Diefet problem - low volume and gain, sounds thin

[Frank_NH]

Interesting.  I hope it is the JFETs - that would explain a lot.  BTW, when you swap out the JFETs make sure to rebias with the trimmer pots.  I assume you are doing that.

For J201s, if you measure Idss and Vp using the JFET tester Paul referred to, Idss should be in the range of 0.3 - 0.8 mA, and Vp should be between about 0.7 - 1.1 V.  These ranges are based on my own stock of J201s (23 JFETs in all were tested).  

[marcus262]

hi,

I just measured vgs(off) using this method, 9V battery and multimeter:



and I get:

a  -1.80 V
b  -1.00 V
c  -1.66 V
d  -1.44 V
e  -1.05 V

so, two of them in the upper limit and three way off.

[Frank_NH]

Yes, a, c, and d are way off.  What are the Idss values for those JFETs??

[marcus262]

I tried to measure, but cannot get any reading.
Either my multimeter has problem with lower current or I am doing something wrong.

[Frank_NH]

If you're setting up your jfets for Vgs (off), the jfet won't flow any current.   You can use the ROG setup as follows:

[Sensei Tim]

going back to my original comment, i don't know why people are still buying through hole J201's from places like tayda.

They are all fake and/or out of spec pieces. Same thing with the 1n34a diodes on tayda.

Spend the extra few dollars/euros and buy some SMD j201's and adapter boards.  if you can solder something onto a very board you can solder an SMD j201 onto an adapter board.

[motterpaul]

I bought mine a couple years ago. They all seem to work but I have seen a lot of complaints lately about recently purchased ones. The SMD ones are said to be very reliable. I am sure I will go this route myself soon.

Does anyone have a good US-based source for the adapter boards?

[Sensei Tim]

i get mine thru oshpark:

https://oshpark.com/shared_projects/8bf7XW05

[marcus262]

This is very strange,
i tried with below picture:



Tried N type - no reading , swapped to P type - get a reading of 0.45 (dont know if amps, miliamps - meter crap), and transistor gets hot very quickly.

maybe I am just slow, what am i doing wrong?


p.s. All of you who have built the diefet, did you make any modifications to the original vero, in parts, pots or pot wiring? I ask this beacuse i noticed on diefet that when i raise the treble to full, I get volume drop, but that doesnt happen on Dr. Boogey.

[Sensei Tim]

the posted layout is verified and should not have any strange behaviour like that.

[marcus262]

I noticed motterpaul had the same observation in some post before about treble on full and volume drop

[Sensei Tim]

here is a playthrough on one of the diefets that i've made. there shouold be no volume drop with treble at full.

https://www.youtube.com/watch?v=sH017xfcol4

[marcus262]

yeah, that playthrough made me want to make a diefet. :)

I noticed when treble full you don't drop volume.
Sounds great.

Will check again when new j201 arrive, hopefully the main problem with gain will be solved, then i'll check that treble pot.

You guys have been great and patient with my problems, and I am very grateful. :)

[marcus262]

didn't know about smd j201,
so it has the same specs as normal j201 and you solder it to the board which you solder to vero?

[Frank_NH]

J201s are n-channel JFETs - you should have a positive voltage at the drain, and a ground or negative voltage at the source.  Don't reverse the voltage!

Your n-channel setup is correct.  The ROG approach has you measuring a voltage drop across a resistor to get Idss = Voltage drop (Measured) / Resistor (100 ohms).  Multiply the number you read by 10 and voila you have Idss in mA.  You have a good DMM that can measure current accurately, then you can use your setup.

The way it works is that JFETs flow their maximum current when Vgs = 0 (see below).  So if you connect a wire between the gate and source (making Vg-Vs = Vgs = 0) and put a big enough voltage across the JFETs you'll get the max current, Idss.  You can measure that current any way you like - usually measuring a voltage drop with a known resistor is easier and can be quite accurate (use a 1% metal film for best accuracy).

BTW, how big a voltage drop across the drain and source (Vds) do you need to measure Idss?  It turns out that the Idss current level is reached for voltages greater than Vgs(off)!  As you can see from the curves below, for any voltage drop greater than the pinchoff voltage, Vp = Vgs(off), Idss remains constant for Vgs=0.  Pretty neat, huh?  And you already know that number, so for most low power JFETs, 9V is enough (maybe not for some others, always check the data sheet for the range of Vp).

And that is your electronics lesson for today - class dismissed.  

 

[Beaker]

Surface mount JFETs are essentially the same as the through hole versions - just smaller, and legless. you will need to buy adapter boards to allow you to use them with vero layouts.

It's not just J201s either, there are lots of builds on here that need 2N5457, and they are as bad in terms of finding good ones - most are garbage, fakes or expensive if genuine.

Advantages are that they are considerably cheaper - even when you include the adapter board. They are much more consistent - little variation from the specified nominal values.
Almost no chance of duds or out of spec values.
No chance (yet at least!) of buying fakes or rejects.
Source and Drain legs are interchangeable, so as long as you fit the gate leg in the right hole, you can't get it backwards - which is pretty neat!
Finally, you can forget having to test them all before using them, so saves you time.

Disadvantages are that they are freakin' tiny - you will need a good magnifier for soldering them.
It will take you longer to solder JFETs on the boards, solder on legs - you can use header pins for this.
They might be unsuitable for some (probably very few) builds, that seem to work better with values that stretch the tolerance band for voltage and current (because they are so consistent).

Here is a datasheet:

http://www.mouser.com/ds/2/149/MMBFJ201-889832.pdf

As Tim says, adapter boards are available from Osh Park, or from Chromosphere's shop. He's in Australia, but his boards are fantastic, and cheap. He's one of the good guys!

http://www.diyguitarpedals.com.au/shop/index.php

There's a few threads now on using these little sods, so using the search at the top of the page will get you a load more info on them.

[marcus262]

thanks for the lesson :)

I just googled them, yeah they are very tiny.
More consistent, no fakes , I will try to get them in the future builds.

[motterpaul]

FYI _ I just spent hours troubleshooting a RAT - I was working on getting rid of some oscillation for a long time but in the middle of that I lost about 75% of my gain but the effect still worked.

It turned out it was my 2n5457. It had been working for months so had soldered it into the sockets - or else I would have tried replacing it sooner. I looked at EVERYTHING for hours (like you) and finally took that out and checked it with a transistor tester - read as either "dual diode" or "no component detected." Never read as a JFET.

So I replaced it and was back where I started with my original problem. Man, I hate days like that. But I don't remember ever having a JFet die like that before. Now I know what to look for - if I lose a lot of gain it is probably one of the gain stages (duh).

[marcus262]

I can feel you pain. Troubleshooting stuff can be really tiring.
Damn those jfets, it seems we put too much trust in them in the first place and check them last.

[marcus262]

Hi guys,

finally managed to solve the problem, it was fake/bad jfets.
The ones I ordered from ebay from the guy in UK work great and now there is full distortion.
I am very happy now, just playing with biasing and final trimmer till I get what I like best.
If you found some sweet spots, please feel free to share :)

Thanks guys for all your help :)

In summary - don't buy j201 from Banzai, obviously bad supplier.