Careful with that voltmeter Eugene!
Ok, so 8uA base current into Q1 and 0.64v on base...(thanks a bunch Eugene!)
Humble pie for me epicyclic.
About 200uA down the 33k connected to Q1 collector, 0.67v across the b-e junction of Q2 and as with the previous set of err...."readings" (rrrrr!) Q2 b-c junction still forward biased.
Voltage at node between 330 and 8k2 indicates that the gain off the transistor you've used for Q2 is too high for an 8k2 resistor in that location. Either test and select a lower hfe transistor (it is worth the bother sound wise) or just drop the value of the 8k2.
Incidentally, the "C" at the end of BC109C means it's a high gain type, not so good in this application.
If you do want to check the hfe of transistors for this circuit, here's a simple way to do it using just a meter and a couple of resistors,
http://www.geofex.com/Article_Folders/ffselect.htm
it works fine on Si transistors too, leakage shouldn't be an issue with these.
Ok, so 8uA base current into Q1 and 0.64v on base...(thanks a bunch Eugene!)
Humble pie for me epicyclic.
About 200uA down the 33k connected to Q1 collector, 0.67v across the b-e junction of Q2 and as with the previous set of err...."readings" (rrrrr!) Q2 b-c junction still forward biased.
Voltage at node between 330 and 8k2 indicates that the gain off the transistor you've used for Q2 is too high for an 8k2 resistor in that location. Either test and select a lower hfe transistor (it is worth the bother sound wise) or just drop the value of the 8k2.
Incidentally, the "C" at the end of BC109C means it's a high gain type, not so good in this application.
If you do want to check the hfe of transistors for this circuit, here's a simple way to do it using just a meter and a couple of resistors,
http://www.geofex.com/Article_Folders/ffselect.htm
it works fine on Si transistors too, leakage shouldn't be an issue with these.
Last edited:
Yes, I'm aware of that. With 1.45V across the 1K, Q2 has no choice but to saturate with the 8.53K collector load and 9 volt supply. It would work with a 24V supply, or simply reduce the 8.2K to something like 4.7K to get the collector of Q2 to mid supply. I think the 1K and 8.2K were predicated on using Germanium transistors.
Last edited:
Good point, 1.45mA is too much for the total 8.53k collector load. More humble pie for me Loudthud.
@Simon: Thanks for the floyd reference and the greek lesson, not to mention the nice pic
So, any suggestions as to what values to try based on these numbers?
I'd go with Loudthud's suggestion of 4k7 to start with, but you'll probably need to cut and try for best results.
Any value is ok - even 0R shouldn't blow a BC108, given the 330R upstream. The max current that can flow is about 25mA, and a BC108 is ok up to 100mA & 300mW.
It is worth experimenting. Don't bother getting the old resistor off the veroboard, cut its body out to leave the two leads sticking up.
Then either temporarily solder resistors on or perhaps use a small bit of connector block, until you find what works best for you. Possibly try a 5k or a 10k pot or trimmer, connecting to the middle and either one of the end terminals to start with.
My mention of epicyclic was referring to a point on which he'd been right and I'd been wrong - though "wheels within wheels" is a bit like this circuit.
Let us know how you get on.
Last edited:
Hi
eugene would you measure the resistance of your 1K0 fuzz pot and repeat your voltage measurements ( with fuzz pot on min fuzz and keep negative lead of meter in same place also include supply rail reading ) as the latest readings dont make any sense ie 1.45V on the emitter of Q2 .
Loudthud ...try www.datasheetarchive.com and search with the trannie number then scroll down for a list of pdfs . I dont know who originally made them .
eugene would you measure the resistance of your 1K0 fuzz pot and repeat your voltage measurements ( with fuzz pot on min fuzz and keep negative lead of meter in same place also include supply rail reading ) as the latest readings dont make any sense ie 1.45V on the emitter of Q2 .
