I don't even know why you would think this??? Each transistor will be biased from the supply to suit its own needs. There is no magic formula that will tell you what voltages to expect on each transistor - that will depend on the biasing resistors in each stage. The only thing that you would expect to be consistent is around 0.6 - 0.7 Volts between each base and emitter but then they could be almost anywhere with respect to the supply.

What are you trying to do? Are you fault finding - if so it seems like you are flying blind without understanding how these circuits work. I'm not trying to be rude but just responding to the question. It seems from the question you would be better off getting someone to look at this circuit for you.

Nope, just wondering. I mean it makes sense though. This is within the context of a simple stompbox circuit. All of the voltage comes from the same place, the source, and follows the PCB which branches off at each transistor. If the supply is putting out 9v and the first transistor pulls 9v then how could any voltage even continue down the line to the next gain stage? Everything that the supply is producing is going through the first transistor straight back to the supply again completing the circuit. The only way that I see it possible to draw 9v over three parallel gain stages is at an increase in current. So do you use more supply current from having more gain stages? That makes sense to me.

Can you post the circuit or a link to where it is?

The full 9V is available for each stage - you could view it as them being connected in parallel on the supply. Thus, as you surmised, three identical stages would indeed draw three times the current of a single stage. In a three stage amp. though you would be very unlikely to find each stage identical and the total current would depend on the design of each stage. Note also that three transistors do not necessarily imply three gain stages. I have to say that this is a very simplistic summary though. As Boswell said a schematic would help any explanation.

Is this something to do with the DIY stompboxes you bought recently?

Well, it's not for a specific project. The projects that I am building aren't complex enough to have three gain stages yet, I am just hypothesizing. So since the current draw increases threefold and the circuit resistance stays the same, according to Ohm's law the voltage draw from the power supply would increase threefold as well. Is this also true?

Ohms law is V = I*R

Three resistors in parallel will all see the same voltage across them, but will have different currents through them (unless they are exactly the same-in which case each resistor will see 1/3 the total current) .

Each transistor in the circuit I think your talking about would be able to have 9 volts on the collector.

Guitarfreak, post: 351924 wrote: Well, it's not for a specific project. The projects that I am building aren't complex enough to have three gain stages yet, I am just hypothesizing. So since the current draw increases threefold and the circuit resistance stays the same, according to Ohm's law the voltage draw from the power supply would increase threefold as well. Is this also true?

Guitarfreak,

You seem to be repeatedly making some rather wild assumptions with your theories. I have highlighted the glaring one in the above quote. If you had three identical stages and disconnected one of them, then the resulting current draw would be reduced by 1/3. How on earth you can imply the circuit resistance stays the same I do not know.... BTW you do not "draw" voltage from a supply, you draw current.

Just another quick thought. As you are aware of Ohm's Law, I would also suggest you google Kirchoff's Law which might illuminate all this a bit more for you.

Will research Kirchoff's Laws, thank you.

I built my circuit today and unfortunately only got about 30 seconds before work to plug it in and play with it. I think there may be a problem because when the circuit is active it makes my gain channel sound like I am playing on the neck pickup. I thought at first that this was because the level was overloading the input of the amp but when i turned the trimpot down it just got quieter. My assumption as of now is that the problem is the output buffer. the volume trimmer is only 15k turned up full and I think I may raise it to 100k. Any other ideas?

And what are you going to do about the difference in voltage drop between germanium and silicon transistors? Which one do you think would work best for your application? Little metal cans or black epoxy? Do you know that some transistors can become micro-phonic? Maybe you actually want to create a guitar amplifier using all thermistors? That way you can tell how hot you are playing. All you have to do is then design a speaker which can also provide a visual display of your playing temperature.

I prefer toasted bagels with blue cheese
Mx. Remy Ann David

Well I think I'll ignore Remy's totally spurious response and go on to suggest that questioning a value for a trimpot without posting any form of schematic is completely unanswerable! It occurs to me that you do not have the knowledge to actually design a gain stage so I can only assume you are using a third party schematic. Does copyright prevent you from posting it?

To be quite honest the sort of questions you are asking give so little information and display a distinct lack of knowledge of the operation of what seem to be quite simple circuits. While I am trying to help, the lack of information you are giving is beginning to test my patience. I appreciate you are trying to learn and that is the only reason I am here but there comes a stage where you really have to help yourself and stop yourself stumbling around quite so blindly.

I am talking about something else now. I got tired of you grilling me about not knowing about gain stages so I moved on to talking about my mail order project but didn't feel like making a new thread. The project is a BYOC confidence boost and the PDF is available for free download there with the schematic. Now go and educate yourself and return with a renewed attitude, or stop posting so I can get information and advice without any unnecessary attitude.

