Rangemaster transistor bias vs. leakage

[mc50]

A little while back I've built a Rangemaster kit with a socketed Ge transistor. The transistor I got from the kit vendor is AC125F. Liked it for a while, then it started to bug me a bit - I just couldn't get why people kept raving about this effect and occasionally found the tone a bit shrill (even with using the pedal "properly" - yep, I know it takes some finesse to make this pedal sound its best).

This is the schematic (it's pretty classic).

My first unwise reaction was to just buy 3 more transistors from Banzai (two AC128s pre-tested for gain - one in the 71-80 hFE range, the other 81-90, and the fabled OC44, which is not the legendary-looking model, but I felt is legit).

The OC44 turned out to be the noisiest transistor I've ever heard, so lesson learned. Unusable. To be fair, Banzai does offer a selection service for additional cost for those too but it's not listed on the website and I learned about it too late (I had already placed the order).

None of the transistors biased better (~7V on the collector), as per RG Keen's article.

So I took it upon myself to pull Rb1 and Re (in RG Keen's article) - or R3 and R5 in the Musikding schematic, and put in an 100K and 10K trimpot (not soldered, so I can pull them out and check values after setting).

The best I could do was 82K for Rb1 and ~1K for Re. That gives me ~6.6V with the "best" transistor of my bunch. But RG Keen says: "If  Rb1  is  below 27K  or  above  82K,  or  if  Re  is  below  2.7K  or  above  5.1K, your transistor either has the wrong gain or is too leaky to use, you'll have to use another one."

Do I do as he says? Do I just solder an 82K and 1K? Do I put the old resistors (68K and 3.9K) back and just ignore this?

I've also noticed that BYOC only provides an 100K trimpot. With that configuration, I could have never come close to 7V. Do BYOC as a rule provide better tested, low-leakage transistors?

Do I just hang up my soldering iron and buy a Beano Boost?

Thanks!

[iainpunk]

i think it wouldn't hurt to try a Si transistor (BD140 sounds great, but the pin-out is different, keep that in mind), a tad harsher, but still quite usable, and easier to get the bias right on.
i do recommend keeping the trim pot in there, if you ever decide you want a worse sounding range master for a certain part, especially when biased above 8v, it can get really rough and harsh, which i personally like. if you go higher, you can get it to gate, which is awesome for high gain sounds, no more hiss.

cheers, Iain

[mc50]

Thanks for the reply. Sure, it wouldn't hurt and if they're easily available around I'll try one as well, and yes, the Rangemaster is definitely usable even outside the recommended transistor and resistor bias ranges.

But assuming I just want to hear for myself what the fuss is all about and have a Rangemaster that sounds "classic": should I use a 1K resistor instead of the 3.9K even though it falls outside RG Keen's recommended range (by quite a margin)? I guess my question really is: is that close enough to whatever tone a Rangemaster expert would expect out of a good pedal? Can I still make it work with what I've got?

And since we got here, is there an EU source for low-leakage, tested hFE Ge transistors people here like (preferably inexpensive), in addition to Banzai and Musikding?

[iainpunk]

as someone who has build quite a few range-masters, and played some originals, i don't think you'll hear the difference in tone between Ge and Si when its in the context of a life full band mix, but you'll hear it in a recording and in your bedroom. it still sounds awesome, but slightly different. power Si transistors (like the BD140) seem to sound better than signal transistors in rangemasters.

i have a rangemaster based schematic i call the star master, you can bias it all over the place, but the gain changes with the bias... its a fun booster, but not quite a range master. i used a BD139. (the npn equivalent of the BD140, just flip the caps and polarities)

you can lower the collector pot to 50k, since about 70% of the 100k's turn is usefull, above 3 'o clock its just too gated and misbiassed.

cheers, Iain

[mc50]

Thanks! I guess I'll see if I can find that Si transistor around and keep my eyes open for another source of Ge transistors.

That circuit sure looks interesting, I'll have to try it when I get more parts in.

[iainpunk]

Thanks! I guess I'll see if I can find that Si transistor around and keep my eyes open for another source of Ge transistors.

That circuit sure looks interesting, I'll have to try it when I get more parts in.

it doesn't NEED to be that transistor, just get any power transistor, (it has to be a BJT, not a power FET) but the BD140 and BD139 are really common for lower power applications. they are quite cheap, often less than 25ct. i recommend doing it NPN
(BD139) with normal ground, so you can use power daisy chains.
the last 2 guitar amps i repaired had a BD139 and BD140 as power transistor pair.

cheers, Iain

[percyhornickel]

I´ve build a NPN Ge with the AC127 and just put a lineal 100K instead 68k resistor, works fantastic so you can set the bias to your taste (standar bias collector 7V is very good) and even I put an input cap selector.

[iainpunk]

I´ve build a NPN Ge with the AC127 and just put a lineal 100K instead 68k resistor, works fantastic so you can set the bias to your taste (standar bias collector 7V is very good) and even I put an input cap selector.