Loudthud ...try www.datasheetarchive.com and search with the trannie number then scroll down for a list of pdfs . I dont know who originally made them .
Last edited:
BC are Pro Electron type numbers, as opposed to, say, JEDEC or JIS.
Pro Electron
Philips, Motorola and Fairchild all appear to have made them. One thing to watch out for with those in a TO-92 package is that there are umpteen variations of this denoted by a final letter after the TO92, they have different pinouts, of course.
Same goes for BCxxx versus BCxxxL - different pinout again.
Plenty of stuff comes back if you google on eg 'BC108 data', I'll attach a couple that I've got stored.
Pro Electron
Philips, Motorola and Fairchild all appear to have made them. One thing to watch out for with those in a TO-92 package is that there are umpteen variations of this denoted by a final letter after the TO92, they have different pinouts, of course.
Same goes for BCxxx versus BCxxxL - different pinout again.
Plenty of stuff comes back if you google on eg 'BC108 data', I'll attach a couple that I've got stored.
Just rechecked every resistor and the 33K resistor on Q1-c / Q2-b seems to be a 3.3K one. Damn! Can't believe i didn't think to do this before (or that I actually placed the wrong value in there). But does the 2,12/2,18 reading here seem too high? and could that be drawing the current that Q2 collector should be seeing? I'll have to change it to a 33K one and check back. I'll have to order new ones so if you think that my choice of transistors would prefer a different value in it's place give me a shout and i'll order some of those as well.
Both of these will prevent it working.
Always a good idea to check components with a meter before soldering in. Especially resistors and pots, are they're so easy, and so prone to confusion as to their values!
Have to admit that I'm now a bit baffled how the voltages around Q2 are so low! Where's the current apparently flowing through the 8k2 resistor going to?
Eugene, may I suggest, given the small component count of this circuit, that you have another go at building it? All new bits, checked carefully etc. Your soldering will probably be quicker and better a second time round too.
Wrt transistors:
BC108, without the C at the end is listed as having gain (hfe) in the range 110 to 800. Popular folklore has it that the ideal values of gain for Q1 and Q2 are 70 and 115, respectively, but you don't want to get too hung up on that.
BC107s, gain listed as being 110 to 450 would be worth a try.
At the moment, your accidental use of 3k3 instead of 33k in the base feed to Q2 means that there's loads too much current flowing into what is probably a high gain transistor at Q2.
As you've got a meter, do think about testing transistors for gain, as detailed by RG Keen / Geofex, at the page I linked too earlier.
Yeah, I'll probably build another one later on, but i feel a need to get this one working first (all other values checked out, and voltages seem even across all joints so i still think it's worth a try). I will, as i said, order some different resistors (around 33K) for both Q1 and Q2 collectors and give a go at biasing this thing. what voltages should I be looking for on Q1-c?
I mean 1K and a little tolerance
I thought I might test some different values in the position if i knew what voltages to look for and try to get as close to the mark as possible... thought I might as well get it right before soldering it in place, but i would obviously start with a 33K and check the voltages.
I thought I might test some different values in the position if i knew what voltages to look for and try to get as close to the mark as possible... thought I might as well get it right before soldering it in place, but i would obviously start with a 33K and check the voltages.
Hi Eugene,
Here is some web site discussion on NPN Si version with some typical voltages. At Q1 collector: 30K at collector will probably be alright, but if you still have too high a DC voltage there just try & add more resistance at Q1 collector. Then adjust Q2 collector resistor to get around 4.5 VDC.
Sili-Face
Here is some web site discussion on NPN Si version with some typical voltages. At Q1 collector: 30K at collector will probably be alright, but if you still have too high a DC voltage there just try & add more resistance at Q1 collector. Then adjust Q2 collector resistor to get around 4.5 VDC.
Sili-Face
I wouldn't try changing the circuit as per the runoffgroove schematic, at least not til you've got it working, though the point about getting the collector of Q2 around about halfway between the two power rails is a good one.
