Newbie needs help increasing gain in a Class A amplifier circuit

Other Hardware General Electronics
1

Hi! I have started working on my tube amp (Bugera V5), and am having a lot of fun doing it, and in the process, am even learning about transistor amp circuits.

You see: my amp comes with a digital plate reverb (Belton Delay Module), which generally sounds pretty good. But it has a tail of high-end sizzle that becomes very noticeble if, like me, you replace the stock speaker of your amp with a new speaker that's more responsive to high frequencies.

So, I decided to look a the amp schematic further (I have included the schem for you to peruse), and combined with that and doing research on the web for general transistor amp circuits, I figured out how to take the offending high frequencies out of the reverb: I installed a 10000pf cap in parallel to the 120k Collector resistor coming from B+. This has definitely brought the highs down, to a very warm and realistic sounding reverb level. But, as all you-who are much more experienced with common emitter circuits than I am-could have predicted, it also seems to have brought down the overall gain going through the transistor, and therefore the reverb has taken a hit in volume.

No problem, right? Just figure out now how to bring the overall volume of the transistor back up, right? That's what my basic Audio Engineering experience tells me. I've done some research on Common Emitters, and based on what I have gleaned, I see that one of the most basic ways to increase the gain level of the circuit is to include a capacitor in parallel to the Emitter resistor. If you look at my schematic, you'll see that there is already a cap there (electrolytic, 100uf, 25V).

I was wondering this:

  1. Am I on the right track as far as a strategy goes (is modifying the Emitter Cap the best way to bring up the gain level of this circuit?

  2. If so, what overall value of new cap do I need in order to bring the gain level back up to where it was before?

  3. What is the overall method/formula one needs to employ in order to figure out such things?

Thanks for your time.

Go Arduino!!

Bugera-V5-schematic-transis3.jpg
Bugera-V5-schematic-transis3.jpg932×657 231 KB

2

No.
R27 controls the gain. The capacitor controls the bandwidth.

By the way once you put an emitter resistor in it is no longer a common emitter circuit.

3

Thanks for the quick reply!!

Are you meaning to say: don't change the cap (C20), but instead change the resistor (R27)?

If so, do you know by how much?

4

Are you meaning to say: don't change the cap (C20), but instead change the resistor (R27)?

Yes indeed that is what I am saying because altering that cap does not increase the gain.

If so, do you know by how much?

Very difficult to say because that schematic was so stupidly large it is hard to make things out.
Did you read the how to use the forum sticky that talked about posting pictures.

5

Sorry, I did not. I will read and comply

6

There-the original pic was 367K, and I brought it down to 68K. I hope that's better.

7

claypruitt:
There-the original pic was 367K, and I brought it down to 68K. I hope that's better.

So have you clicked on it to see?

It is exactly the same size as before you have just increased the compression so that now it is even harder to read.

It is the number of pixels not the file size that is important. Make it no more than 1000 pixels in any direction.

8

K-I misunderstood you before. I initially included the schematic for the whole amp because I thought some people may need to see how the reverb circuit related to the whole amp.

9

Basically what you have here is not a common emitter circuit but a Class A amplifier.

The gain will depend on the gain of the transistor and the component values.
Do you know what gain you want?

This page tells you about it.

10

If you don't want to learn the theory behind transistor amplifier design (clearly explained in, among many other places, "The Art of Electronics" by Horowitz & Hill), it is easy to simulate the circuit with quite reasonable accuracy using the free LTSpice simulator. Then you can change anything you want and observe the consequences.

11

I don't like that circuit, the transistor's biased all wrong, its relying on the current
gain of the transistor being a constant which is never a good idea. Normally you'd
bias a class A with two resistors to the base so its DC bias voltage is well defined.

12

R27 controls the bias point of the circuit, and it affects the gain.

R18 only affects the gain. Reduce this to increase the gain.

C20 will not change the gain if you increase the value.

As pointed out, that is not a stable way to bias a BJT (Bipolar Junction Transistor). What you want to use is called voltage divider emitter stabilized bias.

A voltage divider provides a voltage to the base of the transistor. A resistor between the emitter and ground stabilizes the bias point, as the voltage across that resistor is equal to the voltage on the base minus 0.6V.

Then you use a capacitor to bypass the emitter resistor for AC signals, boosting the gain.

Here is an example:

From question 20:

This one page gets right to it.
http://smendes.com/el22c/exer1.htm

This is still classed as a common emitter amplifier, because the emitter is the common point between the input on the base and the output on the collector.

Another website describing BJT bias:

13

You don't want a capacitor from ground to the emitter in an audio amp, severe
distortion will be the result. That's only useful in narrow-band amplifiers where
the harmonics are irrelevant and filtered out anyway.

The orginal circuit had an R and RC link from the emitter setting the
DC and high frequency gains (but preserving the shunt-feedback to
keep things more-or-less linear).

These days its all opamps everywhere for audio signal processing.

14

Normally you'd
bias a class A with two resistors to the base so its DC bias voltage is well defined.

MarkT-I thought that that was what R29 and R31/R30 were doing: functioning as a, more or less, biasing voltage divider to the base, albeit in a highly irregular way (with the caps inserted, and all). That's not correct?

A voltage divider provides a voltage to the base of the transistor. A resistor between the emitter and ground stabilizes the bias point, as the voltage across that resistor is equal to the voltage on the base minus 0.6V.

Then you use a capacitor to bypass the emitter resistor for AC signals, boosting the gain.

Polymorph-Isn't that exactly what C20 is doing-bypassing the emitter resistor? I know that R18 is on the same line-are you saying that that inclusion alters C20s ability to bypass R27?

15

Yes, R18 provides partial AC bypassing to help keep the amplification more linear. As MarkT says, merely bypassing the emitter resistor will cause large signal distortion. That is because with full emitter bypassing, gain is dependent on the internal Re (emitter resistance) which is nonlinear and dependent on emitter current. But emitter current changes when a large signal is present.

As long as R18 is much larger (greater than 10 times) Re, it will be fairly linear. The gain is Rc/(R18//R27) where // means "in parallel" and Rc is the equivalent resistance of all the stuff attached to the collector.

Look carefully at the circuit I posted. Read those links I gave carefully. Can you not see the difference between those and the schematic you gave? It isn't about the number of resistors, but their configuration.

The two resistors on the base of your circuit are 2 instead of 1 only because R30 and C18 are a lowpass filter meant to act as noise bypass, to prevent noise on the 15V line from being amplfied by they transistor. Forget those and only see R31 to the base. The bias point depends very much on the exact Beta of the transistor. Which will vary with time, temperature, manufacturer, batch, etc.

Personally, I'd just use an Op Amp. You can build it as a tone control with bass and treble to control the frequency response.

Half of the stereo circuit on page 4 will do:

TI released a slightly abridged version of the Second Edition of Op Amps for Everyone for free:
http://www.ti.com/lit/an/slod006b/slod006b.pdf

The Third and Fourth editions are print editions and are quite good.

I'm not surprised that your 10nF (aka 10000pF) capacitor kills the gain. The RC time on that puts the approximate -3dB point at about 800Hz. I mean, the coupling capacitor C16 is only 22nF, and you are bypassing the output with 10nF.

Actually, C16 seems a little low, but it is meant, I think, to keep bass frequencies out of the reverb.

16

Please note that the circuit on page 4 of that PDF is really a single Op Amp tone control. The first Op Amp is merely a buffer.

Here is another page talking about tone controls which includes a single Op Amp tone control: