Hi guys, i cant get 2N5458 anywhere, can i swap it for J111 or J112?
What are the voltages on the output buffer JFet please?

I have a 1M resistor on the gate to ground. Opamp is 0P07CP. C12 is 2n7.

Use J212 and 12k to 15k Source resistor to bring Source closer to Vcc/2..
(although, retaining 10k Source resistor should be OK due to low amplitude input..)

You have to aim for Source voltage NOT lower than D2/D3 pair forward voltage drop..
(but delete R13 'cause it forms a voltage divider with R12 halving current sinking ability..)

P.S.
If you insist to retain R13, tweak R12 for Source voltage equal to (or higher than) 2 X D2/D3 forward voltage drop..
(e.g. > 1.3V for Si clipping diodes)

In case of asymmetrical clipping, take into account D2's forward voltage drop..

edit: Sorry, just noticed you refer on J111/J112..
Use J111 and 15k Source resistor..
(keep aiming for Sourve voltage a bit higher than double the diodes forward voltage drop..)

J212 is obsolete. J111 is still in production by ONsemi, it will do fine and give about 3.7V at source with 10k resistor. But no two J111's are the same of course, so you can select one that gives Vsource closest to half Vsupply.

While J111 shows about 7V at the source, J112 measures just 3.4V (socketed). I left the 10k res on the source just to start with. I guess it will be easier to continue with J112.

Another one i've tried is J201, i brought 4.5V at the gate by a 2M2/2M2 divider and placed 4k7 as the source resistor. That shows 4.5V at the source.

As long as D2/D3 are 1N4148, ALL of the above are more than OK..!!
(you don't need an output positive voltage swing more than 650mV and negative more than 1.3V ..)

Actually, you can go ever lower due to JFET low transconductance (compared to BJT), hence lower than unity voltage gain but let's keep it as a safety margin..

P.S.
I'd go to about 6V Source bias, in case of prospective change of clipping diodes with another pair of forward voltage drop up to 3V..

Hi there


Try getting J113 and it will bias around 2V
J113 will give lower output impedance than 2N5485 ( see next paragraph) but also marginally less noise. Noise gain is less than 1 times i this application.

The J111 and J112 both have too high Idss
and Ugs for the application that are here low current and low voltage so they may bias far too high or be locked.
For the J112 you could increase current by lowering the Source load to around 2K.
You would get lower output impedance but that won't make any audible difference as long as Volume control is not at max but it may let J112 rest more towards mid supply .

https://www.onsemi.com/download/data-sheet/pdf/mmbfj113-d.pdf

At your service
Bjorn Juhl
BJF Electronics
Sweden

Quote from: BJF on January 19, 2025, 01:31:20 AMThe J111 and J112 both have too high Idss

I don't see the problem, the Jfet in this circuit is not running at Idss. A J113 would run at about 0.2mA, a J111-J112 maybe at 0.4-0.5mA.

Hi there,
Yes all good

Oh not so much of a problem if you are OK with the Source biasing at 7Vs.Then you might get a problem with the transistor locking itself if battery voltage drops a little bit. From Idss and Ugs you get the approximate resting point the Source will have. See the ranges in the data sheet.Yes the transistor will not work at Idss but the actual Id it will have is dependent on the Idss and Ugs. Yes typically a J113 will rest in such a circuit at about 2Vs at Source. Typically J113 has an Idss of 2mA but Ugs varies in my experience.
So if you want to define the voltage at Source above ground Idss and Ugs will let you know which type you want to use to get desired voltage.
Arguably the range of through hole J Fets is limited with many dealers but most suppliers would have J111,J112 and J113.
Then if as suggested above you want to go with a resting point a little higher than 1,3-1,4V J113 would be a good candidate whereas J111 and J112 would rest at a lot higher voltage.

As sidenote
A typical 2N5458 would rest in this circuit at a voltage between 1,3V and 5V and still be well within specified data

https://www.onsemi.com/download/data-sheet/pdf/mmbfj113-d.pdf

https://www.mit.edu/~6.301/2N5459.pdf

At your service
Bjorn Juhl
BJF Electronics
Sweden

Pft, just use a BJT when plenty of hfe instead, and connect R11 to Vref . Maybe change it to 470k.

Quote from: BJF on January 19, 2025, 10:11:48 AMOh not so much of a problem if you are OK with the Source biasing at 7Vs

As you can see in post 3, a certain J112 gives a source voltage of 3.7V. Problem solved.

