M.Leach jfet broadcast-limiter

[richiyobs]

hi folks,
this time i m playing with Marshall Leach Paper about broadcast limiter
attached LTspice sim only for the limiter part... it s working great!
tried to search something more about this design but nothing found on how to implement a different design for recovery....
Varying R18 can slow the attack, instead varying R21 since it also changes U6 gain i get strange behavior
in my poor experience never saw a timing net like that.
also noticed that in Q2 and Q3 rising R19 and R20 i get a sort of log response

also added a meter drive as per RM57 style but not so linear, just an attempt

thanx for your attention

 

[zamproject]

Not sure about the release time constant ? R19/C2 ?
What about changing C2 value ? (but this will be stepped setting)
Or R21 as a pot but with C2 at the wiper ?

 

[richiyobs]

Not sure about the release time constant ? R19/C2 ?
What about changing C2 value ? (but this will be stepped setting)
Or R21 as a pot but with C2 at the wiper ?

My first approach was to switch C2... According to the Paper release is R21/C2

 

[richiyobs]

R19 is part of the current mirror as R20... Advised 100ohm

 

[richiyobs]

Lowering R21 I get of course a different compression... So believe the recovery time is fastening due to the shape of wave out that looks more similar to the input one except the first transient...
Even if it alters the gain of U6 the Cv at jfet gate remains quite constant.
Also added a cap across the meter and its response smoothed...
Is there a way to visualize Timing on Ltspice?
Have to use slewrate measurement?
Any suggestions please?
Best

 

[richiyobs]

attached a comparison of R21 values
i m very confused, can someone please clarify about this behavior?
also added different caps for test... hope to get good calculations with 75k for R21
thanks for your attention
best