single transistor treble booster not working right

[C Bradley]

I'm trying to build a single transistor treble booster and things aren't going all that well. I'm getting some distortion of the signal when bypassed (SPDT switch on the output) and it doesn't have no where near the gain or volume I'm looking for. I'm using three resistors and three caps. 12k collector resistor, 1M from 9V to the base, and either 1k or 470 ohm for the emitter. The transistor is a 2N2222. I've got a 4.7uF emitter bypass cap, a .022uF input cap, and a .002uF output cap. Can anyone help me figure out what's wrong here?  

Chris B

[RDV]

What circuit is this?

Regards

RDV

[gez]

I'm trying to build a single transistor treble booster and things aren't going all that well. I'm getting some distortion of the signal when bypassed (SPDT switch on the output) and it doesn't have no where near the gain or volume I'm looking for. I'm using three resistors and three caps. 12k collector resistor, 1M from 9V to the base, and either 1k or 470 ohm for the emitter. The transistor is a 2N2222. I've got a 4.7uF emitter bypass cap, a .022uF input cap, and a .002uF output cap. Can anyone help me figure out what's wrong here?  

Chris B

I take it that your circuit isn't true by-pass? How have you got the SPDT wired up, i.e. is the input always connected?  Does the distortion only occur when by-passed, or do you get it when the booster is kicked in?

The low gain is probably as a result of using the 1M resistor.  The collector current is probably too low meaning that the gain of the circuit will be poor.  I don't know typical gains for this trannie, but you could try recucing the 1M and change the other resistor values to compensate, or stick in a higher gain trannie.  Divider bias would be a better option IMO.

[gez]

Sorry, just noticed that you said the switch is on the output!  How have you wired it up? (just to confirm).  If the switch is dodgy it may be feeding back some of the signal to the input causing distortion which in turn is bleeding through to the output.

Edit: If you increase the value of the emitter bypass cap it might help things.

[C Bradley]

What I've attempted is to build a circuit like the Electra distortion with the 2N2222 transistor. I had a 47k resistor on the collector at first, but it was even lower output than it is now. The next lowest resistor from a 1M that I've got is a 220k. Is that low enough? My voltages were like this:

Vc = 8V
Vb = 0.6V
Ve = 0.02V
Vcc = 9.36V

It would seem to me that it's biased correctly, but then again, why is the collector voltage so high? Shouldn't it be closer to 5V? The switch is wired up with the center lug going to the output jack, outer lug 1 goes to the effect's output and outer lug 2 goes back to the input jack. The treble loss is terrible when it's bypassed and there's some nasty clipping, like it's not biased right (this is still when bypassed!). The effect sounds OK when not bypassed, except for a lack of gain. Board space is an issue, which is why I didn't go with the voltage divider bias, but I might be able to squeeze in another resistor if I try. I could also go to a PNP 2N3906 and reverse the supply if that would help.

Chris B

[gez]

What I've attempted is to build a circuit like the Electra distortion with the 2N2222 transistor. I had a 47k resistor on the collector at first, but it was even lower output than it is now. The next lowest resistor from a 1M that I've got is a 220k. Is that low enough? My voltages were like this:

Vc = 8V
Vb = 0.6V
Ve = 0.02V
Vcc = 9.36V

It would seem to me that it's biased correctly, but then again, why is the collector voltage so high? Shouldn't it be closer to 5V? The switch is wired up with the center lug going to the output jack, outer lug 1 goes to the effect's output and outer lug 2 goes back to the input jack. The treble loss is terrible when it's bypassed and there's some nasty clipping, like it's not biased right (this is still when bypassed!). The effect sounds OK when not bypassed, except for a lack of gain. Board space is an issue, which is why I didn't go with the voltage divider bias, but I might be able to squeeze in another resistor if I try. I could also go to a PNP 2N3906 and reverse the supply if that would help.

The collector current is definitely too low by the look of things, which is why the collector is biasing so high and it's probably why the gain is so low.  The base current is too small due to the 1M resistor.  If you have a tester on your multi-meter, check what the gain of the trannie is and this will help you to work out resistor values.

The treble loss in bypass 'might' be due to the input impedance of your circuit being low.  The signal is always connected to the input, and the input's impedance is in parallel with the 1M bias resistor, so it may not be as high as you think, especially as you've bypassed the emitter.  All this means that you're probably loading your guitar's high end.  Better to use a DPDT and make sure the input is disconnected when the effect is bypassed.

Not too shure why you're getting distortion.  I'm just off to bed now, but I'll sleep on it!  

