How to measure germanium Hfe TUTORIAL!!! (pcb & spreadsheet)

[slacker]

If you are going to bother testing germanium you might as well put the effort in a do it right.

It's not really a case of doing it right, it's about deciding what level of accuracy is useful for your purposes. R.G's method using the notional values over states the gain by about 1% and that's likely more than good enough for anyone. If you don't need that level of accuracy then you can be looser with the values used, for example 1 Volt out either way on the supply voltage gives a ~10% error, if that's good enough for your purposes then there's no need to tweak it.

[Govmnt_Lacky]

I wonder if the original makers of Fuzz Faces, Tonebenders, etc. (Vox, Fender, etc) squabled about this kind of stuff....

I mean really... do you think they worried about "Exactly 9VDC" or "2.472K ohm resistors" when they built the things?? 

Just get it as close as you can and audition by ear and be done with it! 

[Carlos Best]

I wonder if the original makers of Fuzz Faces, Tonebenders, etc. (Vox, Fender, etc) squabled about this kind of stuff....

I mean really... do you think they worried about "Exactly 9VDC" or "2.472K ohm resistors" when they built the things?? 

Just get it as close as you can and audition by ear and be done with it! 

All said.

Regarding that I was wondering if the new fuzz faces from dunlop take hfe matching into account

[Arcane Analog]

I wonder if the original makers of Fuzz Faces, Tonebenders, etc. (Vox, Fender, etc) squabled about this kind of stuff....

I mean really... do you think they worried about "Exactly 9VDC" or "2.472K ohm resistors" when they built the things??  

Just get it as close as you can and audition by ear and be done with it!  

If you are going through the motions of building a tester, why would it be anything but close? Is getting a 2k4 and a 2M2 resistor and 9V hard to do?

When you are building certain types of Fuzz pedals, including a few you listed, you need to measure them properly if you want them to work properly. Getting them to sound good is a completely different thing. The original builders used certain types of transistors in certain circuits because they inherently had the characteristics required to make the pedal work and sound good. If DIYers want those designs to work well and do not use the original transistor they need to ensure the transistor has the appropriate HFE or the right amount of leakage, whether that be low or high leakage, to work. That, or modify the circuit. Put the wrong transistors or the wrong values in certain fuzzz circuits and you might not even get signal to pass. I have circuits where I want the leakage to be within 25uA +/- of my target.

Edit: For that matter, there are circuits using silicon transistors that have to be measured for the pedal to sound right. Auditioning by ear can be a nightmare if you do not know what you are doing or if you are using untested transistors.

[Govmnt_Lacky]

The original builders used certain types of transistors in certain circuits because they inherently had the characteristics required to make the pedal work and sound good. If DIYers want those designs to work well and do not use the original transistor they need to ensure the transistor has the appropriate HFE or the right amount of leakage, whether that be low or high leakage, to work. That, or modify the circuit. Put the wrong transistors or the wrong values in certain fuzzz circuits and you might not even get signal to pass. I have circuits where I want the leakage to be within 25uA +/- of my target.

Edit: For that matter, there are circuits using silicon transistors that have to be measured for the pedal to sound right. Auditioning by ear can be a nightmare if you do not know what you are doing or if you are using untested transistors.

My point is....

Do you honestly think that Vox, Fender, or other Fuzz marketers actually had a group of people that weighed and measured EVERY Germ transistor for gains, leakage, and loss prior to putting them into their circuits? I seriously doubt it. The fascination of getting the "perfect" matched set of transistors is something that was created out of hype and nostalgia. I know this may be a taboo thing to say but.... really..... do you think one of the above companies wasted THAT much time doing this. Hardly!

As for getting the right sound..... Isn't it well known that Hendrix used to audition new batches of Fuzz type circuits.... Often going through dozens of them.... until he found the one that sounded right? If the transistors were matched and the circuit was built around them as you elude to... then he could have just walked in, tested 2 of them, and known that they had the IDENTICAL properties and moved on.

This is why I suggest auditioning by ear. Pick what sounds best and right to YOU. Don't be so concerned about gains, losses, whatever! Just go for it! 

