JFET Common Source DC Bias Spreadsheet

[Frank_NH]

Hi all,

I spruced up the spreadsheet I've mentioned in other posts and am making it freely available for others to use.  Here is the link to download it:

JFET Spreadsheet

(UPDATE: 2/25/2015 - New Version 2.0 at the Link Below!)

JFET Spreadsheet v2.0

Basically you provide your JFET's Idss and Vp (which I call Vgs(off) in the spreadsheet) and the desired source voltage Vdd (e.g. 9V).  You then have several choices:

(1) Midpoint bias.  This assumes the drain current is 1/*Idss and Vd(bias) = 1/2*Vdd (i.e. 4.5 V drain voltage).  A good reference point.
(2) Known source resistor and drain voltage.  Supply those values and it calculates the drain resistance and other parameters.  This is what you can use if you have an amp sim (like one of the Runoffgroove circuits) with a known source resistance and target drain voltage (e.g. 4.5 V).  You could use this to determine a single drain resistor which would bias your JFET (approximately), or you can just make sure you have the right size trimmer.  Note that your stage needs to have the common source configuration for this to be valid (mu amp and other complex JFET amp circuits do not apply).
(3) Known source and drain resistors.  In this case, the calculator figures out the bias voltages (Vd, Vs).  If these look out of whack (i.e. 1 V or 8V), then try a different JFET (Idss, Vgs(off)).  This is what I used to determine what JFETs would work in the Plexi-Drive.  Note that the Plexi-Drive uses the same source and drain resistors for each stage, so having a matched set of J201s makes sense.
(4) Fetzer Valve Bias.  I put this in for comparison purposes using formulas at the Runoffgroove site.  It is interesting to note how different this is from other types of biasing.  Is it better?  Only your ears can say.  It does seem to have less gain than other bias points.

Please note that the gain calculation (Av) is a crude approximation, and is mainly there for comparison purposes.  If someone wants to enhance this with more complicated formulae, feel free to do so.  The main thing is that you're getting some amplification (and NOT one or less than one!).

Have fun, and let me know if you discover any mistakes or problems.  There is NO warranty expressed or implied, because it's free!

Cheers,
Frank_NH

[rocket88]

thanks buddy. now once i finish sorting the 500 ME4003's i got, i can start sorting my JFETs. btw, frank, do you think all fets should be sorted like this, like 2n5457's?

[Frank_NH]

Hey Zach,

I just sorted through and tested a bunch of J202s.  What I've done for all my JFETs is to get a sheet of paper, some labels (small pieces of white masking tape work well), and the fire up the JFET tester.  I mark the sheet A, B, C... If you have more than 26, then A1, B1 etc.  I test the JFET, record Idss and Vp, then stick the label to the legs of the JFET.  Once done I put all the JFETs in a safe bad for storage.  Later on I'll put the data in spreadsheet.  If I use the JFET in a build, I'll make a note about which build used it.  (That way, if I'm desperate, I can pull it out later ).

Interestingly, all of my batch of 20 J202s were very similar to one another (Vp ~ -2.0 to -2.2 V, Idss ~ 2.5 - 3.0 mA), and so should find homes in effects which use 2N5457s.

[rocket88]

i was just asking, cause i stated reading the articles you mentioned about JFETs and their ability to emulate tubes. so i started looking at the circuits in which that seems to be the ultimate goal, and i noticed that the vast majority of them use J201's, rather then 2N5457, catalinbread seeming to be the exception which use 2N5457 & BS170 for their amp-in-a-box, pedals. so i guess then i have a lot more to sort then i had originally planned.  but if it's worth the extra work, then it would be silly not to do it.

[Frank_NH]

What I've found is that the lower Idss and Vp values in J201 lead to larger stage gains versus other types.  This apparently is more tube-like than other JFETs, but then there are exceptions.

BTW, BS170s are MOSFETs and thus would not be applicable for the sorting we're doing with N channel JFETs like the J201.  I'll have to read up on those, as there are a lot of overdrive/distortion designs which use a MOSFET stage driving a JFET stage (e.g. ROG Peppermill).