If you read my posts again you should realise that there is no "attitude", only what I intended to be helpful pointers in response to the questions you asked. It is neither my fault nor my attitude that each of your questions clearly showed some fundamental lack of understanding.

Would you be better advised if this was not pointed out to you? Probably not if I did not bother trying to point you in the right direction as well.

Again I will say asking for suggestions regarding your circuit is hopeless without even telling us what the circuit is. What do expect me to say? Suggesting you adjust R3 from 10k to 22k is hardly possible is it? If you really want some useful help then you will need to help us to understand what it is you are trying to achieve.

Sorry if you think this post represents continued "attitude" but please try to see what I have actually said and that I have tried more than anyone on this thread to help you.

R3 in that schematic is 1M, I am talking about the volume trimpot which came with it and is rated 15k. After getting a little more time to screw with it today I think that there is a problem with the circuit or how I built it. It is very gainy, and sounds almost 'fuzz' like. Under the tone there is crackling and intermittent volume swells. I checked my soldering today and it seems to be inconclusive. All joints seem solid and nothing seems to be touching or grounding out. Here are some pictures I just took.

Is this the booklet of the confidence boost unit?
Secret

It seems to be one of those designs that's unduly sensitive to battery supply voltage, but it's worth a go.

Guitarfreak,

You really don't make it easy! :

I agree with Boswell that this design is certainly not going to yield consistent results and if just for the op-amp, you will need to make sure the battery is always healthy. As to the buffer stage, bearing in mind that it seems they supply almost any transistor for this and also with transistor gain spreads, the biasing is going to be very variable. If this "confidence booster" is intended to boost your confidence in building something that works (I can see little other purpose to the circuit) then I don't think it's a great design.

The first thing I suggest checking is the voltage at the transistor collector. This really should be sitting around 5 V or so (it's probably nearer 3V) but this is likely to vary significantly. Also, the transistor output is surely going to be loaded by the 15k trimpot, so it could be a good idea to increase this to the 100k you said. That is assuming you are taking the output to a high impedance guitar amp. Many mixing desks have relatively low input impedance so this would tend to load the output anyway.

(Sidenote: Technically the op-amp is an input buffer here and the transistor is the gain stage.)

The biggest downside I can see with the hugely variable collector current of the buffer is that the high frequency roll off could become significant which is maybe why it sounds as if you're on the neck pickup (less harmonics).

BTW if the sound is fuzzy, you may well find that the collector voltage is either very low or very high. The collector needs to be a little above half the battery voltage to get the maximum ac voltage swing without clipping.

MrEase, post: 351975 wrote: Guitarfreak,

You really don't make it easy! :

I agree with Boswell that this design is certainly not going to yield consistent results and if just for the op-amp, you will need to make sure the battery is always healthy. As to the buffer stage, bearing in mind that it seems they supply almost any transistor for this and also with transistor gain spreads, the biasing is going to be very variable. If this "confidence booster" is intended to boost your confidence in building something that works (I can see little other purpose to the circuit) then I don't think it's a great design.

The first thing I suggest checking is the voltage at the transistor collector. This really should be sitting around 5 V or so (it's probably nearer 3V) but this is likely to vary significantly. Also, the transistor output is surely going to be loaded by the 15k trimpot, so it could be a good idea to increase this to the 100k you said. That is assuming you are taking the output to a high impedance guitar amp. Many mixing desks have relatively low input impedance so this would tend to load the output anyway.

(Sidenote: Technically the op-amp is an input buffer here and the transistor is the gain stage.)

The biggest downside I can see with the hugely variable collector current of the buffer is that the high frequency roll off could become significant which is maybe why it sounds as if you're on the neck pickup (less harmonics).

I did mention that the project was a BYOC confidence boost when I said the PDF is for download, but that's beside the point. Lets move on and push forward.

Your analysis makes sense because I looked up a few op-amp configurations and the schematic matched none of them. I did think that the first resistor which goes parallel downward (the 2M2) was the buffer. The one that points upward I am not sure about, I've never seen a double resistor configuration like that. At least I think not.

I still plan to build a pedal out of this and mod it using extention PCB's to make it better. I bought it to experiment so it allows me to do that.

New news. Now the circuit does nothing. It gives very low very distorted tone only, with some intermittent very loud banging/popping noises. I don't even want it plugged into my rig while it's doing that or that shit could blow my speakers. Any idea what that sounds like?

Maybe I'll look into a different clean boost circuit to work on?

Can you use stranded wire in a PCB project like this? Or should you use only solid core wire?

