Titania 2 Transistor Choice

bobson dugnutt

Active member
Quick question, which I'm sure will seem a little silly in retrospect, but I'm a bit at a loss:

I just finished up a titania 2 build, and everything works and all, but..................it does not sound much like the original. For lack of a better descriptor, I would describe my build as "farty," but in a bad way, not in a badass oldschool fuzz sort of way. While it's supposed to be sensitive to input gain, I'm finding that turning the guitar volume from below 9/10 silences the damn thing. Compared to some of the demos I've been watching, it seems like there's less treble, even with the tone control dimed.

Since they're readily available and cheap, I went with 2n2222a's for all three transistors, which I figured would be fine since the build docs (and the effectslayouts version of this) list them as an option. I figured the Si transistors are close enough in tolerance measuring the hfe for all of them wouldn't be necessary and just kinda went for it (regretting not socketing them now).

So I guess my question is a two-parter:
- Has anyone built one of these and experienced anything like this?
- Any suggestions for substitutions or obvious stuff I missed with these transistors?

Thanks in advance! I feel like I'm losing my mind on this one because comparing my own sound to demo videos is kind of an apples and oranges situation. My gut feeling is something is wrong, as this is not how any fuzz I've played sounds/responds.
 
Building one right now, I’d like to know the answer as well. Further, did you use a 27nf cap for C7, or a 22nf/33nf?
 
Last edited:
I believe not measuring and testing your transistors could be the issue. You really need to make sure they are between 150-200 HFE (180-190 is ideal, as per measurements of the original), I’ve had some 2n2222s be way above spec. Ya gotta be sure!
 
Thanks for the reply! Still learning the subtleties of transistors in fuzzes; this absolutely has to be the issue! Is the 150-200 HFE a recommendation for this specific circuit, or would that apply to a good chunk of other fuzzes with a similar architecture as well?

edit: revisiting the build guide it _does_ specify low gain! that explains the 150-200 recommendation I believe
 
Every pedal has different transistor requirements. The ones with a stable bias circuit aren't too picky about HFE. The biasing in this pedal is very picky about transistor HFE.

"Low gain" doesn't mean much. Specifying HFE without specifying collector current isn't enough either. I have one of these in the queue and I'm gonna breadboard it first so I can hand pick the right transistors. All three will influence each other.

Nostra can chime in, but my guess is you want the collector of Q2 & Q3 to be somewhere between 4.5V and 6V.
 
It’s been awhile since I’ve had it, I don’t recall the collector voltages but you are most likely correct Chuck!

EDIT: The deadendfx guys have some measurements from an original in their build docs for the same project. No input signal, all controls at noon.

2E01F9CA-0228-4712-9285-A108915FC37C.jpeg
 
Makes sense! Thanks for all the info. Kicking myself for not socketing the transistors when I built this (what feels like ages ago). Will hopefully get that going and take some measurements soon!
 
Mixing NPN will it make a huge difference or should they all be the same? For example 2N3904 and two 2N222A.
 
It's not about part numbers, it's about specs. You can shoot for the numbers Nostradoomus posted above as a starting point. If you get within ±20% that should be good enough.

I think you mean 2N2222A.
 
My DMM and Transistor tester is wrong as everyone of my NPNs are 280+ hFe.. My 2N2222A are all 480+ hfe. I got some from Tayada, Mouser, Newark, and Amazon.
 
DMM I don't trust to measure HFE accurately. Not sure why your transistor tester would be wrong. What kind of transistor tester?

It's easy enough to gin-up an HFE measurement with a breadboard, a 9V supply, a DMM and two resistors.
 

LCR-T4 Mega328 ...

My DMM is being a pain and will have to get my fluke from work. For some reason it's not reading µA and the mA is sporadic. I'm using a 470K on the base with a 9v input. Going from E to GND and then putting 9v into the C to GND and not getting a good reading either. Tried voltage and convert it to mA, but that's kicking my ass as well.
 
Try 1M from +9V to base. 1K from +9V to collector. Emitter to GND. Measure the voltage across the 1K resistor. 1V = 1mA. The base current will be approx 8.6μA with a 9.2V supply. HFE = Ic / Ib. If you measure 2.15V across the 1K res, then Ic = 2.15mA. HFE = 2.15mA / 0.0086mA = 250. This method only works with low-leakage (Silicon) transistors. Polarities given are for an NPN transistor. Reverse the voltage for PNP.

You have to have a current-limiting resistor in series with the collector, otherwise you risk burning up the transistor. The 1K resistor protects the transistor.
 
Try 1M from +9V to base. 1K from +9V to collector. Emitter to GND. Measure the voltage across the 1K resistor. 1V = 1mA. The base current will be approx 8.6μA with a 9.2V supply. HFE = Ic / Ib. If you measure 2.15V across the 1K res, then Ic = 2.15mA. HFE = 2.15mA / 0.0086mA = 250. This method only works with low-leakage (Silicon) transistors. Polarities given are for an NPN transistor. Reverse the voltage for PNP.

You have to have a current-limiting resistor in series with the collector, otherwise you risk burning up the transistor. The 1K resistor protects the transistor.
My LCR-T4 is fubar'd
2N2222A with the meter hfe = 431, using the bread board = 270
2N3904 with meter hfe = 430, bread board = 127
BC337 meter = 430 - bread board 244
S8050 meter = 420 - bread board 160
 
HFE varies with collector current. You can get two different HFE reading and both can be correct.

Does look like your meter is OTR though. Does yours have some SMD pads for measuring SOT-23 package devices? Mine does and the pads got dirty which threw off the readings. A little IPA and a cotton swab fixed it right up.
 
Yes, I have the SMD pad. I will clean it up and test it again.

I guess the 2N3904 is too low of a gain for this pedal?
 
Not necessarily. I just pulled a 2N3904 from parts tray and it measured HFE = 151. I used a couple of medium-low gain trannys for Q1 & Q2 in my build. K Pedals gave them to me. I don't recall measuring those guys, but I just measured a spare one and it came out HFE = 140. I colored outside the lines just a little bit and used an MP38A for Q3. All of my MP38As measure between 60 and 80 for HFE. It sounds great.

I'd suggest either breadboarding it or install sockets so you can audition transistors.
 
Back
Top