What do I need to know before I order my first transistors?

"Zero leakage"
Umm?
Some transistor testers, such as the Peak DCA55, can’t measure less than 9 µA of leakage using their test parameters. If there’s less than 9 µA of leakage, the DCA55 reports leakage as 0 µA. There are several Soviet transistors that frequently have leakage that low. For example, at 70°F, about half of my GT308, GT309, GT310, and P416 transistors read as having zero leakage.
 
You inspired me to make progress on an abandoned project. This is the Fuzz Face half of an "all you need" classic drive section. The other half will be a Rangemaster
 

Attachments

  • PXL_20240212_041305965.jpg
    PXL_20240212_041305965.jpg
    559.1 KB · Views: 13
I just wanna say give Germanium a chance.

Once I was able to distinguish between germanium and silicon fuzz, the latter became unusable to me, for the most part. Too grainy and fizzy. Especially on cleanup.

However I don't claim to be able to tell the difference every time and I'm ready to be stumped and humbled.

It's it fun to explore and find what you like.
 
@MBFX StompBox won't ship to France unless I have an EORI number and the French website tells me that I don't get one if I don't have a business... stuck in red tape. So, your source is safe (at least from me ;))
 
@MBFX StompBox won't ship to France unless I have an EORI number and the French website tells me that I don't get one if I don't have a business... stuck in red tape. So, your source is safe (at least from me ;))

Your English is so good! I had no idea English was not your first language, and I assumed you were from the USA. I'm sorry about that, and doubly so if you mentioned it elsewhere. Your most reliable source might be eastern Europe, via the EBay sellers other builders mentioned.
 
I've gotten some good Ge transistors from Ebay lots... amid a majority of low gain and/or leaky ones. It's nice to have a stash and the low-gain ones can be put to some use, but I'd look for tested single transistors, or sets. For example, these always seemed like a good deal:


I haven't bought any from them so I can't vouch for them. I've gotten a few tested Ge transistors from Cameo's Components on Ebay. It works out to around the same cost per "good" transistor.
 
I didn’t know much about fuzzes, but as a first step I knocked together three Ge fuzzes from Aion and they all work perfectly and sound great. From there you can start tweaking, making adjustments that you might like etc. Or just play ‘em.
 
Ge fuzzes can really test your sanity, especially if they are going into a MK1 Tone Bender. I spent $40 on a transistor set from Small Bear and my MK1 sounded pretty "meh". Thankfully a very generous member here took the time to sort through his stash and test a bunch of Ge transistors until he found three that he thought would work well. What he sent ended up sounding fantastic!

Depending on the circuit you may end up spending quite a bit of time swapping transistors around until you find a combo that sounds right to you. This means buying several "matched" sets for whatever circuit you are working on or several lots and testing them yourself.

So far I've built a Ge Hudson Broadcast clone and a MK1 Tone Bender. Next will be a Ge Fuzz Face and then I'm more than likely done building Ge pedals.
 
If these are your first builds, I'd hit the pause button for a second.
Ordering a set of pre-matched transistors for those builds, or sticking to silicon, would be my recommendation. Maybe even asking nicely here and someone would set you up with a pair. Do you have a means/knowledge to measure/match leakage?
I'll echo @MBFX , Ge is a huge pita. So much so that, personally, I'm almost over it. It's almost not worth the financial turmoil and time to me at this point. I would also caution, that if you do go down the Russian Ge road, take notes and ask questions about which ones to order for which circuit. Even then, you may order 40 and find only a few, at best, that hit the needed specs. You can't really trust the published specs on sovcom's listings. Maybe I've just have bad luck but I've bought several lots that all measured significantly lower than published specs. One in particular that measured in single digits hfe or diodes. I'm fairly sure most have been pre sorted/were rejects. This is the case for most Ge transistors today. In other words, a lot of them are useless.
Silicon fuzzes can sounds awesome, it doesn't have to be Ge. And it's way cheaper to get a variety of Si transistors.

TLDR:
Either order from a person/vendor that will sale you measured and matched setd(gain and leakage) or stick to Si for your first couple. It's very easy to waste a significant amount of money and time going deep down the Ge rabbit hole. I say all this because I don't want you, or any other new builders reading this in the future, to get frustrated early in your building career trying to chase the mojo ghost. And to answer your second set of questions regarding other mojo parts, no. Stick to new stock known good parts. You won't hear a tropical fish cap in a cranked fuzz. You will waste extra money and have to deal with the pita of trying to get that capacitor to fit the board/in the enclosure. Again, no need for the added frustration and cost starting out.
That's just like, my opinion man.
Happy building.

Edit:
P.S., I wouldn't trust Small bear to match any transistors at this point in time. I would go to pedal hacker before small bear in this instance.
? Curious about the comment as relates to Small Bear. We continue to offer measured transistors and sets, and Pedal Hacker is us too...
 
? Curious about the comment as relates to Small Bear. We continue to offer measured transistors and sets, and Pedal Hacker is us too...
I’m not sure what @jwin615 was referring to, but the one time I purchased a set of measured transistors from Small Bear (for a fuzz face), I received transistors with gain and leakage values that were nowhere close to being in spec for a Fuzz Face. I tried them in a PedalPCB Twin Face and the germanium side sounded weak and gated. Later, after I bought a DCA55 and pulled the transistors out to measure them, I found that they were not suitable for a Fuzz Face. Q1 had an hFE of 115 and 300uA leakage, and Q2 had an hFE of 178 and 350uA leakage at room temperature. I don’t know how typical that is, as I stopped buying anything from Small Bear after that experience. But I’d be surprised if I was the only person who had that happen to them.
 
