I've made small amplifier for hand-made electric guitar - it is just two transistor cascades - and I've found one curious effect when increasing amplification. Some annoying noise (just typical electric interference from nearby devices) appears if I raise amplification on the last cascade, but not if on first.

And I don't understand why is it so.

Now here are details:

![two-stage transistor amplifier](https://sun9-72.userapi.com/c855432/v855432016/12a074/YS7V_Y1QWsQ.jpg)

The amplifier is necessary because my hand-made pick-up has comparatively small coils (in comparison with normal electric guitar).

Here you see the amplification of each transistor is determined by relation between resistor in collector and in emitter.

E.g. with values shown, first transistor amplifies 1100/39 = 28 times, while second only by 1100/220 = 5 times.

If I increase amplification by changing 220 Ohm resistor to 68 Ohm - i.e. roughly 3 times - I get the noise described above - it occurs when the hand is close to or touches strings.

However there is no such noise if I raise amplification on the first transistor instead. I.e. I changed 39r to 4r7 and got louder but still clean sound.

Well... How does it happen?

Below is the picture of the whole assembly for better idea - power is from two batteries, to exclude noise from power supply...

transistor amplifier on hand-made guitar

  • 3
    \$\begingroup\$ The first stage introduces more noise into the signal which gets amplified by the second stage. So you tend to want as much gain as possible in the first stage, subject to other constraints (linearity, bandwidth, etc). \$\endgroup\$
    – DKNguyen
    Oct 14, 2019 at 18:55
  • \$\begingroup\$ use thus for last stage tinyurl.com/yxm9hegn \$\endgroup\$ Oct 14, 2019 at 19:04
  • \$\begingroup\$ @SunnyskyguyEE75 thanks - I'm aware of such pulling-up base from collector though not quite sure why it should help in this case, honestly :) Anyway thanks - I never thought falstad simulator allows schematic sharing via url :) \$\endgroup\$ Oct 14, 2019 at 19:12
  • 1
    \$\begingroup\$ You could wrap another set of windings around the first, the same number of turns. Go around in the other direction this time however, then wire these in reverse polarity in series with each first coil. This then becomes a "hum-bucking" pickup. Coils electrically opposite = EMI cancels out, and oriented magnetically opposite (while wired opposite) = signals add. It won't help with pure amplifier noise, but it will reduce 50/60Hz mains hum. \$\endgroup\$
    – rdtsc
    Oct 14, 2019 at 19:28
  • \$\begingroup\$ @rdtsc thanks for suggestion - I do know about hum-bucking pickup but it never occurred to me it could be wound that way. Though this specific noise doesn't seem to occur from pickups, you see - otherwise it would be amplified by either stage. \$\endgroup\$ Oct 14, 2019 at 19:35

1 Answer 1


Q1 does not clip, but Q2 does clip from insifficient base bias current and Vc rides too close to the 3V rail.

There are a lot of interactions with changing bias resistors.

  • Q2 has RC1 as the source impedance 1k1
  • The input Rbe diode AC impedance lowers with rising base current from 2.6uA to 24uA (Rb=680k to 300k, Rcb= 100k) and drops from 8k to 1.1k thus attenuating the Q2 base voltage 15% to 50%
  • raises Ic = 1.6mA =100(3-0.6V)/150k to lower Vc=1.6mA*1k1= 1.8Vdc from 3 to 1.2V to prevent clipping
  • Reducing Rb alone however causes more positive base current thus lower Rbe thus more negative swing gain and THD
  • overall Neg feedback V gain is ~ 22 as shown with no impedance gain 1k1 in:out

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.