Anyone play with J.c. Maillet's triode emulator?

[WhiskeyMadeMeDoIt]

Has anyone tried this circuit as shown here on the bottom of this page http://www.lynx.bc.ca/~jc/transferCurvature-TubeSimulation.html?
I've been playing around with fetzer valves and decide to see what else was possible.
I am sure I have made a mistake somewhere but I can't get it to do anything.
Is there something missing from this basic circuit ? Any guidance would be appreciated.

[george]

check the pinout on the FET you are using - they can be different depending if it is 2N5457 or MPF102 or J201 that you are using.   Just google for the datasheet it will give you the pinout.

hth ...

[WhiskeyMadeMeDoIt]

Pinout is known and correct I am using 2n5457.  I have disabled the opamp feed to the source and subbed a resistor in place and it is working fine.  As soon as I hook up the opamp to the source it is dead as hell.  I metered the output of the opamp and it is working as expected.  Any suggestions?

[tca]

How are you biasing the opamp? Can you post the schematic?

[tubegeek]

How are you biasing the opamp? Can you post the schematic?

Well, that's kind of part of the issue. The schematic is a little unclear. The description is included below.

It looks to me like the op-amp is biased by a 100K trim pot (that's my interpretation of the squiggle at lower right) - the op-amp is being used as a stiff DC voltage source, purely to bias the JFET.



"If we are seeking to replicate the intrinsic curvature of a Triode circuit by way of using the similar Transfer curvature characteristic of a jFET device then introducing resistance in the Source circuit as an exclussive means is not the way to go. In other words, we seek a way to move the operating point inside the curved area without loosing the curvature - this means producing a DC shift while introducing no appreciable amount of resistive NFB in the Source circuit.

There are several ways of doing this - the method shown below simply consists of applying a DC voltage to the Source via an op-amp. In effect, this shifts the Transfer curve over the Quiescent operating point (@Vg=0) thus achieving our two-fold goal. There are several other ways of doin this that I know of but the practicality of this approach also allows continuous variation (tuning) of the relative position of the Quiescent Bias point inside the curve and at the same time setting the idling current level. This does a few things for us : (i) by being able to set the bias point we can control the AC gain at the Quiescent point and (ii) also the standby Noise level, (iii) we can set the signal limits for desired output shape and headroom, and (iv) the exact value of Vp isn't required to be known in advance - only a rough estimate suffices.


By using an op-amp as a near-ideal voltage source and then converting the jFET Drain/Source Current to a voltage through a Drain load resistor we can produce an output voltage that mimmicks the shape of the non-linear jFET Transfer current since a scalar product (r*I=V) is linear by definition. By making the resistor variable we obtain a simple volume control and thus implement a variable output non-linear booster circuit. When Triode circuits operate without localized negative feedback in the Cathode circuit they roughly produce a 3/2 (1.5) power Transfer function; and when jFET circuits operate without localized negative feedback in the Source circuit they roughly produce a 2nd power Transfer function. We can then regard this non-degenerated/shifted jFET circuit as producing a greater output of similarly induced harmonics than its Triodic counterpart by way of a greater order of curvature - more importantly this curvature profile similarity extends to both extreme limits of operation (cutoff and saturation). Hence this circuit can be seen to produce an exaggerated Triode action in terms of dynamic harmonic production."

[tca]

It does work but that bias arrangement is very touch sensitive...

[tubegeek]

I have disabled the opamp feed to the source and subbed a resistor in place and it is working fine.  As soon as I hook up the opamp to the source it is dead as hell.  I metered the output of the opamp and it is working as expected.  Any suggestions?

So you can get the opamp to match the Vs as measured on a resistor only?

What would happen with a small resistance in series with the opamp output to give that node a little LESS TIGHT hold on Vs?

[Frank_NH]

Pinout is known and correct I am using 2n5457.  I have disabled the opamp feed to the source and subbed a resistor in place and it is working fine.  As soon as I hook up the opamp to the source it is dead as hell.  I metered the output of the opamp and it is working as expected.  Any suggestions?

