JFET Buffer and buffer cable

[JFrankParnell]

Sept 30, 2006 23:33:58 GMT -5 JohnH said:

as shown on the LP Max thread:

guitarnuts2.proboards45.com/index.cgi?board=schem&action=display&thread=1138768962

Page 2 of the LPmax thread has further samples, through the internal buffer:

guitarnuts2.proboards45.com/index.cgi?board=schem&action=display&thread=1138768962&page=2

These links dont work for me

[JohnH]

I know - the schematics section was reorganised since then. Plaese see link in reply 56 above
J

[JFrankParnell]

goes to guitarnuts2.proboards.com/index.cgi?board=schem&action=display&thread=3159&page=1 Jimmy Page mod, with and without buffer. Is that the LPMax?

[JFrankParnell]

Are there any gotchas to look out for if I do a kill switch like: guitarnuts2.proboards.com/index.cgi?board=wiring&action=display&thread=5090 with either built in buffer or the buffer cable? I'll prolly do the buffer cable.

[sumgai]

jfrank,

In general, low-power transistors are pretty hardy - they don't care if you short their output for a little while. Leaving the output shorted for a long time will just drain the battery real quick, and in some cases the transistor may get warm, or it may even be "touchy" about the shorted output and get upset with you. But for momentary periods, like your spring-loaded kill switch, everything should be fine.

Shorting the input of any amp circuit (tube or ss) is never a bad thing, no damage of any sort ever occurs in this scenario.

HTH



sumgai

[ashcatlt]

That is, to be safe, put the kill switch before the buffer.

[asmith]

I'm currently ordering parts for John's build of the JFET Buffer Cable from Mouser and they've just told me they've found out Fairchild's standard "MPF102" is out of stock. Minimum Order Quantity is therefore 1,000.

There are a couple of other MPF102's available but with slightly different specs.

For example, standard MPF102 has a Drain Current '(Idss at Vgs=0)' of "8 mA to 80 mA," whereas MPF102_D27Z's is "2 mA to 20 mA."

I'm unsure of how much of a difference to the operation of the circuit those spec changes will make. I can Google and self-educate to the ends of the earth; what it comes down to is that I'm not yet confident enough to put my money on buying these transistors, and I'd like the advice of people with more experience in this.

Are any of the MPF102 variants good-to-go, or is there another JFET that I might purchase instead? (The innerwebz seems to push BF245 as a possibility)

Cheers

[sumgai]

Ace,

Hmmmm. Fairchild's own datasheet says "2 - 20ma", with a nominal drain of 10ma, under normal circumstances/usage. Makes me wonder what you've been reading....

But nonetheless, for your needs, I'd be very surprised if anything the family of alternate MPF102's didn't work just fine. The short and sweet is, those ratings are minimum and maxium values, and what you need to accomplish falls well within the normal parameters.

Because the BF series is pretty old, they might be a bit noisier than the MPF stuff, but they should still work. Always worth a try.

I also note, with no small grin at the irony, that Radio Shack wants a $1.49 (USD) for a single piece, whereas Amazon wants $4.93 for the same thing - one single piece. I ask you, ain't The Free Market a grand place or what?!

HTH




sumgai

[reTrEaD]

I'm no expert, but I don't think that particular spec means much in this application. The only time Vgs could be zero for this circuit would be if R1 was shorted. Vgs(off) would be a more important characteristic, but that's not listed in the specs for the D27Z. I wouldn't expect a huge difference in that spec between two variants of the same basic designation.

Because it's a source-follower, this circuit should be tolerant of a fairly wide deviation of device characteristics. At most, you might need to change the value of R1.

[asmith]

I've been reading Mouser's website.

Thanks SG. Substituted for MPF102_D27Z on my order.


EDIT: Ninja-d by rT.

Dec 13, 2011 13:28:27 GMT -5 reTrEaD said:

The only time Vgs could be zero for this circuit would be if R1 was shorted. Vgs(off) would be a more important characteristic, but that's not listed in the specs for the D27Z.

OK. That makes sense.

I wouldn't expect a huge difference in that spec between two variants of the same basic designation.

Good to know for future builds. Cheers.

For future thread-readers: what is the difference between the variations of the device?

[JohnH]

Im not sure about the variations on the MPF102, but Id expect they would work in this circuit. I dont normally use them, but in this case, the generally more useful 2n5457 and J201 are not suitable for the cable buffer because they need too many components. We need the source to sit several volts above ground so that it can move up and down without hitting the power rails. With a device with a highish Vgs[off] like an MPF102, this can be done with just a resistor to ground at the gate, but with others youd need two resisitors and a cap.

J

[warmstrat]

Hi guys (and probably JohnH specifically)

I'm becoming interested in the idea of buffering, and doing the reading up - but I don't have a concrete idea of how much improvement I can get. Tried to listen to your original recordings on the 1st post, but the links 404'd on me. Also, could you comment on the benefits of gaining the signal more than the 1x buffer?

