why not higher than 1M input impedance?

[Alpha579]

lo all,
if you look at most jfet boosters, they use 1M grid resistors, limiting the input impedance to 1M(the input Z of a fet is immensly high), so why not go for a higher grid resistor ie 10M or more...could this create any problems at all, or is it just because 1M is a  more common value...

Thanx,
Alex

[niftydog]

there has been some discussion about this previously - a search will turn up more opinions.

The main thing I would think is that larger resistors mean more noise. (google "Thermal noise" and note the "R" in the formulas!)

Also, larger input impedance can lead to more noise due to microphonics and a greater tendancy for electromagnetic interference to be allowed into the amplification stages.

The effect of greater input impedances is also not linear. That is to say that a 2M impedance is NOT twice as good as a 1M input impedance. You only need to have it as high as needs be. There's more on this concept in this thread. (look at my 2nd to last post, near the bottom)

[bwanasonic]

Note that the AMZ Mosfet boost has a 5M input impedance. But I was wondering a similar thing recently and was also curious about lowering output impedance. Specifically was wondering what effect raising the input impedance and lowering the output impedance would have on the performance of simple JFET boosters. The answer is only a breadboard away, but I haven't conducted the experiment yet.

Kerry M

[zachary vex]

blah blah blah theory blah blah...

my SHO circuit has 5M input impedance at low gains and it sounds deliciously more sparkly than any 1M input impedance circuit.

[R.G.]

blah blah blah theory blah blah...

my SHO circuit has 5M input impedance at low gains and it sounds deliciously more sparkly than any 1M input impedance circuit.

Indeed.

**My** S... er... mosfet booster has a 2.734M input impedance, and it's not only sparkly, it has a full body, a bit cloying on the tongue, but with a clean finish and slight overtones of parsnips and mint.
8-)

Ever'body loves their own babies, I guess. And that's how it should be.

[Tim Escobedo]

I'm wondering of the sparkly tone is a result of some kind of subtle nonlinearities. I've come across clean, "sparkly" tone using circuits with significantly less than 1M input Z. And I've done some clean circuits with input Z >2M without the sparkle.

It's easy enough to use, say, a TL061 with 10Meg input Z as a clean boost and sound quite good, perhaps even biasing it a bit asymmetrically for some sparkle, even though the circuit (with a max gain of about 10) is configured not to clip.

Also, bootstrapping the input may be a nice way of boosting input Z without some of the usual downsides of large resistances.

[puretube]

one more for the bootstrap! (in any topology...)

[Brian Marshall]

one other thing to consider is that most amps have 1m input impedance (well tube amps anyways)  If you want something to have the least amount of coloration a 1m input is probably the easiest way to get there.  The sho, and the mosfet booster enhance trebble specifically because they have a higher input impedance.

[Alpha579]

blah blah blah theory blah blah...

my SHO circuit has 5M input impedance at low gains and it sounds deliciously more sparkly than any 1M input impedance circuit.

well said, i think i like that explanation    i love the video of that @ ur site hehe,

so what'dyu guys think about the trade off between noise and 'sparkle'? 10M in z do alright without to much impact on noise dya reckon?

EDIT: niftydog, just read your explanation, fantastic explanation! easily the easiest guide to impedances ive ever read, it took me a long time to grasp the concept of impedance, shame i didnt have that  

[zachary vex]

one more for the bootstrap! (in any topology...)

i use it in the lo-fi loop junky for the direct guitar.  nice texture, bootstrap.  8^)

[puretube]

a nice combination of SRPP/bootstrap/conventional is used in here:

[spongebob]

Also, larger input impedance can lead to more noise due to microphonics and a greater tendancy for electromagnetic interference to be allowed into the amplification stages.

I can only second that, usually I don't go any higher than 1M because of the danger of cable microphonics, especially with not-so-expensive cables. Bigger isn't always better! :wink:

[zachary vex]

bah.  guitar cables make noise, but going above 1M impedance isn't going to cause more trouble than it's worth.  8^)

[Pedal love]

I vote higher than 1M.

[WGTP]

Could it be that higher impedance/noise is less of a problem with lower gain and more of a problem with higher gain?  

Didn't Jack lower the input impedance (100K) on his AMZ Dist. Pro to reduce the highs and noise?  

[niftydog]

Could it be that higher impedance/noise is less of a problem with lower gain and more of a problem with higher gain?

absolutely. The noise that gets into the system then becomes highly amplified and is more likely the transfer all the way through the circuit and appear at the output at a relatively high level.

Also, running active components at high gain makes them introduce more noise than they otherwise might.

niftydog, just read your explanation, fantastic explanation! easily the easiest guide to impedances ive ever read

cheers!

[amz-fx]

Note that the AMZ Mosfet boost has a 5M input impedance.

Actually it is 10M, but who's counting  

Didn't Jack lower the input impedance (100K) on his AMZ Dist. Pro to reduce the highs and noise?

That was the Fat Gnat...  you design into a circuit the features that produce the result you desire...  the Fuzzface has a very low input Z and it has plenty of highs.  The reason is that even though you are losing some response of the signal on the front end, the distortion mechanism creates a spectrum of higher harmonics and therefore the signal doesn't sound muffled or dark.  This idea would not work too well with a clean boost, for example.

regards, Jack

[bwanasonic]

Note that the AMZ Mosfet boost has a 5M input impedance.

Actually it is 10M, but who's counting  

Yeah that's what I meant!  :lol:

The ouptut impedance part of the equation is something I want to work out, as it applies to booster/buffer circuits. Looking to explore lowered output impedance.

Kerry M

[Paul Perry (Frostwave)]

There is also the question of the impedance of the signal source.
Consider a piezo pickup.. this typically looks like 'quite a few' meg with a couple of pf across it. When you put it to even just 10 feet of guitar cable, the cable capacitance sucks the highs to nothing.
So it is likely to be more vital to put the preamp right at the pickup, rather than worry whether it is 1M or 10M.
Variation in source impedanc eis one more reason (anong with amps, ears adn musical taste) why one size will never fit all.