[rocket88]

that's pretty interesting actually. can't want to try building with and without testing to compare the differences. i knew that BS170's were MOSFETs, just commenting that catalinbread likes to use them in some of their amp emulator, amp-in-a-box pedals. . i just remember reading something a year or two ago that tried to explain that different FETs were created to mimic tubes for amps, and most amps were either triode or pentode. i remember it saying something like JFETs are supposed to behave like pentodes, and MOSFETs are supposed to behave like triodes, or it could be the other way around. not so sure anymore cause of how long ago i read it.

[Frank_NH]

To be honest, I don't think it matters if JFETs or MOSFETs act like tubes or not.  All that matters is how the amp-sim/overdrive/distortion effect sounds in the end.   And a lot of that has to do with the tone shaping between stages, and the tone stack itself.

My hope is that by testing JFETs and recording their Idss and Vp values, they can be used more predictably.  One thing I would like to try on the breadboard is to make a simple common source stage (like the Fetzer stage) with some coupling caps, and try out different JFETs to see if the signal boost is in line with the spreadsheet predictions for given bias point (as defined by the source and drain resistors).

[IvIark]

Excellent job on this Frank, I'll have a play around with it.  Although I'm not happy, I've got about 2000 JFETs to test now

[Surgeon]

This is awesome Frank! Can't thank you enough for sharing this!

[rocket88]

God damn Mark. Just when I think in creating a stash like yours, you make me realize how far off I really am. Lol

[Frank_NH]

No problem Surgeon

IvIark - 2000 JFETs??  I think that would last me at least two lifetimes   Of course, given that J201s are now at least $1/piece and all through-hole JFETs are increasing in value, maybe you can recoup your investment ten fold in a few years.  And if you can sell matched sets, they can sell for even more!  (Someone should start a commodity exchange for JFETs - ha ha ).

[rocket88]

I forgot to mention about the triode/pentode thing with FETs. I agree it really doesn't matter as long as they sound good. Just thought it was interesting as to why FETs were made. I mean it kinda makes sense, specially when you think about all the solid state amps, and claims of how some can match the sound and behavior of tubes. Just my 2FETs.

[dbat69]

Frank, many thanks for the spreadsheet and information - this will be very useful although I think I'm going to have to buy a lot more JFETs to be able to match them

Anyone got some good sources for cheap J201s  ??  

[rocket88]

I get mine from smallbear, as they're good quality and they're going for $0.55 each. I can order a bunch and send them over if the shipping direct from them is outrageous.

[dbat69]

Thanks Zach.
Tayda were doing them for $0.48 so I have just been getting a few each order I place - I've got about 60 now but figure I will need more.

Lucky Mark with 2000

I just noticed your comment above about the MEs - I should have read all the posts before posting in the other thread

[Surgeon]

I have a bunch of through hole ones but I almost exclusively use SMD ones from mouser with adaptor boards... they're very very good so far. The only problem is that the header pins don't fit in regular sockets for through-hole components, making measurements a bit more complicated...

[rocket88]

I know the SMD ones are supposed to be more consistent then the through hole, and now I assume they're cheaper, but any other benefit to the SMD?

[Surgeon]

Not that I know of.

They are also not necessarily cheaper if you take in account the fact that you have to buy boards (oshpark has great prices though) and header pins. IIRC they're pretty close to the price SmallBear sells them for... I just find this more convenient than having to rely on suppliers that may not have stocks of them for much longer and/or dealing with sketchy ones from eBay. I have no proof but there's been lots of talk of rebranding/sanding and renaming jfets for a while. Plus, the consistency of the batch I have (100 mmbfj201) has been great, which can't be said of the through-holes I have on hand.

[IvIark]

They'll get much cheaper when more people start selling them.  At the moment the few sellers on eBay who do sell them don't have much competition but when more people are buying in larger quantities the price will drop a lot.

[j-pee]

a precious package of knowledge :)
and a great "tool"...

thank you

PS:

suggestion: put "calculator" in the title... and a couple of times in the opening post, too :)
cause it only appears 1nce :)
and both google- and site-searches could find it more easily...

it could be also mentioned that Mircosoft's cloud spreadsheet app  sometimes displays ohm sign as "Watt" :)
I mean, this calculator-spreadsheet is not only for the experienced... :)
like myself, those who learn will LOVE to come across this nice post and the calculator...
and since resistors have W values all the time in parts list, it might help someone who
sees it for the first time if s/he knows it's, of course, ohms, not watts :)

oh, one more note:
online it is read only, evidently...
so it might be worth mentioning that downloading is not  a kind suggestion but a must :)

thanks, really!