OP has that trimmer as well, but still doesn't reach the desired bias.

cheers

[ElectricDruid]

But assuming I just want to hear for myself what the fuss is all about and have a Rangemaster that sounds "classic"

Don't forget that back when Rangemasters were in current production, they certainly didn't go to all this trouble to sort parts to put in them. They got a bag of transistors, fitted them, shipped the product, job done. Even beyond the huge variation in the transistors, the tolerance on the caps and the carbon resistors would have been huge (+/-20% on the caps, and maybe +/-5% or +/-10% on the resistors).

It's no wonder there are stories of musicians playing all the pedals in a box to find a good one. The originals must have varied like hell.

[mc50]

I´ve build a NPN Ge with the AC127 and just put a lineal 100K instead 68k resistor, works fantastic so you can set the bias to your taste (standar bias collector 7V is very good) and even I put an input cap selector.

Sorry for the late reply, Gmail has, for some reason, marked reply notifications as spam and I wasn't getting them in my inbox. Yes, I've done all the legwork for biasing, and I've been able to get around 7V, the problem is this (quoting RG Keen's article):

If Rb1 is below 27K or above 82K, or if Re is below 2.7K or above 5.1K, your transistor either has the wrong gain or is too leaky to use, you'll have to use another one.

If you're needing to go above 82K for the 68K resistor you're in the same boat as me.

[mc50]

OP has that trimmer as well, but still doesn't reach the desired bias.

I do reach the desired bias, but only by replacing the 68K with an 82K (so still within RG Keen's range for that resistor), but I also have to lower the 3.9K to 1K (which puts me out of RG Keen's range for that resistor).

With that combination (82K/1K) I do get close to 7V on the transistor collector.

[iainpunk]

dudes, i advice you ignore those ''proper'' numbers and just change the trimmer until it sounds good! the sound is primary, the actual resistance values don't matter if the pedal sounds good, just go to the full resistance and check if it sounds better and dial back until it sounds the best!!!!

cheers

[mc50]

dudes, i advice you ignore those ''proper'' numbers and just change the trimmer until it sounds good! the sound is primary, the actual resistance values don't matter if the pedal sounds good, just go to the full resistance and check if it sounds better and dial back until it sounds the best!!!!

Well TBH it doesn't sound bad at all like that, it's just that I keep wondering if I'm missing out on anything.

Anyway, I've ordered a small batch of supposedly measured Ge transistors from eBay (no magic mojo transistors, just proper hFE and leakage current), hopefully one of them will bias within RG's resistor limits.

[iainpunk]

i strongly believe ''mojo'' is mainly how it looks visually!
if the parts look sketchy/coroded/funky/old they have more mojo! its just that most ''mojo'' parts numbers are in old style packages.
there is this effect called confirmation bias, which makes metal can CA3140's sound better than the plastic DIP8 versions!

cheers

[PRR]

....None of the transistors biased better (~7V on the collector), as per RG Keen's article.

R.G.'s plan has collector going to 9V. Your plan has collector to zero. Have you accounted for this? Can you get 2V to ground?

Unless you know your gains are way under 50, I would leave the base and collector resistors alone (they also set audio parameters), and diddle the "3.9k" for a happy collector operating point.

[mc50]

Thanks for the post! Not sure I follow, this is RG's schematic:



and this is my build:



The transistor emitter and collector are wired the same way, except that (as far as I can tell) my DYI schematic has ground at the bottom and RG's schematic has ground at the top.

I could very well be wrong, of course. I'm at a quite humble level of knowledge with this.

[iainpunk]

the original Range master had a ''positive ground'', the +of the batt connected to ground, and the power line is -9v. this was standard practice in the Ge era, since PNP's were more common and easier to produce.
you schematic is reworked to make the PNP work with +9v, which means that the 7v RG talks about, should be 2v on yours, because the ground is flipped to the other side of the circuit.

cheers

[mc50]

Gah! Embarrassing stuff. Nope, I've not accounted for this. I do remember that the voltage there was close to 4V, so still off, but I might be able to work with that with the resistors still within RG's limits and use the transistors I have.

Thanks for the help!

[Phend]

Found this on the GGG site for the RangeMaster.
Might be of interest. 

[mc50]

As it turns out, the OC44 I got from Banzai does bias to ~2V with the 10K trimpot (which replaced the 3.9K) lowered to 3.2K. It's still noisy as hell (I hear the frying eggs crackle even with the amp very quiet - it's too late to crank it up now). I'll hear how it sounds pushing the amp tomorrow, after my neighbours wake up.

I think all the transistors I have probably now bias within RG's resistors limits, but I'm still hoping that the ones I've ordered will bias with no special tuning at all (and will hopefully be quiet unlike the OC44).

Thanks for the help, I did find it pretty strange that I had to resort to such drastic resistor changes to get 7V collector-to-ground. 2V is a much more reasonable proposition.