Thanks

[ashcatlt]

Have you got pedals which are not true bypass? Boss, DOD, Ibanez, most big name pedals are buffered even in bypass mode. Have you got cables of various lengths? Plug the pedal into your amp using whichever (working) cable you want and then plug your guitar into that using shorter and shorter cables. Should give you an idea. If you can figure out a way to strum the guitar while it's connected to the pedal by a short patch cable (like you'd use between two pedals on your board) you'll be about as close as you can get to on on-board buffer or buffer cable.

[JohnH]

Hi WS, nice to hear from you again.

ashs suggestion of trying with an extremely short cable will give you a good impression of the extra high-end clarity that a buffer at the guitar can provide.

Sorry about the link, I try to upload something again this weekend.

Adding gain to the buffer stage (meaning, in this case, just a clean boost at the guitar, maybe a small amount of extra harmonics) is useful if you want to push your amp a bit harder. or if you need it to match levels of different pickups or piezos etc. Generally though, I think the x1 buffers are a better thing to build in, and use a stompbox for extra boost when you want it.

John

[warmstrat]

Thanks, guys.. I'll have a look around and do some experiments... and start a new thread if it becomes any more complex/interesting than this

[warmstrat]

I'm actually going out to buy components this week. It's been a hectic few months, but thesis is handed in and I'm pretty much on holiday, so time to start tinkering!

You say in one of your posts that you modified the circuit to include a 1M resistor from output to ground to prevent "clunking". I'm not sure if you've updated your circuit diagrams with some other method of achieving the same thing (I can't see a 1M anywhere) or if I should stick it in myself... ?

Will record some stuff once I've got it working for the benefit of those who come back to this old thread

[JohnH]

Hi warmstrat,

So which type ae you palnning to build - is it a built in x1 buffer?

The output tie down resistor is not a very critical value, in this thread I think I showed 3.3M in one place on post 1. 1M is fine. On this one I showed 1.5M:

guitarnuts2.proboards.com/index.cgi?board=schem&action=display&thread=4595

The thing that is slightly important about selecting the values is what you want to be able to output to. If its just into a guitar ampm you can run at lower current, (eg with a 100k source resistor) smaller output cap etc. But if you want the option of going straight into a mixer line in, then it needs a bit more juice flowing, such as the schematic with a 22k source resistor and 2.2uF output cap.

[warmstrat]

It's going to be the x1 cable version. For now all I'm powering is my guitar amp, but I'll leave it as your diagram states for the sake of getting it working without tinkering too much.

Finding (either of) those transistors is proving to be tricky! I might have to order them off eBay - unless anyone knows a replacement/equivalent part. They all seem to be discontinued where I am.

[JohnH]

The buffer cable is a good little project, at no risk to your guitar!

The jfet for the buffer cable is the MPF102. It's picked because in this circuit it only needs to be biased to ground and will still give a reasonable source output voltage of around 2.5V which is enough to allow for voltage swings from a guitar signal. For other circuits I use 2n5457 or j201, which need an extra resistor at the input and so also an input cap, too hard to all get into a jack.

2n3819 would work here. Or, are there any other jfets that you can get? In most ways it's not a very critical circuit.
John

[warmstrat]

Looks like I'll fare slightly better with the 2N3819, thank you

[warmstrat]

Finished and working!

I'll post some clips during the week. Thanks for your help again, JohnH +1

[cnb]

Hey there everyone, new here.
Quick question: would this cause any problems with effects pedals? As far as I know, the Fuzz Face is very sensitive to impedance. What about something like this revolutiondeux.blogspot.com/2012/03/tim-escobedo-tytewadd-fuzz.html?

[JohnH]

Hi cnb - welcome to GN2.

The buffers will drive just about any kind of input, including fuzz-faces. But the thing is, fuzz-face circuits have inputs that are (so Ive read) somewhat interactive with the particular output characteristics of a guitar. They will definitely work with the buffer, but the tone may change, and the issue is with the fuzz-face rather than the buffer. Any other usual kind of stomp-box works fine with the buffer.

And Warmstrat, I forgot to offer congrats at getting the buffer cable going. So what do you think of it?

cheers

John

[cnb]

Thanks, JohnH. I'm usually pretty bad with electronics because I try to take on way too ambitious projects. Your source follower seems like a fairly friendly circuit, so I'm going to try to put one in an Altoids tin so it's fairly portable. Just have to find an easy but reversible way to mount it to a guitar so it can be swapped around.

I'm also eying for the future that piezo design you came up with as I have a guitar with a strat trem and no pickups laying in my rack at the moment.

[warmstrat]

Hi John

I haven't posted any clips because my recording rig is... well. it's not. Lets leave it there.

The change in tone is more subtle than I expected - but then again, I'm not accustomed to using very long cables anyway, confined as my guitar playing currently is to the walls of my bedroom.

It is very nice and clear and crisp though, which is exactly what I was looking for - that extra clarity. Reminds me of what I know as "the EMG sound", that extra clarity you get from active pickups. I've also started rolling down my tone knob, for the first time ever, to get warmer jazzy tones, so it must be doing something

The unexpected highlight of the whole thing, though, is that I have "normal" tone at all positions of my volume knob! Things used to get really muddy when I turned it down, but now it's staying sweet and clear the whole way.