[Arcane Analog]

My point is that they were picky about their transistors in that they designed the circuits around the transistors. So yes, they were picky, but they did not need to be picky in the same way you are describing. They did not need to be picky the same way someone using a now hard to source germanium device would be. What you are not grasping is that some circuits need lots leakage, need high/low HFE or need a combination elements to work. Try building a good sounding MKI without paying close attention to your transistors.

You see it all the time on these forums.  A prime example is "Why isn't my MKII working? I used all the recommended HFEs!" A common answer is "Do you have enough leakage in Q1?" There is a reason OC75s were used in MKIIs. They have lots of leakage and they bias up perfectly with stock values giving ~8V at Q3C. Try getting that to happen with near zero or low leakage germanium devices using stock values. You can swap out transistors until the end of time and it will not make a lick of difference if your devices are not within the correct tolerance for the circuit you are working on or you do not modify the circuit.

It is not always nostalgia as in your Fuzz Face example - which by the way, if the builder had simply measured the transistors, Hendrix would not have had to go through a large number of crappy pedals to find a good one. That is case in point - thanks for the analogy. Starting with proper transistors is better than blindly swapping them out. Or, you could pick out a set by simply swapping them out only to find that the pedal no longer works or sounds decent when he temperature goes up/down a few degrees. It might have been a good idea to properly measure the leakage, champ. Now you have a pedal that only works properly when the temperature is the same as when you did your blind taste test.

Taking the extra few minutes to properly test your transistors is all about getting a circuit to actualy work properly and not fizzle, gate out, fart or sound extremely thin and harsh. Many fuzz circuits are not paint by the numbers modulation effects where you follow the picture and everything works out. You need to properly measure your transistors, let them settle, listen, etc if you want a decent pedal.

[zombiwoof]

Actually, on Small Bear's site, he has a very good explanation of how to measure germaniums, it's adapted from the GEO method.  I found it the easiest to understand.  It's in the section of articles on the site.

Al

[Arcane Analog]

If you are going to bother testing germanium you might as well put the effort in a do it right.

It's not really a case of doing it right, it's about deciding what level of accuracy is useful for your purposes. R.G's method using the notional values over states the gain by about 1% and that's likely more than good enough for anyone. If you don't need that level of accuracy then you can be looser with the values used, for example 1 Volt out either way on the supply voltage gives a ~10% error, if that's good enough for your purposes then there's no need to tweak it.

I did not say you had to use RG's values to do it right but if you are using RG's values why not use all of his values. I would say that there are circuits where 10% difference may give you trouble.

If a person is willing to invest in the germanium, the parts for the build and building a tester it does not make any sense to deviate from the values provided unless you have the knowledge to compensate.

[Govmnt_Lacky]

It is not always nostalgia as in your Fuzz Face example - which by the way, if the builder had simply measured the transistors, Hendrix would not have had to go through a large number of crappy pedals to find a good one. That is case in point - thanks for the analogy.

I understand some of the points you are making but.... how is the statement above contradictory to the point I was making? In fact, you just amplified the point I was making......

The builders, in fact, DID NOT measure the transistors. That is why he auditioned them.... which was the point I was trying to make.

I was NOT implying that the proper transistors should not be used. In fact, I am in favor of it. What I was trying to imply was that the reason for all of this hoopla regarding weighing and measuring Germs is because of hype and the quest for the perfect sound. Which was mostly because the artists took the time to PICK out the pedals that sounded good. The actual "measuring" was a result of the modern quest for tone..... not because it was the way it was always done.

In the spirit of this, I would say do what sounds good to you. Don't focus on getting EXACTLY 200uA of leakage OR that 70-80Hfe. If it sounds good to you... then ROCK IT!

Hope that is a better explanation 

[Arcane Analog]

It is not always nostalgia as in your Fuzz Face example - which by the way, if the builder had simply measured the transistors, Hendrix would not have had to go through a large number of crappy pedals to find a good one. That is case in point - thanks for the analogy.

I understand some of the points you are making but.... how is the statement above contradictory to the point I was making? In fact, you just amplified the point I was making......