You can use either type of wire, it makes no difference for a project like this electrically. Stranded wire is more durable though until the PCB is housed. If you use stranded wire, make sure you strip the wire, twist the strands and tin the ends before you fit them to the PCB or jacks.

The two 2M2 resistors are just providing a voltage divider to bias the input of the op amp. The op amp itself is just a unity gain buffer (the output is taken back to the negative input). Again the design is not great as the 4558 input bias is not the lowest so the input impedance of this circuit is only going to be several hundred kohm. The general wisdom is that you should aim for at least 1 Mohm for guitar pickups so as not to load them and hence dull the tone.

With a little careful modification there is no reason that the existing circuit board could not be improved significantly. At least better consistency could be achieved. What I would try doing first is to look at the voltages around the circuit - particularly the collector voltage as I mentioned earlier but also the op amp output. Let us know what you get as this will give some clues to what's going on with your board.

To understand op amp circuits a bit better have a look at this link. http://www.national.com/an/AN/AN-31.pdf it gives some basic op-amp circuits with the maths to work out the gain etc. These circuits and equations apply to almost all op amps - certainly audio types.

The circuit reads ~3.5v at transistor collector and ~2.5v at IC pin 1. Ground measured from 'Out sleeve'. Battery reads upwards of 9v so it checks out.

OK That is pretty much what I would expect. There are two different things you could try for the input buffer. The op amp output should really sit at about half voltage which is what the 2 * 2M2 resistors will provide. Unfortunately the input bias current of the current op amp is pulling this "ideal" down. You could either

a) Substitute the op amp with a FET input type - say a TL072. These require virtually no input bias so you should then see around 4.5V on the op amp output. This may result in a slight increase in noise.
b) Compensate the input bias current (likely to drift more with temperature though). To do this, change R2 to 3M9 or 4M7. This will get the op amp output nearer the ideal. You would normally just use lower value resistors (say R1=R2= 470k) but that would also considerably reduce the input impedance and you don't really want to do that to guitar pickups.

On the transistor gain stage the simplest mod would be to reduce R6 to 6k8 or thereabouts. This will slightly reduce the maximum gain of the booster but should make it sound a bit cleaner.

Any other mod's (and there are many you could try) will get more involved and will very much depend on what you are trying to do. Note that with this kit and the variability of the design, the two changes I've suggested are really for this unit only as other samples may well need something else.

It also seems that you are thinking of building this circuit into a stomp box. I would say that I don't think that is a great idea as the overall design seems to be exactly what it says on the tin and is certainly not a design that is ready for production!

That sounds like a plan. I'll have to search Mouser/Digikey to see if I can get my hands on a few components to try out. Until then I still need help debugging this thing. I played it for a few minutes today and it sounded pretty cool. Very fuzzlike with lots of sag, but cleans up to light grit when you roll back on the volume knob. The sound is cool but it's not a clean boost at all, even though the guide says it is. After about 15 minutes of playing there came intermittent noises and then eventually the noise grew into a wall of white noise and there failed to be any guitar sound coming through. This happened the last time that I used it for upwards of 15 minutes as well. Any ideas?

I presume you mean the sound changes when you alter the guitar volume rather than the preset on the PCB. That is most likely due to the poor biasing the unit currently has and certainly why it is not clean.

As to the thing going off after 15 minutes the first thing to check is the battery (do this while the unit is still on and failing). If the battery voltage drops below an ill defined level (7 - 8 Volts) the op amp internal biasing will start to fail and anything could happen. If that's not it, recheck the voltages I mentioned before without powering down the unit.

While failing, the battery reads a steady 9.2v. The collector reads anywhere from 3.5-9v and fluctuates very rapidly. The IC pin 1 also fluctuates very rapidly and shows between 1.5-4v.

There is nothing obvious as to why this would happen but it is clear that the transistor stage is only reflecting the fault that seems to be occuring in the first stage. The op amp is a dual type and it seems the second amp is left disconnected. This is never good practise. What I would suggest is to link pins 1 to 5 and pins 6 and 7. This will at least ensure that the second amp has known conditions and is within limits. Other than this I would say that the only possible problem parts are R1, R2 and the op amp itself. Be certain to reflow the solder on all three of these components to eliminate dry joints. A bit of flux on the joints would be good when you do this or at least apply a fraction more solder but remove any excess. Do you have a solder sucker?

If you repeat this test, make sure to turn the guitar volume to minimum and leave it connected when measuring voltages. This ensures the input is effectively grounded. I guess you are taking these measurements with a digital multimeter. In the presence of any AC waveform, they can get very confused but at least the max and min voltages you see should be "real". If you use an analog meter then the measurements are very limited by the ballistics of the meter itself. It would be handy to know what you are using.