I’m not sure what @jwin615 was referring to, but the one time I purchased a set of measured transistors from Small Bear (for a fuzz face), I received transistors with gain and leakage values that were nowhere close to being in spec for a Fuzz Face. I tried them in a PedalPCB Twin Face and the germanium side sounded weak and gated. Later, after I bought a DCA55 and pulled the transistors out to measure them, I found that they were not suitable for a Fuzz Face. Q1 had an hFE of 115 and 300uA leakage, and Q2 had an hFE of 178 and 350uA leakage at room temperature. I don’t know how typical that is, as I stopped buying anything from Small Bear after that experience. But I’d be surprised if I was the only person who had that happen to them.
We're certainly sorry to hear of that experience; I don't recall seeing a customer service ticket related to it but I'll ask the team, and your post does not refer to a timeframe. Small Bear changed ownership a few years back. As everyone here knows, germaniums can be, well, inconsistent and we do our best with tools and measurement techniques and a long track record of happy people. It's interesting to note --one issue that does come up is when customers measure our sets and get different results; I am not suggesting that this is the case here, but often we'll take a set back and find that it measures the same way here it did when it went out. In this case, you also built the pedal and it sounded wrong. But listen-- a satisfied experienced pedal builder customer is more important to us than the cost of a transistor set; please give us the opportunity to win back some of your parts business.
 
We're certainly sorry to hear of that experience; I don't recall seeing a customer service ticket related to it but I'll ask the team, and your post does not refer to a timeframe. Small Bear changed ownership a few years back. As everyone here knows, germaniums can be, well, inconsistent and we do our best with tools and measurement techniques and a long track record of happy people. It's interesting to note --one issue that does come up is when customers measure our sets and get different results; I am not suggesting that this is the case here, but often we'll take a set back and find that it measures the same way here it did when it went out. In this case, you also built the pedal and it sounded wrong. But listen-- a satisfied experienced pedal builder customer is more important to us than the cost of a transistor set; please give us the opportunity to win back some of your parts business.
Thanks, I bought the set in January 2020, which I think predates the ownership change. I didn't bother putting in a ticket at the time because I had chalked the issue up to user error—after all, germaniums are finicky, I had used sockets, and it was only my fourth or fifth build. It wasn't until I decided to debug the build three years and 50+ builds later that I discovered that the gain and leakage values were out of spec. I don't have the transistors anymore, otherwise I would gladly take you up on the offer to return them.
 
? Curious about the comment as relates to Small Bear. We continue to offer measured transistors and sets, and Pedal Hacker is us too...
I sent a very detailed email about this to which you never replied. No sorry. No f*** off. Nothing.
This was a year after ownership change. I haven't given you guys a dollar since.
The quoted comment was pre-pedal hacker acquisition as well.

I'm not trying to kick up dust but you came here and necro-quoted my post so...
 
My 2C worth - when buying transistors don't get sucked into the whole "mojo parts" thing. Builders like Analogman will tell you that NKT275s are the only true transistors until they run out and then they'll find another holy grail transistor which is the only one. I've built some of my favourite Ge fuzzes with boring old Russian MP38s. If you use sockets on the board you can just go through all the transistors you have until you find some that sound best to you.

The biggest problem with Ge transistors is all the snake oil. I do wonder if it has been worth all the trouble building all these old school fuzzes. I rarely play them!
 
use sockets on the board you can just go through all the transistors you have until you find some that sound best to you.
this or breadboard the circuit and then you can solder in your picks nice and neat with no sockets.

I do wonder if it has been worth all the trouble building all these old school fuzzes. I rarely play them!
curiosity is a strong force i guess 😅
 
this or breadboard the circuit and then you can solder in your picks nice and neat with no sockets.


curiosity is a strong force i guess 😅
It may be worth looking at the RG Keen piggyback method. I have had success with this way of getting Si T's to sound great. This method allows you to use two T's with a trimmer to get the lower Hfe you need. Then they sound great. It's the lower Hfe that makes for a great pedal. You all know that.
 
I’m not sure what @jwin615 was referring to, but the one time I purchased a set of measured transistors from Small Bear (for a fuzz face), I received transistors with gain and leakage values that were nowhere close to being in spec for a Fuzz Face. I tried them in a PedalPCB Twin Face and the germanium side sounded weak and gated. Later, after I bought a DCA55 and pulled the transistors out to measure them, I found that they were not suitable for a Fuzz Face. Q1 had an hFE of 115 and 300uA leakage, and Q2 had an hFE of 178 and 350uA leakage at room temperature. I don’t know how typical that is, as I stopped buying anything from Small Bear after that experience. But I’d be surprised if I was the only person who had that happen to them.
Just a sanity check. I have read before that Germs should test lower than 500uA leakage for a fuzz.
 
Just a sanity check. I have read before that Germs should test lower than 500uA leakage for a fuzz.
Depends on the fuzz. I’ve built Tone Bnemder Mk. II’s with Q1 and Q3 that leak 500-600 uA and sound phenomenal. Likewise, I’ve used AC187 and GC521 transistors that leak over 1mA in Tone Reapers, and they sounded really good, too. I’ve heard of people using extremely leaky transistors for some slots in the Tone Bender Mk. I, germanium Fuzzrite, and FZ-1, among others. So I think that you can discard that rule. Breadboard before you build and trust your ears.
 
Back
Top