Your statements do not make any sense in the context of the thread and you left out the part of the quote that explains why it does not make any sense. The topic of conversation here is the importance of measuring transistors. You advocate for simply relying on the "just trust your ears" method which suited Mr. Hendrix. That is not good advice nor a good practice - quite the opposite. It is apparent that you have little to no experience with germanium devices.

Let us use the particular case of a Fuzz Face. I can work with that but I would say there are many fuzz designs that are a hundred times more picky about the devices used than is the Fuzz Face - which is pretty damn easy to select transistors for.

If you simply plug in transistors to your Fuzz Face which sound good at your testing time without any other control measures those transistors may be leaky devices which sound terrible with a few degree temperature swing. You sound awesome in the basement but you sound like crap on stage.

What's the problem? I swapped out a hundred transistors yesterday, picked two amazing devices and they sounded great! Now they sound terrible!

Well, you used leaky transistors that sound good at room temp or in a cool environment but sound terrible under the stage lights or outdoors in the sun. That 250uA in the basement quickly became 500+uA in the heat. If you had of measured the transistors first you eliminate the leaky transistors and stick with the on spec devices which will be much more likely to tolerate temperature swings. You then audition the devices that are on spec. Put another way, you can have a great sounding set of leaky devices that become poor sounding devices when the temperature changes a few degrees. In fact, you could very well go from "sounding great" to a pedal is no longer passing signal at all. Auditioning is important but you need to screen the transistor first. I would go as far to say that if you could only chose one method of selection that a builder will have a better chance of success simply measuring the transistor to ensure the values are on spec. For a great sounding pedal that is relatively stable it is essential to do both tests.

Even if the builders in Hendrix's day auditioned the transistors for sound they would still have pedals that sounded poorly when Mr. Hendrix took the stage at Woodstock. If the builders in Mr. Hendrix's time had of properly measured the devices he could grab one off the shelf and have an excellent chance of selecting a great sounding pedal that stays great sounding when he leaves the music shop. In other words, if you measure the transistors you have a greater amount of success for a good pedal and you eliminate a completely avoidable scenario. If a builder simply does your "trust your ears" test you could still end up with a poor sounding pedal the very next day.

The scene becomes even more important with other designs. Try swapping out untested transistors in a MKI or MKII that have "zero" or low leakage. The popular Russian transistors are great for Fuzz Faces but nearly useless in Q1 of a MKII. If you had of measured the devices first you would not have wasted your time plugging them into the pedal and swapping them around. Perhaps you improperly determine you have made a mistake in your build as it is misbiased or not passing signal. If they are all zero or extremely low leakage devices you will never get a good sounding pedal. You may never have anything but a useless pedal that does not pass signal.

Is testing germanium import? Only if you like your pedals to work properly and sound good.

[karbomusic]

Just want to point out that properly working AND sounds good is an assumption and that sounds good is 100% subjective. Obviously there are specs that when stayed within guarantee that certain audial issues won't rear their ugly heads but there is a limit where one must stop measuring and start trusting their ears regardless; the ears always win at the end of the day because their satisfaction is the entire reason for the product to begin with. Tell any musician who doesn't like what they hear "well I measured it and it is in spec" and see what happens.

That being said the trick is to know when which one of the above matters and which one to give precedence to. Doing that well is key. One can spend countless hours taking measurements and be off by a mile as far as the ears go and vice versa. The last thing one wants to do is get hung up on being too like "Spock Logical" and miss the magic; it's good to be both left and right-brained about this and not get too caught up in the minutia. 2 cents.

[Arcane Analog]

My point is....

Do you honestly think that Vox, Fender, or other Fuzz marketers actually had a group of people that weighed and measured EVERY Germ transistor for gains, leakage, and loss prior to putting them into their circuits? I seriously doubt it. The fascination of getting the "perfect" matched set of transistors is something that was created out of hype and nostalgia. I know this may be a taboo thing to say but.... really..... do you think one of the above companies wasted THAT much time doing this. Hardly!

No. They certainly did not. If they did, they would have not had a bunch of crappy pedals and made Hendrix wade through a bunch to find a good one as you mention below. This is why testing is important. If they tested them all Hendrix would have been able to grab and go.