MrEase, post: 352068 wrote: There is nothing obvious as to why this would happen but it is clear that the transistor stage is only reflecting the fault that seems to be occuring in the first stage. The op amp is a dual type and it seems the second amp is left disconnected. This is never good practise. What I would suggest is to link pins 1 to 5 and pins 6 and 7. This will at least ensure that the second amp has known conditions and is within limits. Other than this I would say that the only possible problem parts are R1, R2 and the op amp itself. Be certain to reflow the solder on all three of these components to eliminate dry joints. A bit of flux on the joints would be good when you do this or at least apply a fraction more solder but remove any excess. Do you have a solder sucker?

The problem seems to be aggravated by loud input levels so I am not sure if that helps you in your analysis. Couldn't I just take the op-amp out and replace the working side with a transistor? I'd need to pull up the ends of those two resistors and attach them to the transistor leads, but I think that it could work. Values need to be tweaked you think? I don't have a solder sucker, but I have wicking braid.

MrEase, post: 352068 wrote: If you repeat this test, make sure to turn the guitar volume to minimum and leave it connected when measuring voltages. This ensures the input is effectively grounded. I guess you are taking these measurements with a digital multimeter. In the presence of any AC waveform, they can get very confused but at least the max and min voltages you see should be "real". If you use an analog meter then the measurements are very limited by the ballistics of the meter itself. It would be handy to know what you are using.

I am using a Radio Shack Digital meter.

[[url=http://[/URL]="http://www.radiosha…"]29-Range Digital Multimeter - RadioShack.com[/]="http://www.radiosha…"]29-Range Digital Multimeter - RadioShack.com[/]

Louder levels will always aggravate the situation - more signal = more hitting the limits of a dodgy design.

No, you certainly cannot just replace an op amp with a transistor! They are completely different devices operating on completely different principles. The tweaks I have already suggested are what I still recommend. They are all intended to get the biasing of each stage optimised (for your particular instance) and hence get you the maximum signal handling with minimum distortion.

As I have said, this design is certainly not ready for production and I think it is a very weak product considering its aim is to boost the constructional confidence of amateurs. While buying kits is a good economic way of getting both the product and experience, you also need to consider whether the design of the kit is worthwhile. Would you ever buy a new guitar or amp without trying it out first?

Alright I'll look into those mods you mentioned. In the meantime I will look for a better circuit to play around with. Do you know of any that are worthwhile and not too many duckets? I was looking for a clean boost at first, but I like the over the top fuzz effect I am getting from this project and I've become more interested in doom metal recently so a nice fuzz would be useful to me. I'm looking at the GGG Dallas Arbiter fuzz box and for only $50 it does seem to be a very good deal. Except that I plan to buy a DA fuzz anyway, so that would be redundant.

Also, do you think my meter is good? I just got it, so if I should get another one let me know now so I can return it. Thanks.

I'm not really into the market for stomp boxes so I cannot advise you on this. As always though the proof of the pudding is in the eating - so try them out if you can.

I don't see anything wrong with your meter - these things are all similar at the same price point and the accuracy is quite typical for the price.

This is a curious circuit. I'd have used the OpAmp for the variable gain and the transistor as a unity-gain follower.

I'll bet it would work a lot better if the trimpot were at the output of the opamp rather than the output of the transistor.

vttom, post: 352086 wrote: This is a curious circuit. I'd have used the OpAmp for the variable gain and the transistor as a unity-gain follower.

I'll bet it would work a lot better if the trimpot were at the output of the opamp rather than the output of the transistor.

The circuit is really intended for assembly practise so I guess the circuit operation is secondary. Why would you even bother with the transistor follower if all you wanted was fixed 13dB or so gain? That is all this circuit (is supposed) to do!

Quick question. The results of this experiment have intrigued and inspired me. If I were to design my own pedal in the future with a pot controlling the transistor bias voltage at each stage, is there anything that can go wrong with a very low bias, or is it just this circuit in particular? How low can the bias voltage safely go?

There is no answer I can give you for such a generalised question except that you could take the bias down to zero if you wanted a class C stage! If you were to design your own pedal, you would need to address and calculate biasing whilst you were doing the design. As you have no design as yet, then there are no calculations I could make for you - nor should there be any need as any calculations should have been made already. After all, it is necessary to establish the DC operating point of a circuit (and the range of variations) before you can really move on to any AC analysis.