Good you feel like sticking on til you've got it working.
Don't stick anything like 30k in the collector of Q2 - if anything, you'll want less than 8k2, not more.
First of all get the Q1 collector resistor sorted - I'll mail you one if you want!
Also Eugene, please tell us, is the convention for decimal point/comma in Norway that 1,003k means 1003 ohms, or that it means 1003000 ohms - it varies country to country - I think in Italy for instance, it means 1003 ohms, but don't quote me on that.
Good you feel like sticking on til you've got it working.
Don't stick anything like 30k in the collector of Q2 - if anything, you'll want less than 8k2, not more.
First of all get the Q1 collector resistor sorted - I'll mail you one if you want!
Also Eugene, please tell us, is the convention for decimal point/comma in Norway that 1,003k means 1003 ohms, or that it means 1003000 ohms - it varies country to country - I think in Italy for instance, it means 1003 ohms, but don't quote me on that.
Last edited:
Breadboarded the circuit earlier today for my own curiosity....results
BC107Bs
Q1.................. Q2
B 0.613V.......... B 1.330V
C 1.330V.......... C 3.18V (8K2 and 330R) , 5.03V (5K6 and 330R)
E 0.00V............ E 0.691V
mid point 8K2 and 330R 8.9V 5K6 and 330R 8.9V
BC109Cs
Q1............... Q2
B 0.594V....... B 1.283V
C 1.283V....... C 3.56V (8K2 and 330R) , 5.29V (5K6 and 330R)
E 0.00V......... E 0.652V
mid point 8K2 and 330R 8.9V 5K6 and 330R 8.9V
I put 27K resistor in series with the input to try and simulate a passive guitars output impedance and the output pot was set on max........AC results
fuzz pot on min..... gain = 1.16X
fuzz pot on max.... gain = 240X
both sets of trannies gave the same AC results
With the 8K2 collector resistor in circuit clipping started to occur on the negative first with an output of approx 65mV rms and the positive at 130mV rms .
With 8K2 replaced by 5K6 the output started to clip symmetrically at 140mV rms.
As i had done these measurements just thought i would post them .
What assymmetrical clipping does for the fuzz sound i dont know as i dont own a guitar and wether this matters when your are basically creating squarrish waves again i dont know . My only thought is youd probably get a louder and perhaps more controllable sound with the 5K6 and resulting
symetrical clipping/squaring/fuzzing.
PS rail volts 9.16V
BC107Bs
Q1.................. Q2
B 0.613V.......... B 1.330V
C 1.330V.......... C 3.18V (8K2 and 330R) , 5.03V (5K6 and 330R)
E 0.00V............ E 0.691V
mid point 8K2 and 330R 8.9V 5K6 and 330R 8.9V
BC109Cs
Q1............... Q2
B 0.594V....... B 1.283V
C 1.283V....... C 3.56V (8K2 and 330R) , 5.29V (5K6 and 330R)
E 0.00V......... E 0.652V
mid point 8K2 and 330R 8.9V 5K6 and 330R 8.9V
I put 27K resistor in series with the input to try and simulate a passive guitars output impedance and the output pot was set on max........AC results
fuzz pot on min..... gain = 1.16X
fuzz pot on max.... gain = 240X
both sets of trannies gave the same AC results
With the 8K2 collector resistor in circuit clipping started to occur on the negative first with an output of approx 65mV rms and the positive at 130mV rms .
With 8K2 replaced by 5K6 the output started to clip symmetrically at 140mV rms.
As i had done these measurements just thought i would post them .
What assymmetrical clipping does for the fuzz sound i dont know as i dont own a guitar and wether this matters when your are basically creating squarrish waves again i dont know . My only thought is youd probably get a louder and perhaps more controllable sound with the 5K6 and resulting
symetrical clipping/squaring/fuzzing.
PS rail volts 9.16V
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.