Thanks again for your help!

[ashcatlt]

Dec 1, 2012 20:41:50 GMT -5 cnb said:

Hey there everyone, new here.
Quick question: would this cause any problems with effects pedals? As far as I know, the Fuzz Face is very sensitive to impedance. What about something like this revolutiondeux.blogspot.com/2012/03/tim-escobedo-tytewadd-fuzz.html?

FuzzFace and the like have a deliberately low in-Z. This combines with cable capacitance and pickup inductance to kill a bunch of the higher frequencies and basically feed the thing all bass and low mids. And that right there is the difference between fuzz and distortion - one clips the low frequencies and the other clips a more full range signal, often with the low frequencies rolled off a bit. If you want distortion out of your fuzz pedal put a buffer ahead of it.

The circuit to which you linked is pretty low-Z also, though maybe not as bad as some. You may need to roll down the Tone control on your guitar to get a convincing effect.

I've heard it said that there are other interactions between the pickups and the fuzz, but these are subtle and device dependent and the "tone suck" is the biggest part.

[cnb]

Dec 2, 2012 13:33:36 GMT -5 ashcatlt said:

Dec 1, 2012 20:41:50 GMT -5 cnb said:

Hey there everyone, new here.
Quick question: would this cause any problems with effects pedals? As far as I know, the Fuzz Face is very sensitive to impedance. What about something like this revolutiondeux.blogspot.com/2012/03/tim-escobedo-tytewadd-fuzz.html?

FuzzFace and the like have a deliberately low in-Z. This combines with cable capacitance and pickup inductance to kill a bunch of the higher frequencies and basically feed the thing all bass and low mids. And that right there is the difference between fuzz and distortion - one clips the low frequencies and the other clips a more full range signal, often with the low frequencies rolled off a bit. If you want distortion out of your fuzz pedal put a buffer ahead of it.

The circuit to which you linked is pretty low-Z also, though maybe not as bad as some. You may need to roll down the Tone control on your guitar to get a convincing effect.

I've heard it said that there are other interactions between the pickups and the fuzz, but these are subtle and device dependent and the "tone suck" is the biggest part.

Cool, very interesting. Thanks for the help!

EDIT: I have started a thread in the effects forum where I will log my attempts at making this and other electrical bits and bobs. Here's a link: guitarnuts2.proboards.com/index.cgi?action=display&board=fx&thread=6658&page=1

[petre]

Sept 30, 2006 23:33:58 GMT -5 JohnH said:

(I am of the opinion however, that it needs another bias resistor , gate to positive, to raise the gate voltage. When I built one exactly to the Tillman schematic, with a J201, it was biased too far ‘off’, and distorted on loud strumming – I’m interested in comments on that).

Hi John,
Do you think your ‘source follower’ circuit can be phantom powered by a standard microphone input the way Tillman suggests on his page?  Maybe more components can be fitted into a larger XLR connector if need be.

Edit: Hopefully any of the technically knowledgeable people can tune in, but can a 9V (or 15V) power adapter be used for the power source as well?  I understand the power adapter will introduce some electrical noise, but is this relevant?  It would be nice not having to worry about unplugging the guitar.  Also I think this can be especially useful if R2 and C3 values can be changed without worrying about power usage to provide a very smooth power on/off operation with a possible stomp switch.

Edit 2: Where would be a good place in the circuit to put in a LED diode and what value.

[JohnH]

Hi Petre and welcome to GN2

All those things should be possible but what is it that you have in mind overall? Is this for a buffer inside a guitar, a cable or a separate box, and what will it connect to?

Cheers
John

[petre]

Hi John,

Thank you for your reply.  I've followed your thread on treble bleed and have played around with your guitarfreak spreadsheet. Reading your posts have been a great learning experience. I currently use RME Babyface (USB interface) and connect my guitar directly to it using 18' cable. I am using Guitar Rig 5 to get all my sounds, and sadly that includes amp simulation because right now I only have headphones and monitors connected in my room.
My goal is to be able to use the volume pot without the treble roll off and have a little more freedom to use different cables (or a wireless system) without having effect on the sound. I like the roll off that happens from the interaction between the pickups, cable and the 470k impedance input on my interface. A treble bleed seems like a possible solution, but using different cables throws off the frequency response curve. If I use a buffer I can adjust the DSP equalizer on the input channel of my interface, but switching to a different pickup will have a different frequency response curve.
I'm thinking if I use your guitar build-in buffer, use a volume pot installed after the buffer, and put a capacitor before the buffer to emulate a cable with 500pf capacitance might be the best solution. I am assuming the volume pot will have to be low resistance (25k?) and I can see having a push/pull pot installed to engage/disengage the 500pf capacitor to get that extra treble. Do you think this is feasible and will get me the intended results?

Edit: What will be the effect of having the volume pot after the circuit?