First builds (some volume problems)

[Obulus]

So, I built my first pedals a couple of days ago but I'm not entirely satisfied with the results.

First up is a DBA Soundwave Breakdown clone. I built it from the layout at tagboardfx, found here: http://tagboardeffects.blogspot.be/2012 ... kdown.html

Problem is that it is very very quiet, I can compare to the genuine DBA one and it's a really big difference. I tried swapping transistors and used the transistors that are last suggested to work well in the comments (replaced the 2N5089's with 2N5088 for Q1 and Q2 and 2N3904 for Q3 and Q4). It got a little better but maximum volume is still below unity. I tried other combinations too, but everything else I tried was even more quiet. Is it possible that I made another mistake? Are there other components I could swap for more volume? It does sound similar to the original as far as I can compare right now. Here are some gut shots (NSFW for big pictures):

NSFW: show

The second one is this 'analog bit crusher' circuit: http://tagboardeffects.blogspot.be/2012 ... usher.html
Kind of the same problem but less extreme, it's pretty quiet. More extreme settings get more quiet than less extreme. I don't have the problem with the IC's that they all mention in the comments. The sound is definitely right, it's just a volume thing. Should I try different transistors for this one too? No gut shots for this one (if required I can make some this weekend).

Ideas? Any help is appreciated! Try to keep explanations simple because I'm just starting out

[skullservant]

That analog bit crusher is quiet, I've built it before and had the same quietness.
As far as the DBA, check to make sure you dont have any accidental bridges, miss connections/links, and all of the cuts cut where they need to be. Nice first go with your wiring man!

[Gone Fission]

Like Skully said. Also, double check your resistor values with a multi-meter, especially if they come out of a mixed pack like those nice big sets from Radio Shack. I swear, the colors they use have got to be by what paint is cheap and not by what is clearly distinct from all the other colors. You get accustomed to telling them apart with time, but I still wonder sometimes and verify. (And I rarely trust my color code reads on a computer screen, so remote tracing is a pain.)

[Obulus]

Too bad about that bitcrusher. Anything i can do about it? Maybe slap an lpb after it?

And thanks for the tips with the dba. Hopefully i have some time tot check it tomorrow!

[skullservant]

Yeah that's always an option to throw a clean boost after it

[eatyourguitar]

because of the reverse biased transistors in the death by audio pedal, cloning involves much more work with transistor selection and debugging. there are no text books that teach BJT abuse. you can kinda sort it out if you think of the transistors as zener diodes with some of the properties being controlled by the base. there is no doubt that the transistors may end up permanently damaged with some additional degradation over time. I would not attempt to clone this unless i had a real one to study or I wanted to sit for hours testing things.

the bit crusher we can fix!

http://www.experimentalistsanonymous.co ... rusher.gif

I think it is this^ or something like this. on the output, all you need to do is add two resistors like this.

http://ecetutorials.com/wp-content/uplo ... ninvrt.gif

100k + 100k is unity gain. 100k loop 22k to ground is a voltage gain of about 4.

[Obulus]

Thanks for your reply! For the bit crusher, just to be sure (because I'm really new to all this and had to do some reading to understand some stuff), I have to add a 100k resistor going from the output of X2 to the "-" (inverting input?) and from there a 22k resistor to ground too? Would it be wise to use a trimmer in place of one of these resistors to choose gain (if I have the space for one)? Or will the 100k/22k combination be okay?

About the DBA: I DO have a real one to study, but no studying skills whatsoever. Do you think I can figure the transistors out or is this way above a newbie's head? I really know shit-all about transistors so... Or am I doomed to just try lots of transistors? I will definitely double check my wiring and resistor values first though, starting with the easy checks!

[eatyourguitar]

I don't think you need a trimmer if you have a volume knob. you said that your signal is less than unity. I'm guessing it is less than 1vpp. if you amplify it like this you will have a 4vpp signal on a 9v supply. it will not clip. you only need a volume knob on the output to bring it back down.

