2sk170 vs 2sk30atm What I wrong?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Violinist

Well-known member
Joined
Oct 4, 2011
Messages
80
Location
België
Hi

Always with the experimentation spirit, I decided to modify my microphone represented here:

http://www.groupdiy.com/index.php?topic=44096.msg648089#msg648089

Instead of a 2sk170, which has an input capacity of 30pf, I wanted to put a 2sk30ATM, which it seems to me very similar to the 2sk170 but with an input capacity of 8.5 pf.
I supposed that with less input capacity, I should get an higher level output signal, rather than with the 2sK170, ameliorating the S/N.

Perhaps my argue is wrong, and I ask you please to correct me.

however once I changed the FET, the microphone still working, but has an output level dramatically diminished.

I suppose there is something wrong, thus, in my assumption.

Can you help me please?

 
HellfireStudios said:
Did you bias the new FET?

-James-

I put it in place of the previous one, taking into account that its Drain and Sources are inverted respect the ones of the 2sk170
 
If you take a car, and you swap the engine, it's very likely that car won't function any more, without necessary adjustments surrounding the engine.
If you take a transistor, be it a JFET or anything else, and swap it for another part, you are in effect doing the same thing, unless you carefully perform said transplant, knowing the circuit operation well in advance.

By biasing, HellfireStudios is referring to changing the voltages and/or currents at key points in the circuit that ensure it's correct operation.
Different parts like different biasing.

Have a good read of something like this:
http://www.electronics-tutorials.ws/transistor/tran_5.html

Another thing to try is taking the schematic, putting it into a simulator (LTSpice is free and exceptionally good for this type of thing), and seeing how swapping different JFETs changes performance.
You might be able to see what's going on this way more clearly.

Simply swapping a transistor, unless that transistor is from a family of transistors that are functionally identical, or, at least, the circuit is self biasing; won't work, or, if it does, can/will produce greatly decreased performance.
 
It's very simple:  it's the difference between the admittance of the two types:  1.2mS min for 2SK30ATM vs. 22mS typ for the 2SK170.

The gain of the new JFET is 1/10 the one that is in there....
 
Matador said:
It's very simple:  it's the difference between the admittance of the two types:  1.2mS min for 2SK30ATM vs. 22mS typ for the 2SK170.

The gain of the new JFET is 1/10 the one that is in there....

right but the gain is 1
thus the admittance should not influence it. here, it seems it attenuates. Am I wrong  ?
 
The 2SK30A is often used in a circuit like this, so it must be the biasing.
Did you compare the drain voltage with the 2SK170 circuit?
By the way: the input capacity plays a role in the S/N ratio, but also the self noise of the FET is an important factor.
(Although the 2SK30A isn't really bad.)
 
RuudNL said:
The 2SK30A is often used in a circuit like this, so it must be the biasing.
Did you compare the drain voltage with the 2SK170 circuit?
By the way: the input capacity plays a role in the S/N ratio, but also the self noise of the FET is an important factor.
(Although the 2SK30A isn't really bad.)

Thank you so much Ruud, This night at 4 o' clock C.E.T. (it was very hot this night and I coudl'nt get sleep) I checked the power supply and the bias.
I'm vitcim of a coincidence :-/ the audioboard to which I normally connect it, got an issue.
Now, (current time), I checked it on my Mixer which has also the Phantom, and the mic works fine.

The only one point is: what I expected there is not.
I mean, having 1/3 of the input capacitance, I expected an increase of signal level, which I have not. It is like before the intervention, but with another FET. This was disapointing a little. So much work and troubles for nothing. Ok it's a learning processes however :)

About the FET Noise Figure, checking the datasheet they are identical: 0.5 dB Typ
 
Violinist said:
right but the gain is 1
thus the admittance should not influence it. here, it seems it attenuates. Am I wrong  ?

Admittance is admittance:  it is delta change in drain-to-source current versus change in gate-source bias.  All other things being equal the gain will reduce.

The circuit in question is an equal voltage, unequal output impedance signal splitter.  If you take a high admittance device and run a simulation, I think you'll see the circuit approaching a unity gain buffer.  If you then substitute something with a lower YFS I think you'll see an output waveform much lower than unity.
 

Latest posts

Back
Top