signal saturation?!?!

[BramcoteLorne]

hi guys,

I've been interested in building effects for about the past 6 weeks, I understand how a basic curciut works and have quite a lot of experience soldering (crazy pickup/tone combinations in my guitars...)

So far I've have had a go at building 'beginers proj, bazz fuzz, EH LPB and other really simple curcuits

ok so that's me...  

my problem is one that I haven't been able to overcome myself in the last few weeks:

I can't make an effect that will work when the strings are played loudly and when they are played gently, at one of the two extremes I get a horrible sort of crackly distorted sound, with a crackly sort of nasty tremelo as the notes die away.

The same is true with changing pickups - it may sound fine with a single coil (eg middle in my strat) but the switch to the bridge (pearly gates humbucker) and the signal breaks down again of heavy strokes.

in my limited knowledge I have attached a trim pot to the collector (this is called biasing....) on the transformer, this does seem to help a little but the distortion is just more in background. And again changing the pickups from single to hum will cause the sinnal to break down again. (is this called saturation...)

The beginners project is a prime example:

I have changed the gain pot cap right down to a 2.2 (from a 47uf) and tacked a 1k resistor on the end to ground, and with this config I only get the sound breaking down on really heavy strokes on max gain.

which I guess is okey but I'd rather be able to build something that works like it should!!!

I have also built it with just a 10k resisitor to ground in place of the pot and the cap - this does warm the sound up but doesn't really give much of a boost.

I have used a 2n2222 metal can transistor for this curcuit. although I have tried BC547's BC549's BC550's and others I can't remember the names of. I can't get hold of a 5088 or a 5089 over here in the UK, as this seems to be the favoured tranny

my conclusion is that the transistor is receiving too much signal, can't handle it and is so distorting the signal.

the bass fuzz is the opposite works fine the harder you play but quite notes sound really crackley and nasty.  All the notes seem to die away really quickly too.  (I don't know if this is the way that the effect is meant to work tho.)

what do I need to do to fix these problems? I really like the effects that I have made so far but of course I don't want to have to be limited to having to tone them down or play in a certain way to use them.

I would love it if you guys can give me some sugestions as to what might be wrong and how I can fix the problems I am having.



oh and BTW happy christmas to all!


dave.

[niftydog]

possibly a combination of too much gain (or not enough headroom - how's your battery voltage?) or incorrect biasing.

Keep in mind that (at least for when you're playing hard) it could be you amp that is overloading. So, it may only be the effect breaking down at very low levels. This would indicate incorrect biasing.

[BramcoteLorne]

hi,

thanks for the quick reply,

I'll have a go at trying a new battery, I didn't think about that.

I've tried the effect on a couple of amps (a big marshall, 15w WEM and a marshall g10 practice amp) and it's the same so I can discount that.

I've tried a couple of guitars too - that's why I figured out that the hotter the pickup the worse the signal breakdown.

I'd like to figure out how to wire it so I can use a 9v ac adapter, jus so I have less variables to work with. Is it just a case of wiring a connector to the + and - and pluging the adapter in?

I'll be happy when I've got all this sorted out! I really like the sound of the NPN boost but just can't seem to nail this problem.



thanks,


dave.

[niftydog]

I'd like to figure out how to wire it so I can use a 9v ac adapter, jus so I have less variables to work with. Is it just a case of wiring a connector to the + and - and pluging the adapter in?

no. You need to "automatically" disconnect the battery when the power supply is plugged in. This requires a switching jack for the DC input. I beleive there is a few useful diagrams already floating around here, check the FAQ and the other usual sources.

Either that, or just ditch the battery entirely.

[BramcoteLorne]

ok IC, maybe I'll remove the battery at the mo, and run just the AC power adapter because I just want to get the thing up and working as it should be.

I tested the bat last night and it was at 7.49. it's an energizer BTW.

I thought about another thing last night:

is it possible that the transistor has too much gain? - on the standard curcuit when the pot was cranked before it was into 'flipping eck that's loud' teritory compared to the bypassed signal.

would a resitor in the signal path upto the transistor bring the level of the signal? would it affect the tone? because I kinda do like the way it sounds right now.



thanks again for your input.

[niftydog]

I tested the bat last night and it was at 7.49

that's about as dead as they come. Also, I assume that was tested out of circuit. Measure it while it's powering the device and it'll be even worse.

is it possible that the transistor has too much gain?

it is possible, but I would urge you to work systematically eliminating the easy things first. Try a new battery first. Futz with the biasing second, and then maybe look at the gain structure.

[BramcoteLorne]

hi yeah,

that was out of the curcuit.

I just guessed that 7.5ish would be ok seen as it was pretty near to 9 I just guessed it had lost a little power.

I suppose thinking about it tho, that if the transistor has to swing - to + it doesn't have so far to swing if the bat is only at 7.5.

am I along the right lines...?