I'm interested in this too.  Did you measure opamp output voltage when it was connected to the FET?  And when you say "dead"  that means no output even when the drain resistance is altered?  What is the drain voltage?

[samhay]

I have played with this circuit. You will probably have better luck using a single supply op-amp like the LM358.

There is a thread here from about a year ago where we played with the circuit, which I can't find at the moment. There is also a parallel thread from that time on the other forum called 'Class-A JFET / Triode Exaggerator'.

Edit - here we go: http://www.diystompboxes.com/smfforum/index.php?topic=108632.0

[tca]

samhay, your reply at the time

> It sounds nice until overdriven, but I'm not sure if it sounds very different to a Fetzer, etc.

is quite right. And it does not clip like a triode, similar clipping occurs in the Fetzer arrangement.

[WhiskeyMadeMeDoIt]

Thank you all who have replied.  I have been away from my workbench and unable to respond.   I will read through the other thread. 

[WhiskeyMadeMeDoIt]

Pinout is known and correct I am using 2n5457.  I have disabled the opamp feed to the source and subbed a resistor in place and it is working fine.  As soon as I hook up the opamp to the source it is dead as hell.  I metered the output of the opamp and it is working as expected.  Any suggestions?

I'm interested in this too.  Did you measure opamp output voltage when it was connected to the FET?  And when you say "dead"  that means no output even when the drain resistance is altered?  What is the drain voltage?

I disconnected the opamp from the FET and measured the voltage to verify it changed when I turned the potentiometer.   When I reconnected it to the FET source and then rotate the drain pot  it had no output.  I tried this with different source voltages.  It had no output no matter what source or drain was set at.

[WhiskeyMadeMeDoIt]

I have read through the other post from this site and FSB. My brain is aching now and I fell like I have absorbed less than 1% of the info provided on those pages. i will be reading them again till it sinks into my thick skull. Thank you to all of you brilliant people who post on this and other sites sharing your knowledge.

TCA- I assumed the opamp was being used as a voltage source so I didn't do anything special to bias the opamp. I followed the schematic as posted. I assumed a TL072 would be fine to provide a voltage but didn't even consider that the opamp wasn't rail to rail. I didn't have a single supply LM358 to fiddle with so i started playing around and tried to side step the problem by changing FET's. I originally had a 2n5457 in the circuit and switched it out to a MPF102. Everything you have stated is on target. The bias is tricky, and it doesn't clip gracefully. I was able to get it to come to life around 2V measuring source to ground.

SAMHAY- Thank you for the information you have posted I will be ordering some LM358's so i can play along and breadboard some of your schematics you have posted.

It is late and i don't have much time to play around with the guitar and listen for the magic. I think the next step will be to build a fetzer with the same FET and do some A/B testing and get the proper opamp and try it with different FET's. I will also be testing a bi-polar supply and seeing what happens when the source tips into the negative region. I will also play around with the gate bias as suggested by slacker and see if i can get around the TL072 inability to hit the bottom rail ( if it even matters ).

J.C. suggest this circuit makes magic when feed into a tube amp or Mu-amp style circuit. I have breadboarded a SRPP using 2n5457's and fed the triode exaggerator into it. I ran the output of the SRPP into the second opamp setup as a buffer. It sounds good on initial testing but I have not had time to test it out and crank the volume. It goes from a clean boost to a nice overdrive. I will post my schematic later once i have finished playing with this thing ( if it is worth posting )

[Frank_NH]

Thanks for the update.  It seems like a lot of trouble when many successful amp-like overdrive circuits have been developed using common source jfet and mu-amp (srrp) configurations in conjunction with an op amp stage or two.  My interest lately is in mu-amp stages that are driven by an op amp stage with diode clipping so as to mimic an overdrive pedal driving an amp.  There are pedals now that do this with a booster stage driving mu-amp stages e.g. Catalinbread Galileo.