1n34a diodes

[Gentle_Jack_Jones]

Sorry, I see now. But what do you make of the high (for a Ge) Vf?

I just looked at a datasheet for an 1N34a, and it says the "maximum" Vf is 1 volt. But I was under the impression from various articles that they generally drop .3 v or less.

Color me confused?

[Silver Blues]

Yeah I have genuine 1N34A that are in a DO-7 glass envelope with a single wide black band and visible cat's whisker that measure far less than 0.5xx volts. I'm a bit confused by this as well.

[rocket88]

there were other markings for them, depended on who produced them. two examples i've got some with 2 black bands on one side (like 1n270s), and ones with a single green band. different production times, and different manufacturers.

even though travis already pointed out that looks can be deceiving, but a good rule of thumb is that germanium diodes will never be in DO-35 packages, small Si size. those that are and sold as Ge are always schottky diodes that are work-a-likes. those do sound very good and very similar to the Ge ones, and they're cheap as chips, but they're not Ge.

i've bought the Ge diodes from tayda before, and had some that are good and some that are way out of spec. the price is also TOO cheap to take the risk. i do have a stock of about 1000 1n34a, 1000 1n60, and 500 1n270 diodes, the price was low because I bought them from a distributor for the manufacturer of them, and you know what some are spot on and some are out of spec with a forward voltage as high as ~0.70V. i'll bet that the ones i have are made by the same company tayda gets theirs from, but mine are more in spec then the ones that seem to be floating around. the NOS 1n34a diodes i've got have a forward voltage of ~0.50V, all of the schottky ones i've got have a forward voltage ~0.25V.

the big picture is, just like buying a Ge transistors unsorted, you have an equal chance of them all being spot on or off. just because the part number is right, doesn't mean they will fit the specs. case in point, MP38a IMHO is one of the best and most reliable gain NPN Ge transistor out there. from one seller my last batch of 200 had gains as low as 20hfe and high as 90hfe with a wide spread, and from another that i've dealt with that i just ordered 250 from the gains were all between 60hfe and 160hfe, with the bulk in the 90's. you have to know who you're dealing, what a fair price (ie not too low), which on average is $0.70 each, and someone who is reliable. i only buy from Ge parts from RELIABLE sellers for rare obsolete parts. part of the problem is that with these, since there's a demand by a large group, the prices go up, stock goes down, so the number of fakes go up, and what's left isn't always the best parts as far as stock goes.

like beaker, i highly suggest the same diodes he does, and from the same seller on ebay. for some of the parts we use it's not really worth the hassle of wading through parts that may or may not be in spec, especially to save a buck when there are parts that available for really good prices still, that do as good a job if not better. with the D9 diodes each model has a slightly different tone and are cheap as chips, so you can get a nice stock of each for way less then trying to get the same amount of the the big 3 Ge diodes (1n34a, 1n270, 1n60). shit i usually get D9's, D310/D311/D312, and D18 diodes in packages of 500 or more of each kind and spend ~$100 for all for it all. also, there's a discussion about the topic of real and fake Ge diodes and testing, here. i forgot about the thread so i didn't post the testing results of diodes since the last post.

side note about FETs from tayda. i've bought a bunch over the years from them and so far so good. i've yet to get any that had been out of spec. but, due to what's happened with the FETs that are circulating, once i run out i'm going SMD with adapter boards. much more reliable, and basically the same cost that i paid for all the through-hole FETs i have.

[Beaker]

"Sorry, I see now. But what do you make of the high (for a Ge) Vf?

I just looked at a datasheet for an 1N34a, and it says the "maximum" Vf is 1 volt. But I was under the impression from various articles that they generally drop .3 v or less.

Color me confused?"

You have answered your own question there Jack. If the datasheet says the max Vf is 1V, that means they can measure up to 0.99V and still be considered to be within specified tolerance, and therefore fit to pass the manufacturers inspection department and out for sale.

So you could easily buy two batches from different sellers and have one batch measure an average of 028V (for example), and the other batch average 075V.

Both batches are within specified manufacturing tolerance, so both batches are "good". One fits the norm, one does not, but it does not mean you have bought fakes or duds. Will they perform and sound the same in your pedal? Unlikely.

It's just the inherent variability of Ge components, which is why they were replaced by silicon alternatives. Silicon parts were more far more consistent, far less sensitive to temperature changes, far less fragile and ultimately, cheaper and easier to produce. And they were smaller to boot. Talk about a win-win situation.


To answer your earlier question about Soviet era Germanium parts, I'm afraid it's not that simple. They made some truly fantastic Ge transistors and diodes, and for far longer too - right up to the mid/late 80's, and were mostly intended for military use. (I believe one of the reasons was that Ge parts won't fry under an massive electro-magnetic pulse (nuclear bomb) attack like silicon components would.)

Unfortunately they had their own design, manufacturing and end-use criteria, so are not simply their versions of once common Western parts. Therefore it's not feasible to say "Soviet diode x is equivalent to Western diode y.

What you can say though, is that if your layout specifies a 1N34A, try a D9V for example - or a D311, or others. I can guarantee it will sound good.

There's a massive sticky thread at the top of the Open Chat page titled " Great Jumpin' Germaniums". It's worth a look through as it collates our experiences and experiments with these Soviet parts.

[Gentle_Jack_Jones]

Thanks Beaker. It's good to know, too, that at least my Fuzz Face will join the cockroaches left after a nuclear blast!

[Sensei Tim]

It's been a while and my semiconductor physics is a bit rusty, but Ge is a direct bandgap semiconductor with an bandgap energy lower than silicon (less than 1.1 eV).  I'm pretty sure this would be fried by an EMP.

A wide bandgap material like GaN or SiC might survive due to the significantly larger bandgap