Quote from: BJF on January 19, 2025, 10:11:48 AMTypically J113 has an Idss of 2mA but Ugs varies in my experience.

Typical Idss for J113 is more like 20mA. Very easy to check, just like Vgs. The source voltage in this circuit will be    a bit less than Vgs. A J112 you are most likely to find with a Vgs of about half Vsupply, 4.5V in this case. If you are building just one pedal it is not a big problem to select a suitable Jfet. For mass production, I would use a bjt.

Hi There

Indeed plenty of buffer options and let's assume 1M Ohm input Z so that even if tone control is at its largest resistance then one could bootstrap a BJT  to get well above or for higher PSRR perhaps use an integrated buffer option like TL 068 or LM310; or why not take the MOS fet option while it takes more parts to make stable?- but it is cool
However J Fets one just has to love as those are individuals but they make such simplicity and low parts count and typically low current draw making them ideal for battery operation

Slight problem with single transistor buffers is that PSRR is very poor but if there is enough voltage one can extra filter the Collector or Drain


https://www.farnell.com/datasheets/105178.pdf

http://www.ic-ts-histo.de/sehenswertes_histo-datenblatt/TL068.pdf

Have fun
Bjorn Juhl
BJF Electronics
Sweden

 

Hi there

In my experience measuring about 1000 J113s over the passed 30 years they have been deviating very little from 2mA just as the data sheet says which is really a rare thing with J fets to have something even remotely consistent as a consequence Ugs have been within shall we say a ballpark.

And that´s just my empirical experience

I would not be surprised if there are variants that deviate from data sheet but I am saying I haven´t seen such a specimen as yet

Have fun
BJ

Quote from: merlinb on January 19, 2025, 11:00:59 AMjust use a BJT

I do love this guy..!!
(with all the respect..)

P.S.1
Nothing better than a BJT buffer except, perhaps, a CFP one..

P.S.2
@Bjorh: I do know that you know that only a cap and a resistor needed to bootstrap Collector impedance raising PSRR to astronomical level..

Quote from: BJF on January 19, 2025, 12:51:18 PMIn my experience measuring about 1000 J113s over the passed 30 years they have been deviating very little from 2mA just as the data sheet says

The datasheet says idss is 2mA minimum for a J113. But this is not typical at all! I have used a lot of them, and the typical Idss I find is 18-19 mA. Also, if I 'measure' J113 in the LTspice simulator, I get 18 mA.
So I am wondering, how did you measure those 1000 J113's? I suspect it is not the way I do, wich is connect gate and source to ground and then supply 9 volts to the drain with an Ampere meter in series..

Quote from: antonis on January 17, 2025, 03:07:42 PMAs long as D2/D3 are 1N4148, ALL of the above are more than OK..!!
(you don't need an output positive voltage swing more than 650mV and negative more than 1.3V ..)

Why do i need twice as much headroom for the negative swing?

Quote from: antonis on January 19, 2025, 01:39:02 PMNothing better than a BJT buffer except, perhaps, a CFP one..

But the question was not 'can you suggest an alternative for a Jfet buffer'.

Quote from: Clint Eastwood on January 19, 2025, 01:52:14 PMBut the question was not 'can you suggest an alternative for a Jfet buffer'.

Given that JFETs are increasingly hard to get and no-one is really arguing that the choice of output buffer technology is a massive part of the sound of this circuit, perhaps "can you suggest an alternative for a Jfet buffer?" would have been a sensible query. We don't *have* to make life difficult for ourselves...

Quote from: ElectricDruid on January 19, 2025, 02:46:30 PMWe don't *have* to make life difficult for ourselves...

That is true, but then you could similarly argue that the OP might as well buy a Proco RAT pedal in stead of building one.
I can understand that when making a clone or doing a repair you want to stay as close to the original as possible/reasonable. In the end, a matter of taste I guess 

Quote from: Clint Eastwood on January 19, 2025, 01:52:14 PM

Quote from: antonis on January 19, 2025, 01:39:02 PMNothing better than a BJT buffer except, perhaps, a CFP one..

But the question was not 'can you suggest an alternative for a Jfet buffer'.

True & Correct..

I wasn't talking to you or to OP..
(excuse me me for deviating from OP's original query..)