I'll try and pick some new ones up tommorow, at what point should I junk a battery then? if it reads anything less than 9?

thanks,


dave.

[niftydog]

the transistor has to swing - to + it doesn't have so far to swing if the bat is only at 7.5.

well, that might not be a problem as guitar signal are quite small (less than a volt) although some circuits will utilise the full voltage swing available - so yes, you're on the right line. The problem is more likely to do with simple biasing methods that do not compensate for the dropping voltage supply.

Batterys will very rapidly drop below 9V. When to ditch depends on the circuit you're using the battery in. I would ditch when they're below 8V or nearing 8V.

[BramcoteLorne]

hi,

I changed the bat for one measuring 8.83 and no noticable differance in the signal distortion,

I'm going to change the pot back to a 47uf and junk the resistor and see how that effects things, with my old batt it was worse but I'll see what happens this time. I'm pretty sure it will make things worse but I will give it a go.

as far as playing with the bias goes would the current to the collector need to go up or down?

what sort of values on the E B an C should I aim for?


thanks again,


dave.

[Transmogrifox]

Your description sounds very much like a biasing problem.  That's often a problem with transistor variance between types. What's happening when the circuit has no signal is that some (or all) of the transistors are resting in an "off" state, or a saturated state and it takes a pretty significant input signal to even get them to swing into operating range...so when your signal dies down, the circuit just kinda shuts off and crackles on and off every once in a while as the fading signal bumps it back on until it's too small to do anything.  Basically you have a crappy version of a 1/2 wave rectifier.

usually you're aiming for about 1/2 the power supply on the collector--I haven't looked at the Bazz Fuss schematic recently, but my memory tells me it's similar to the fuzz face.  Take the Fuzz face for instance (so we're on the same page), the first transistor's collector is actually at about a volt, or a little over, but the second tranny is biased at 4.5V ideally.

here's a test that will give you a good clue about whether the bias is anywhere near what it should be:

measure between the base and emitter of each tranny.  If it's not within the range of .5 to .7 volts, you have a bias problem.

Next, measure from collector to base.  If this voltage is less than about .7 volts, you definitely have a bias problem   For most gain stages, this voltage should be around 2 to 4 volts as .7 would likely give you a pretty farty sound even in a fuzz face Q1.

[BramcoteLorne]

hi guys,

I took some measurements from my beginners project I guessed it was as good a place as any to start.

here they are:

col to ground : 5.8
base to ground : 2.04
emit to ground : 1.49

col to base : 3.51
emit to base : 0.58
col to emit: 4.33

batt 8.73


does anything jump out at you?

I'm guessing from what you have said that the collector is too (5.8 instead of 4.5)  high??

all the other reading seem to be in the same sort of area as you mention.

where do I need to go from here?


thanks,


dave.

[niftydog]

base to ground : 2.04
emit to ground : 1.49

they seem kinda low to me. Have you double checked and measured all your resistors to ensure they are the correct values?

[BramcoteLorne]

hi,

I'll check it through again tonight, if I don't get any joy I'm going to build it from scratch again and measure everything before I put it on there.

what sort of figures should I be reading from the base and emmiter?

will this curcuit def work with a 2N2222?



thanks again,


dave.

[niftydog]

the difference between the base and the emitter is always going to be the same. Set the base voltage and you're also setting the emitter voltage. The difference for most transistors is in the region of 0.6-0.7 volts.

From a quick forums search, I would think the base needs to be closer to 2.5V or maybe even higher. But, I have never built nor analysed this circuit so I can't give any specific help.

[BramcoteLorne]

how do increase the voltage at the base?

would I need to increase the resistor running from before the base to ground or increase it?

or am I on the wrong lines completely???

oh and BTW the transistor is a 2n2222a I measured it and it measured  (I think) at 215.

this is the schem I'm working from:




dave.

[BramcoteLorne]

hi,

I had a go at changing the 10k resistor (from before the base to ground)

results:

anything over 10k makes the distortion sound worse,

anything less reduces the volume and so kinda reduces the distortion, but the signal just sounds kinda weak and muddy.


I'm starting to wonder why I can't make this thing work - I've seriously spent weeks on the thing and still the same breakup on the sound.

could it be the components that I bought were at fault?



dave.  :cry:

[niftydog]

could it be the components that I bought were at fault?

Hence why I said;

Have you double checked and measured all your resistors to ensure they are the correct values?

how do increase the voltage at the base?

I would try increasing the value of the 47k resistor at the very bottom of the picture... but only after I checked and double checked (out of circuit) the values of all the resistors in the circuit.

would I need to increase the resistor running from before the base to ground or increase it?

There are no resistors running from base to ground. There are two resistors in that path, a 10k and a 47k.