I'm glad you're inspired and I have no wish to discourage you but if you are considering designing pedals then you will have to know how to design them properly. There are no real short cuts and only so much can be achieved with a "suck it and see" methodology. What I would recommend for the beginner is to stick with op amps or published designs. Op amp applications to audio circuits are generally much simpler to calculate and the idealised formulae will generally be OK. This is why I gave you a link earlier in this thread and I'm sure you will also find suitable books in the library. Whilst I am happy to try and help you with specific problems it is just not practical to give you an education in electronics design on this forum!

Guitarfreak, post: 352109 wrote: Quick question. The results of this experiment have intrigued and inspired me. If I were to design my own pedal in the future with a pot controlling the transistor bias voltage at each stage, is there anything that can go wrong with a very low bias, or is it just this circuit in particular? How low can the bias voltage safely go?

It seems you have got some valuable experience from this kit, but as Mr Ease indicated, it's an educational design and does not have the pedigree for serious guitar use.

By the way, the bias for a transistor is specified in terms of current and not voltage, so you are doubly dependent on the resistor values at the input and battery voltage to get this particular design to behave in an acceptable fashion. That's not a recipe for a production design! If you can specify what aspects of the sound from this unit you would be interested in translating to a pedal that can stand the rigours of normal use and abuse, we may be able to make some suggestions about how to modify this teaching aid to get to those conditions.

I liked that it was like an over the top fuzz effect, it would go really well with some of my doom projects I've just written. I was thinking of designing a two or three gain stage (all transistor) pedal which I could adjust the bias of one (or more?) of the gain stages via an external pot. Most fuzzes are On/Off and what I hope to accomplish with this project is a fuzz that can go really from almost no gain at all to really over the top squashing while not really coloring the tone much if at all. Cold biased gain stages seems to be the way to accomplish this. To me anyway.

I see what you mean, I was trying to calculate the Resistor values last night and I realized that without more information it doesn't work. I should be using Ohm's Law right? V=IR. What do you use for the current value?

Transistor amplifiers work like this....

Use a resistive divider to establish a DC bias voltage at the base of the transistor. Couple in your AC signal with a capacitor (as they've done in the Confidence Booster).
Put a resistor (we'll call it Rc) between the supply voltage (VSS) and the collector, and another resistor (we'll call it Re) between the emitter and GND.
The voltage at the emitter (we'll call it Ve) will be the base voltage minus a diode drop (0.4v-0.6v).
The voltage at the collector (the output of the circuit, we'll call it Vc) can be calculated using V=IR as follows:

VSS-Vc = I * Rc
Ve = I * Re -> I = Ve/Re

In a transistor, the collector current must equal the emitter current, so we can make the following substitution for I:

VSS-Vc = (Ve/Re) * Rc = (Rc/Re) * Ve

Then solve for Vc:

Vc = VSS - (Rc/Re) * Ve

Basically, the gain of the amplifier is the ratio of the Collector resistor to the Emitter resistor.

BTW - There are lots of things to watch out for with the type of amplifier.

The biggest thing to keep in mind is that it will amplify the DC bias as well as the AC signal. So you have to make sure you set the bias voltage and gain such that you avoid the situation where Vc

Then again, it's probably in that region where the cool sound comes from.

Also, this amplifier can't tolerate any load on the output (that'll break the assumption I relied on in which the currents through Re and Rc are equal). But this is easy enough to fix by tacking on an emitter follower (that's where you connect the collector output of the first stage to the base of another transistor whose collector connects directly to VSS and has a biasing resistor between the emitter and GND). Emitter followers have very high input impedance (in other words, look to the previous stage like a very light load), and a very low output impedance (meaning they can drive a hefty load).

While vttom makes some valid points there are a few omissions. In the design you have, the source voltage for base bias is just 0.81V (1/11*9(V)) and has a source impedance of 90k or so, so the voltage at the emitter lies somewhere between 0.21 and 0.41V (using the Base Emitter voltage of 0.4 to 0.6V) less the base bias source drop which will be significant. In other words the emitter current can vary by considerably more than 2:1. This is not good. Also the base source impedance is around 300 times the emitter impedance which is far too high a ratio for predictable biasing due to beta variations.

As far as AC gain is concerned, Rc/Re can often be near enough but in this design, the bulk emitter resistance (re) will be significant (around 50 Ohm) compared to the emitter resistor (Re) of 360 Ohms. The load impedance must also be accounted for in the gain calculation. This is why I find it crazy that the part list for this project can include any value of trimmer from 5k to 100k! Of course the load impedance does not take part in the DC bias calculation.

As far as I can see, Guitarfreaks "confidence booster" should have done just that. It seems to be working as well as the design allows! I would still recommend using op amps until you really know how to design with transistors properly.