[gez]

Excuse me if this is really dumb (bear in mind I'm about to turn in so my brain isn't that efficient at the moment) but are you sure you're not confusing the bypassed signal with the FX signal?  With the voltage readings you've given it's quite possible that you're booster is clipping - maybe it's this you're hearing in 'by-pass mode'.  This wouldn't account for the treble loss though...oh well, it's definitely time to turn in!

PS yes the collector should bias up around the 5V mark.  Try reducing the 1M and re-jig some resistor values...night night!  

[gez]

When base currents are too low, the amount of holes that are temporarily 'liberated' at the base-emitter junction isn't enough to create the slight charge which attracts other electrons from the emitter towards the junction.  As a result gain tends to be low.  Going by your figures it's about 12 at the moment.

You could try disconnecting the emitter resistor, wire up a trimpot (croc clips and a breadboard?) to the emitter, reduce the 1M to 220k (if that's all you've got), try something like a 10k resistor at the collector and adjust the trim.  This 'might' give you the correct bias at the collector (that's the problem with this type of bias), and if it does measure the pot and work out the correct emitter resistor to use.  

If you can get away with it, a divider would be the better option.  Either way, it sounds as if the input impedance is too low so a DPDT would be the way to go - hopefully this will sort out your other problems too.

[C Bradley]

I tried a 220k in place of the 1M resistor last night, and the distortion I was hearing on the bypass is now gone. The gain of the effect is still pretty low, and just manages to put a slight grit on the signal, not what I wanted. I guess I'll have to try a 100k instead of the 220k now. :?

I want something similar to the Brian May sound, like a Rangemaster or some such circuit.  

Chris B

[bwanasonic]

I want something similar to the Brian May sound, like a Rangemaster or some such circuit.  

IME- a Rangemaster  type circuit really needs an at least slightly overdriven tube amp after it do that trick. Although Brian May did use that modified car radio amp to good effect. I have found my Rangemaster clone does not work the same way with any two amps (which isn't a bad thing).

Kerry M

[C Bradley]

I messed around with the circuit a little more last night, and got a decent tone out of it (still low output). I changed the 220k biasing resistor back to 1M, and changed the output cap from .002uF to 1000uF!! I also changed the emitter resistor to 2.2k. According to my Circuitmaker software, that gives me about 1.5V on the base, 1V on the emitter, and about 5.4V on the collector.

I'm still going to be tinkering with it, but it's much better now. I think maybe I expected too much.

Chris B

[gez]

I changed the 220k biasing resistor back to 1M, and changed the output cap from .002uF to 1000uF!! I also changed the emitter resistor to 2.2k. According to my Circuitmaker software, that gives me about 1.5V on the base, 1V on the emitter, and about 5.4V on the collector.

Unless you know the exact gain of your trannie, Cicuitmaker is a bit of a waste of time, because of the nature of the biasing method.  Use a voltmeter to influence your choices if you don't have a meter that measures transistor gain.

1000u - way to go!!!!    

[C Bradley]

OK, with a voltmeter I'm getting 8.1V on the collector, 0.69V on the base, and 0.13V on the emitter. So it appears I've still got biasing problems.

I'm seriously considering the fixed bias supply now. Hope I can fit it on the board...

Chris B

[gez]

You could try the following to get the bias right:  

Use a 9k1 for the collector resistor, 1k for the emitter resistor and wire up a 1M pot in series with a 'stop resistor' for the base resistor.  Measure the collector voltage as you turn the pot, till you get around 4.5V.  Measure the pot resistance, add that to the stop resistor's value (look at a data sheet to calculate what the max base current this trannie can take and sort out a resistor so that the current doesn't exceed this), then wire in the nearest value resistor you've got to replace this chain.  By-pass the emitter with a cap and Bob's your uncle (please don't anyone start THAT post off again!).

[C Bradley]

I've got it working now! Lots of gain now! My transistor was in backwards, and after I figured that out, I fixed that and re-biased it. I'm still having a problem with the dynamics. If I really hit the strings, the sound gets buzzy for a second and then becomes clearer. It sounds like my transistor is getting saturated during the attack, or it could be going into cut-off? Whatever is happening, I'm getting some nasty clipping on the attack.

I used a 10M resistor inplace of the 1M to bias the transistor, but I think I'm going to have to go to Radio Shack and look into getting a 2.2M resistor or two for this type of thing.

Chris B

[gez]

Try removing the emitter bypass cap, or partially bypass the emitter.  This will lower the gain.