As for getting the right sound..... Isn't it well known that Hendrix used to audition new batches of Fuzz type circuits.... Often going through dozens of them.... until he found the one that sounded right? If the transistors were matched and the circuit was built around them as you elude to... then he could have just walked in, tested 2 of them, and known that they had the IDENTICAL properties and moved on.

You said yourself they did not test Fuzz Face transistors. If they did, Hendrix could have walked in and discovered there was little difference in them. What I said was that certain circuits matched transistors to the circuit or designed around them. I never said they did that with the Fuzz Face although low leakage devices were certainly used. They could have weeded out the bad ones testing them. The prime example for a circuit design based on the transistor are the MKI and MKII Tonebender. They need certain amounts of leakage to get the pedal to function properly.

This is why I suggest auditioning by ear. Pick what sounds best and right to YOU. Don't be so concerned about gains, losses, whatever! Just go for it! 

If you only audition by ear you can easily end up with the Hendrix effect.

[Arcane Analog]

Just want to point out that properly working AND sounds good is an assumption and that sounds good is 100% subjective. Obviously there are specs that when stayed within guarantee that certain audial issues won't rear their ugly heads but there is a limit where one must stop measuring and start trusting their ears regardless; the ears always win at the end of the day because their satisfaction is the entire reason for the product to begin with. Tell any musician who doesn't like what they hear "well I measured it and it is in spec" and see what happens.

That being said the trick is to know when which one of the above matters and which one to give precedence to. Doing that well is key. One can spend countless hours taking measurements and be off by a mile as far as the ears go and vice versa. The last thing one wants to do is get hung up on being too like "Spock Logical" and miss the magic; it's good to be both left and right-brained about this and not get too caught up in the minutia. 2 cents.

Which is why I said this:

Auditioning is important but you need to screen the transistor first. I would go as far to say that if you could only chose one method of selection that a builder will have a better chance of success simply measuring the transistor to ensure the values are on spec. For a great sounding pedal that is relatively stable it is essential to do both tests.

It is important to read all of a post - especially so when you quote one line of many.

[Govmnt_Lacky]

Just want to point out that properly working AND sounds good is an assumption and that sounds good is 100% subjective. Obviously there are specs that when stayed within guarantee that certain audial issues won't rear their ugly heads but there is a limit where one must stop measuring and start trusting their ears regardless; the ears always win at the end of the day because their satisfaction is the entire reason for the product to begin with. Tell any musician who doesn't like what they hear "well I measured it and it is in spec" and see what happens.

That being said the trick is to know when which one of the above matters and which one to give precedence to. Doing that well is key. One can spend countless hours taking measurements and be off by a mile as far as the ears go and vice versa. The last thing one wants to do is get hung up on being too like "Spock Logical" and miss the magic; it's good to be both left and right-brained about this and not get too caught up in the minutia. 2 cents.

AWESOME! Agree 100% 

[karbomusic]

It is important to read all of a post - especially so when you quote one line of many.

Either properly working = sounds good or it is subjective. I don't see where multiple subsequent paragraphs are required to address that as written. However, no dog in the fight, as I said, just pointing out what I felt was a discrepancy. I don't get the argument passion for such a small thing to address to be honest. Carry on!

[Arcane Analog]

It is important to read all of a post - especially so when you quote one line of many.

Either properly working = sounds good or it is subjective. I don't see where multiple subsequent paragraphs are required to address that as written. However, no dog in the fight, as I said, just pointing out what I felt was a discrepancy. Carry on!

Properly working = passing signal, not noisy, not hissy, good range on the fuzz pot, etc. This is objective.
Sounds Good = subjective.

Sounds good and working properly are two distinct features. A pedal can sound great but have little temp stability, range on the fuzz pot, etc.

[Arcane Analog]

AWESOME! Agree 100%  

I don't get the argument passion for such a small thing to address to be honest.

I am curious. Have either of you ever built MKI Tonebender or an FZ-1?

