MU Booster FET selection question

[CheapPedalCollector]

I recently bought 40 J201's from Small Bear when they were shortly available and I want to build a MU Booster, I've searched all over and I can't find any post on how to choose devices that will work the best in the circuit and if they should have some parameters matched or not.

Any advice or link is appreciated, thank you.

[idy]

The famous Keen article mentions that no matching is necessary because they are doing different jobs.
http://www.geofex.com/article_folders/foolwfets/foolwfets.htm

At least that part is easy.

[CheapPedalCollector]

OK That's relieving then. I bet certain characteristic sound better than others though, hmm I don't really want to swap 40 FETs in and out. Maybe someone has some "secret sauce" they can share about this circuit.

[Steben]

J201 is a scarce product. I would not use them in non-specific positions. They are great as single amp input stages. In a MuAmp you can combine jFET types.

[Fancy Lime]

In my experience and for my personal taste, JFETs with low transconductance are better suited for a mu-amp. High transconductance types are too gainy for me in that role.

Andy

[R.G.]

JFETs are hard. The industry essentially quit using them in any situation where something else could be used. I think this is due to the extreme variability of the JFET. It never really got any better, as bipolars and MOS devices did.

On top of that, the J201 is odd, even for a JFET. To me, it seems like a razor-edge device with its very low Vgsoff and other quirks.

The trick in JFET circuit design that the industry adopted (where it used JFETs) was to add external parts to trade away gain for predictability. The Mu-amp is almost the antithesis of this approach IMHO. It's very high gain, all right. But it will be variable with the JFET until you put in, for instance, source resistance or some other feedback mechanism to trade away. Which is a way of saying that I'm not aware of any secret sauce to make subbing in different JFETs unnecessary without making them sound less like JFETs.

[CheapPedalCollector]

R.G. so I just need to try them until I find a pair I like the sound of then?

[R.G.]

Pretty much. On the other hand, given the differences between exactly what the circuit does and how it sounds, you may not need to do a whole lot of swapping.
I suspect you may have read the article at geofex on the mu-amp. If you haven't it may help with your understanding of the circuit. http://www.geofex.com/Article_Folders/modmuamp/modmuamp.htm
The mu-amp gives you a bit of a leg up on raw single JFET amps because it fixes the bias point of the circuit (kind of) with a bias string. So you can adjust the bias voltage by tinkering these two resistors to compensate for a given set of JFETs. Sometimes just swapping upper and lower JFETs will change things substantially - they do different jobs.
As with all high gain clipping circuits, how it sounds in the end will depend not only on the circuit and its clipping but what filtering and loading happens before and after the clipper.

[amptramp]

The famous Keen article mentions that no matching is necessary because they are doing different jobs.
http://www.geofex.com/article_folders/foolwfets/foolwfets.htm

At least that part is easy.

There is one thing that is not clearly spelled out in the geofex article.  The first and second schematics are the µ-amp and a µ-amp with a source follower output.  The third circuit, although it looks similar, is an SRPP (shunt-regulated push-pull) circuit.

The µ-amp uses a normal grounded source amplifier in the lower section and a load made from a FET biased as a current source with capacitive coupling from source to gate to keep the operating point the same.  A current source is a very high-impedance load, so the gain can be quite high.

The SRPP adds one resistor but the operation is entirely different.  The upper transistor source is connected to the lower transistor drain through a resistor with the gate driven by AC coupling from the lower FET drain.  Suppose the lower drain voltage goes down due to a positive input and an increase in current.  This pulls the upper gate down. causing a reduction in current in the upper FET.  Similarly, if the lower drain goes up, the increase in gate voltage in the upper FET causes it to draw more current.  If the value of the resistor between the lower drain and the upper source is the reciprocal of the transconductance of the upper FET, the circuit is balanced and the change in upper drain current is equal and opposite to the change in lower drain current.  This is why the gain appears to double in an SRPP compared to the same devices in a µ-amp.

One thing all three of these circuits have in common is they need a load to avoid saturation.  The upper and lower FET currents are in series so the difference in current has to go to the load.  In both the µ-amp and the SRPP, you have a difference in upper and lower drain current meaning any difference between lower FET current and upper FET current goes into the load and if the load impedance is too high, it cannot take the difference in current and it slams against the rails.

You could make a neat fuzz by using a variable resistor as a load.  Low resistance gives linear operation and high resistance results in almost square waves as the difference in current has nowhere to go.

[Steben]

Once you have those J201s, you MUST use them for quircky circuits.

[CheapPedalCollector]

R.G. Yeah I read the article years and years ago (I've been around since the 90s and just lurk) and I'm a repair guy so I'm familiar with the circuit, I just never built one before.

amptramp Thank you for the technical explanation of the two circuits, that gives me some ideas. I'll build both versions and see if I can get them balanced with minimal tweaking.