Re: DBA
here is the first half of it


on this datasheet the reverse bias conditions are not listed. does it break down at 9v? we do not know.
http://www.mouser.com/ds/2/302/PMST5088_5089-353460.pdf

check the datasheet here. reverse bias is also called breakdown voltage. it is 30+ volts. so according to the datasheet we should not have any current flowing on 9v supply reverse bias. either the datasheet is wrong about leaking small amounts of current or we have some kind of voltage follower. I would be reluctant to say we have voltage gain but you get the idea. http://www.mouser.com/ds/2/149/MMBT5089-195858.pdf

my best guess would be that the NPN junction may act like some very early transistor designs when used in reverse bias. if you look at it that way, it has similarity to a regular common emitter (collector here) NPN amplifier using a point contact transistor.

[Obulus]

I already played around trying to mod the bitcrusher circuit this afternoon before I read this post. Tried with the 100k loop and 22k to ground at first but it didn't do anything then so I removed the 22k to ground and hooked op a pot instead to see if I could get it to work with another value. Then it worked when I turned the knob, but I didn't have time left to measure the value and solder a resistor in (my bandmates showed up and it was time for rehearsal). Now I won't have an opportunity to work on it until saturday, bummer! Anyway, thanks for the tip, I'm pretty sure I will get it to work now! I don't have a volume knob on it, but I will probably just select the right value resistor to get it to unity (or I may decide to include a volume knob).

Thanks for the information on the DBA, I will have to do some more reading to understand it all (which is cool, I may learn from it)! Hopefully I can read up/understand everything by this weekend so I may have an idea what to do once I have the time!

[eatyourguitar]

from the experts

it relies on Vbe breakdown to 'bias' and is using the reverse current gain

so remember I said I would be reluctant to confirm it has gain? it has gain! but you are basically getting the right transistors at random. it should be the same for an entire batch if one works from that batch. buy different brands from different suppliers.

http://electronics.stackexchange.com/qu ... -operation

[Obulus]

Cool! I'll order some batches this week (or maybe I'll try contacting the guy on tagboardfx who got it to work to see if he still has some transistors of that batch). Thanks again man, really appreciated!

[Obulus]

Finally got round to finish the bitcrusher board in a way that I'm satisfied with. Had to pull an ugly trick to add both resistors though:

Went with a 100k loop and a 200k to ground. I tried out some different values with alligator clips at first and this combination was significantly louder than unity gain (but not too loud) so I decided to go with that and add a volume pot. Drilled the enclosure for the extra knob and hope to spray paint it this weekend and finally have it finished!

Thanks again for the advice eatyourguitar! I messed around with it already before adding the volume knob and I really like the pedal now!

[morange]

Man, debugging is the hard part. Congrats. You're getting into the fun part now, painting outside the lines.

That schematic, eatyourguitar. I'm imagining the nightmare that would be to get to behave predictably.
You have quite some experience torturing bipolars, don't you? I remember reading some stuff you posted about your experiences using them like zener diodes for white noise generators, and burning them in. Fascinating.

[eatyourguitar]

yeah I was told by one of the proper engineers that I was a fucking idiot for trying to get white noise from a damaged silicon junction that clearly only gives burst noise at a non-white spectrum. I learned a lot that day. it is tricky business knowing just enough to ask the right questions while making frequent failures and still trying to look respectable. anyway I got to reading all this.

http://en.wikipedia.org/wiki/Burst_noise
http://en.wikipedia.org/wiki/Shot_noise
http://en.wikipedia.org/wiki/Flicker_noise
http://en.wikipedia.org/wiki/Johnson-Nyquist_noise
http://en.wikipedia.org/wiki/White_noise

and after reading enough of that you start to get hungry for learning calculus and semiconductor physics. so I did.

[morange]

Niiice. I'm not hungry for that yet, lol.