[deafbutpicky]

@arcane analog
I don't get where you're pointing at. Your almost obsessive praise for accurate measurement contradicts  the overall given fact
that leakage is temperature dependent. With fixed variables it's quite easy to determine the result but temperature is like weather... Who knows if Hendrix's tech didn't have a stack of FFs at the gig to choose the best for the evening. Good measurement at a well known place makes it a bit more predictable but it's still a guessing what happens on stage a week later. I'd say go for a ballpark value and cope with a bias pot (maybe with external measuring points) in risk to loose this one sweet spot with the output cap, it' all compromise. Leakage, bias, whatever, I'd like to hear a sound guy say "this place shouldn't take more than 257 people or the resonance frequency about 300Hz gets unpleasant". If you know the weakness and can't make it predictable  make it adjustable within a range, typical engineer's advise...

[Arcane Analog]

@arcane analog
I don't get where you're pointing at. Your almost obsessive praise for accurate measurement contradicts  the overall given fact
that leakage is temperature dependent. With fixed variables it's quite easy to determine the result but temperature is like weather... Who knows if Hendrix's tech didn't have a stack of FFs at the gig to choose the best for the evening. Good measurement at a well known place makes it a bit more predictable but it's still a guessing what happens on stage a week later. I'd say go for a ballpark value and cope with a bias pot (maybe with external measuring points) in risk to loose this one sweet spot with the output cap, it' all compromise. Leakage, bias, whatever, I'd like to hear a sound guy say "this place shouldn't take more than 257 people or the resonance frequency about 300Hz gets unpleasant". If you know the weakness and can't make it predictable  make it adjustable within a range, typical engineer's advise...

A huge point about measuring them first is that they are temperature dependent! Leakage will increase with the temperature - that IS pridictable. Why would you start with a device that has a lot of leakage or not enough leakage? The result IS pridictable and it makes no sense not to make sure you are using the right tools. It is very apparent that anyone that does not think starting off with a good idea about the transistors they are using do not know much about germanium based circuits. That is like saying getting the proper voltages on your pedal or using established voltages as a reference on a circuit is not important. It makes zero sense.

Do you think Small Bear would have a lot of happy customers who pay for transistor sets if they decided to stop measuring the device before they sell it for a specific circuit? Does Small Bear waste their time when they audit they transistors for gain and leakage?

I am saying eliminate the leaky/unstable germs first and then pick them by ear. Have you ever watched an unstable transistor in a test circuit? Measure first, then listen. If you do not think measuring devices is a key component you do not have a lot of experience on the subject matter.

I am saying if you are working on a circuit that needs leakage, you have to measure the germanium device to make sure it will work before you audition the device.

Go look at the main page and see how many people have problems with their fuzz because they have not used the right range of transistor. How many times have I read a post by Lucifer's Trip explaining "leakage is more important than HFE for Q3 of the Buzzaround" or "you do not have enough leakage for your MKIIs Q1" or "you do not want 4.5V on your MKs Q3 collector". There is at least one thread a week with someone having trouble because they try to plunk a random device into a circuit. This is because the person does not have the insight or has relied on someone who does not have the insight. I appreciate the fact that most of the people in this thread rarely use germanium and when you do it is probably a Fuzz Face or Rangemaster. Try building something a little more difficult and then explain to me that measuring is not all that important. Go build a MKI with low leakage transistors and tell me how much fun you are having getting rid of the noise and fixing the decay. 

[deafbutpicky]

Why would you start with a device that has a lot of leakage or not enough leakage? The result IS pridictable and it makes no sense not to make sure you are using the right tools. It is very apparent that anyone that does not think starting off with a good idea about the transistors they are using do not know much about germanium based circuits. That is like saying getting the proper voltages on your pedal or using established voltages as a reference on a circuit is not important. It makes zero sense.

That's the whole point, getting a "good idea" about the transistors is not getting a grasp of the environment they're used at.

Have you ever watched an unstable transistor in a test circuit? Measure first, then listen. If you do not think measuring devices is a key component you do not have a lot of experience on the subject matter.

yes, but measuring gives you an idea about the range to make it work with, it's not absolute.

I am saying if you are working on a circuit that needs leakage, you have to measure the germanium device to make sure it will work before you audition the device.

Again, this is not about the quality of measurement as it will be not so necessary to determine absolute values as they would change anyhow with two more light bulbs on stage. Go easy, make a range of possible values and deal with it.