ad: elecraft

RF Oscillator & Buffer Amp

Discussion in 'Homebrew and Kit Projects' started by KB1LQC, Jul 30, 2008.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: abrind-2
ad: L-MFJ
ad: Left-2
ad: L-Geochron
ad: Left-3
ad: l-BCInc
  1. KB1LQC

    KB1LQC Guest

    OK, So I settled on quite literally redesigning my Local Oscillator for the SDR project. I am now inputing the LO @ about 28 MHz into a 74HC161 but am still having trouble driving the chip. I can get about 20 mV P-P directly out of the oscillator (all measurements are with a X10 probe) and after a buffer amp (2n3904 biased at about 2V on the base with about 10 volts on the collector (2.5K Rc) I can get about 50 mV P-P. Even experimenting with a pot in the bias would still yield no higher output. Cascading two buffer amps resulted in about 100mV P-P. Even with biasing the input of the IC with two 10k resistors at 2.5V, there was not enough to drive it.

    Should I go the route of a comparator such as the scheme that the Softrock 40 employs with the LT1719 or equivalent? Or maybe a different setup of the buffer amplifiers? Or a switch from a 2N3904? I did some research on the switch and found some design articles talking about how to derive the maximum frequency and it resulted in a switching max of about 10 MHz... I do not know if that is in fact the case. I would post schematics but I have to get up at 6am and its 2am... so i hope the description works hihi! Any help would be very appreciated! I am already incredibly happy with the help other forum members have given me on this issue!
    Last edited by a moderator: Jul 30, 2008
  2. VK2TIL

    VK2TIL Ham Member QRZ Page

    Designing stuff like oscillators & amplifiers (and even dipoles :) ) in a forum environment is just too difficult.

    I would make two recommendations; one general & one particular.

    General; get a copy of Experimental Methods In RF Design (EMRFD). It will answer your questions and provide you with a lifetime resource and a great read. No homebrewer should be without EMRFD.

    Particular; I would buffer the oscillator with a FET buffer. Then, by reference to EMRFD, I would design a suitable amplifier stage or stages to get the desired output plus a little margin that will be useful for impedance-matching with a pad.

    It's great to see another enthusiastic home-brewer.

    Have fun!!
  3. AB8RO

    AB8RO Ham Member QRZ Page

    Ok, I agree, but, it's possible to do better than that. The OP should take a picture or two. Maybe one of the schematic. I'd sure like to see much more of these types of discussions and even better if they are in color with pictures!
  4. KI6NNO

    KI6NNO Ham Member QRZ Page


    A 2N2222 should switch to at least 100MHz. It should also drive a reasonable 3v p-p square wave output if you drive it into saturation.

    Here's a data sheet.

    although the 2n3904 should work above 100MHz as well.

    Maybe try a darlington configuration with two NPN's?

    Are you getting 50mV output without the 74HC161 connected, or is the 2n3904 loaded? I'm wondering if you just have too much load on the transistor. (seems odd, but it's late and neurons do strange things when it's nap time...) hihi.

    I'll think about this some more.

    73, Dave
  5. M0DSZ

    M0DSZ Ham Member QRZ Page

    An HF op-amp should do the trick. I have used Elantec amps before but the ones I had in mind are now obsolete. It's easy to set the gain, even into clipping, and the output should drive anything. I don't recall trying a comparator for this but if one has gain enough that should work also. An LM293/393 looks promising.
  6. KB1LQC

    KB1LQC Guest

    I am having some issues with loading :cool:
  7. AI3V

    AI3V Ham Member QRZ Page

    As well as the book EMRFD, I Highly Recommend "Introduction To Radio Frequency Design, By W7ZOI.

    Mr Hayward's description of transistor biasing is the best I have ever seen. It is so clear and straightforward, That even I get it! :)

  8. KB1LQC

    KB1LQC Guest

    I actually Own EMRFD (Proudly wearing out the pages quite quickly :)) but for some reason I am having some problem with this. Maybe its me not quite understanding everything I read in there... Like when I bias a transistor, what type of values am I looking for to get maximum voltage or power gain? It seems that lowering the emitter resistor is one way but when I do it in practice (always different that the mathematical "it should work" :D) the small signal amp turns into a small signal brick :(. It is very nice to be able to converse my questions while going through the book and designing a project. I really am grateful that there are others out there trying to help me through this problem, and I am learning a lot just by trying suggestions, coming up with ideas of my own along the way, experimenting with them, etc. This will eventually be figured out, time and experimentation will tell!
  9. WB3BEL

    WB3BEL Ham Member QRZ Page

    It is impossible to understand from your post the configuration of your buffer amplifier.

    You must write a more comprehensive description or post the schematic.

    I am certain that a bipolar transistor based buffer amplifier will work satisfactorily in this HF application if designed and built correctly. There are lots of other circuits that will also work.

    I would be very careful in using a supply voltage of 10V in a buffer amplifier that will interface with 5V CMOS logic. If it is DC coupled you are looking at a big problem. Do not exceed the CMOS supply voltage or go below GND on the signal pins! Read the data sheet for the maximum and minimum voltages.

    Is the topology a Grounded Emitter amplifier with resistive degeneration and a bypass cap across the emitter resistor? Do you derive the bias with resistive divider to the base from the supply voltage? When you state the base bias voltage is it relative to GND? What is the no signal voltage on the emitter? Where do you take the output signal from in your buffer amplifier? Is your buffer amplifier DC coupled or AC coupled?

    What current is flowing in the base with no input signal? What current is flowing in the collector with no signal? What current is flowing in the emitter with no signal? Does it agree with your calculation for the DC bias?

    What are the resistor values that you have chosen for the bias and are they impacting the AC signal as you expect? What values have you chosen for capacitors in the circuit if any?

    Good luck with your project.

    -Harry WB3BEL
  10. KL7AJ

    KL7AJ Premium Subscriber QRZ Page

    Whoa! You're bias voltage is WAY too high....2 volts will drive that right into saturation. I'd try it with JUST a smidgen over .6 volts. You can try a current drive bias by a simple dropping resistor from your VCC, on the order of 200K to 1Megohm.

Thread Status:
Not open for further replies.

Share This Page