrow I measure mine special R1 trimmers & let you know the exact value.

Ok I leave the 68pF in both channels.

I can't understand the simulator values, only understands green best values but see only Left channel in both simulations?

Sorry, the right table should read Merlin Right.

I see that your made new measurements so I must restart again.

Please send me the following values Left and Right:

R1 3k3
R3 47k
R4 6k8
R5 1M

I will use your previous info:
C2 46.3n both channels
C3 16n both channels

[Salas]

My guess is that he is over compensating with R1 trimmer for bias voltages and mismatch for the burned duet. Which is normal for when he was seeking bias to make it initially just work. If he can set them for 3.3k sharp, and the bias voltages will be still in acceptable bracket, then you will have a reference point. I also run a Lipshitz inverse feeder in his circuit. The main arbitrary points are R1 and C3. With his now bold typeface deviations, it seems that a basic C3 without the extras, can lift it a bit, because it looks indeed dropping enough, especially if he had overshoot 3k3.

[RCruz]

Ok

With the new value of R4 right, the picture looks much better now:

My suggestions are:
Left channel place a 1M4 over R3 and a 470p over C2
Right channel place a 1M9 over R3 and a 450p over C2

LEFT
R1 3k3
R3 47k // 1M4
R4 6k8
R5 1M13
C3 15n6 + 330p + 68p
C2 46.3n // 470p

RIGHT
R1 3k3
R3 46k8 // 1M9
R4 6k79
R5 998k
C3 15n6 + 330p + 68p
C2 46.4 // 450p

[RCruz]

Off course, just lifting the 330p and 68p from C3, will lift everything producing imediate high definition but IMO, bypassing R3 with the big resistors will add so much more presence in bass and highs that C3 should be kept as is 16n.

After the big resistors, bass can become overwelming so I would increase C2 by 470p to bring back Low mids and snap.

[Salas]

Fixed 3k3 is key for any attempt non the less.

[RCruz]

Yes, because it has other functions than be just part of the passive filter.