Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

JFET, Mosfet, BJT Design Question

  1. Apr 6, 2016 #1
    In my electronics class we are learning BJT, Mosfet, JFET theory. My gripe at the moment is that when teaching this you are taught how to troubleshoot and solve for circuits already made. This means you are always solving for various voltages, currents, Gm etc etc. With no previous explanation of how We did a lab today where I had 2 design a 3 channel Jfet amplifier using a 2n5459.

    I sort of have an idea but I am mostly lost.
    This is school work, so don't give me answers just nudge me in the direction.

    I know from the data sheet:

    VP is -2 to -8V
    Idss is 4 Ma - 16 Ma
    Gfs - 200 umhos - 600 umhos

    I tested my jfet and found my Idss to be 13.46 mA using a test circuit
    I also found Vp to be -4.8V
    I plotted and graphed a detailed Vgs Vs ID plot.

    Other info needed for design of circuit:

    - Av 3-4
    - Input signal 200mBpk
    - Power supply 9V (Can be 12V if needed due to high Idss)

    Things I think are correct

    - I know that if I make this a voltage divider circuit R1/R2 ratio will have to give me my desired Vp. I also know from seeing circuits these resistances are usually around the 1M ohm area.

    - I know that Vds is a value I must choose to try and solve for.

    - I I feel like Rs doesn't really change Vout alot but I could be wrong. It would however at to total current flow from Drain to Source.

    - I know that Vds must be in the active region, not the ohmic region.


    - I dont know which thing to solve first. I have no intuition of where to start because Im not really 100% sure how R1/R2 (Zin) affect Zout or any sense of order because I've never had to think of it this way.

    - I've tried analyzing it as DC, as the small signal model used for AC analysis, and as the graphed parameters.

    - Can anyone explain or help me get started? The only thing I have figured out is that I can solve GM. I'm not sure which variables are things I choose, how to pick a number for R1/R2 or how to even manage to find values with Av needed.

    Thanks in advance.
  2. jcsd
  3. Apr 7, 2016 #2


    User Avatar
    Science Advisor

    Start with the input section. You know what an OpAmp input section should look like?
  4. Apr 7, 2016 #3
    No :-(
    We've done diodes, BJT's, Jfets and I've gone ahead a bit and looked at mosfets because they are similar.
  5. Apr 7, 2016 #4


    User Avatar
    Science Advisor

    Start by reading http://socrates.berkeley.edu/~phylabs/bsc/PDFFiles/bsc5.pdf [Broken].
    Last edited by a moderator: May 7, 2017
  6. Apr 7, 2016 #5


    User Avatar

    As an example - in this last sentence you have used three abbreviations (symbols: GM, R1/R2, Av). In some cases, the reader can guess what the symbols mean.
    However, if you wait for a really helpful answer you must try - from the beginning - to make your question clear in order to avoid misunderstandings. It is good and helpful practice that in each technical text the meaning of all symbols is explained. Such an explanation could (and should) be supported by a circuit diagram or any other relevant graph.
  7. Apr 17, 2016 #6
    Sorry for not being clear,
    For example I'm trying to just make a JFet amp circuit. Its biased in the middle just so I can make sure it works.

    Attached is my calculations.

    Output measured is
    Vd = 7.44V
    Vs = 4.73V
    Vg = 2.238V

    For Ac:

    Originally I wanted to use 200mVp but My function generator doesn't go that low so I used 350 mVp @ 1 kHz
    Input = 0.355 Vp
    Output = 0.705 Vp
    Av = 1.98

    Things I'm unsure about:

    - In my calculations I went for an AV of 3 but ended up with 2. Is that just that the gate isn't creating the depletion region as expected?
    I previously had done some work to find Idss and Vp of the same transistor I used.

    -If I wanted to hook up a tiny 8 ohm speaker to the output to hear my 1k signal would anything else be required aside from a decoupling cap?

    - I was told its good practice to put a parallel capacitor to Rs however I don't really understand what it does. I get that it blocks the DC and lets AC go to ground bypassing Rs but I don't understand the influence that has on Vout.

    -Vd calculated was 6 measured was 7.44V is that acceptable?

    -My next project is to build a VU meter using op amps as comparitors. If I were to attach that after to a jfet/mosfet/bjt amp where in theory would you attach that? (I don't necessarily have to, I'm just curious)

    Thanks in advance

    Attached Files:

  8. Apr 17, 2016 #7


    User Avatar
    Science Advisor

    Put a capacitor across Rs. Any value above 1uF. Get the polarity right if you use an electrolytic ('+' to the FET Source, '-' to Ground). And we can continue from there.

    Looking at the datasheet, Vd of 7.4V is reasonable due to the wide tolerance of the transistor characteristics.
  9. Apr 17, 2016 #8
    Thanks for the reply. I put a 10uf cap there. I just don't understand what it does.
  10. Apr 17, 2016 #9


    User Avatar
    Science Advisor

    Did the AC gain change any? If so, what is gain with the cap in there?
  11. Apr 17, 2016 #10


    User Avatar
    Gold Member

    Did the professor cover amplifier configurations? Take a look at the difference between the common source and common source with degeneration, and compare your circuit while looking at the small signal model compared to the operating point or (DC) biasing. What happens to large capacitors in each of these models? Recall that [itex]\tilde Z_C = {{1} \over {j \omega C}}[/itex]
  12. Apr 18, 2016 #11


    User Avatar

    Have you ever heard about feedback and its consequences?
    Depending on the feedback network (in the prent case, a simple Re||Ce combination) we must discriminate between DC and AC (signal) feedback.
  13. Apr 23, 2016 #12
    Thanks for all your replies guys/gals.
    Appreciate it.

    LvW + JP - I do know the consequences. I'm still unsure in a BJT, Mosjet, Jfet circuit how I would calculate what capacitor to use
    as a decoupling capacitor.

    I know RC makes up the time constant but which resistor values in say a CS circuit do I use to calculate the time constant?
  14. Apr 24, 2016 #13


    User Avatar
    Gold Member

    Does your circuit have bandwidth or stability requirement? There's a technique called the Zero Value Time Constants that might be helpful for you. I used an online search on this technique and found seemingly good slides from a university on this technique [UNM's, UCB's]. It's an approximation strategy, but if you want definitive answers that take a while, then I would put a hypothetical/test capacitor between the stages, derive the transfer function, and look at the bode and Nyquists plots.

    Am I answering your question or does this make sense?
  15. Apr 25, 2016 #14
    Just audio frequency range, so 20hz-20khz.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted