Two Germanium Fuzz Bias Questions that Keep Me Up at Night.

[mattypalm]

Not sure where the best place to post this is...any thoughts are welcome...

1. How does Benson tune a Germanium Fuzz circuit to work at 100 - 110 degrees F, when all my Ge transistors really hate going above 80? And the datasheets usually say max operating temp is 80 or so.
When I first saw the Benson temperature controlled idea I thought it was a bit gimmicky, since it heats but obviously can't cool, and to me a Ge trannie that's too hot is more of a problem than a cold one. But I recently saw on the JHS show that they solved this by tuning it to work at those high temperatures. What kind of Ge transistor / circuit combo would be ok with that? Kudos to them for figuring that out, or finding those transistors.

2. This question is about KingTone copyrights. KingTone apparently have a patent pending on their "Bias Light Indicator" on their MiniFuzz. I have an idea for something along the same lines but definitely not the same. I have no idea what their patent entails - surely it can't be any kind of anything indicating the bias voltage is correct or incorrect? I reached out to them to ask but they haven't responded yet. Anyways I build in such small quantities it probably won't matter but I still wanna check.

(In the guitar pedal builders community I have encountered almost no-one trying to copyright what they do. People steal and mod/improve on things left and right - and everyone's nice about it - and understand when it's appropriate to do so and when it's not cool, so this KingTone thing kinda stumped me).

[PRR]

...Germanium Fuzz circuit to work at 100 - 110 degrees F, when all my Ge transistors really hate going above 80? And the datasheets usually say max operating temp is 80 or so....

Eighty CEE, not Eff.



80C is 176F. You may not wish to play at 176F.

[bluelagoon]

A Patent Pending is not enforceable unless it converts to a complete patent. Patent pending may, therefore, never amount to protection for a provisional patent. However, as there is no disclosure of a provisional patent application to the public, it is impossible to know whether the provisional application will proceed to a complete application before the 12 months comes to an end.

Could be just smoke and mirrors to deflect others getting onboard with such a concept. As converting to a full Patent protection is very costly, and would have to be weighed well in favor for any potential projected profits.
Not to say they wont take up the full Patent, but all too often these preliminary patent pending's, fizzle to nothing, But are in place just long enough to give the innovators the best foot in the door and hence market leadership and brand recognition with the new novel ideas.

[amptramp]

A germanium transistor circuit can be roughly stabilized by having a reverse biased germanium diode from base to emitter.  As the temperature goes up and the transistor collector to emitter leakage increases, the diode leakage will increase also, shunting the extra bias current away from the base.  This was a trick used in early transistor radios to extend the temperature range.  As long as you had a decent match between leakage currents, it sorta worked.

There is nothing saying the bias light indicator has to use the same technology as the signal path.  If you have something like a fuzz face, you could put a window comparator on the first or second collector voltage to light up when the collector voltage is between specified limits.  I don't know of any adjustment you can make if the bias light is not on, but you can at least know if you are within the correct operating parameters.

[mattypalm]

...Germanium Fuzz circuit to work at 100 - 110 degrees F, when all my Ge transistors really hate going above 80? And the datasheets usually say max operating temp is 80 or so....

Eighty CEE, not Eff.



80C is 176F. You may not wish to play at 176F.

Haha, thanks, good to know! I grew up on metric but now live in the US so it's a bit of a jumble in my head - and when I see 80 I guess I automatically read Fahrenheit even when it's not. I do find the characteristic of the sound can change pretty dramatically once you go above 80 F though.

[Phend]

You may not wish to play at 176F.

Unless you are in Finland playing in a Sauna band.
It's the jump thru the ice after the makes many Ge's go dead.

[PRR]

I dunno what you need a bias light for. Generous design leads to very self-stable circuits. The BBC did this for remote amplifiers. Battery like 10X the signal swing and base networks carrying like 10X the worst-case base current.

Here is 48V (a standard switcher voltage) and 75uA bias bleeder for 2uA base current (hardly optimized but it is late). Over -40C to 80C the collector point varies much less than a Volt. Over a player-practical (if uncomfortable) 20F (air over Phend's sauna pond) to 104F (Mojave desert in the shade) less than a half-Volt on about 25V idle point.


The transistor model has a way-high Is value so it leaks like Germanium.

Yes, you could rip-off a verse of "Auld Lang Syne" at -20F but not a lot more. Very few players could get through "Summertime" at 104F without fainting. (At 80C your skin will blister in minutes.)

[bluelagoon]

Benson Germanium Fuzz Teardown

https://www.youtube.com/watch?v=UfqdW3_kuZw

Schematic:

http://guitar-fx-layouts.238.s1.nabble.com/file/n49945/868551CB-EECF-416D-9A07-37FBCCA1FA27.png

[Phend]

Would a bias adjust knob remedy this temperature issue ?

[amptramp]

I suspected they might use a comparator that was not in the signal path and they did.  Instead of the pair of 68 ohm resistors used as a heating element, you could use a reverse polarity comparator to drive a Peltier-effect cooler that will keep the transistors colder than ambient.  It keeps the hfe low but it reduces the noise and aging of the transistors.

[Phend]

Hum, This from Amazon...12V TEC1-12710 Thermoelectric Cooler Peltier by Pinkcoo
Interesting...But it looks like condensation of h2o can be a problem.

Short and simple. the Peltier 12710 worked flawlessly even at output a bit above its rated 154 watts. Im using it for scientific testing now and have it hooked up to a monitored LAB power supply and monitored its reaction at different ramped power levels (from 3v-12v). At 3v it can be used without a heat sink and you can easily find the cold side as the hot side just gets to about 100 degrees (f) which is well below the tjmax temp. Also I noticed was the side with the model number PRINTED on it is the cold side - I ordered 2 and its the same on both.

Does is freeze? Ohhh YES! I'll post some images. I am able to have freezing temperatures within 10 seconds. Water droplets will freeze at this point in seconds (which is neat to watch!). I highly suggest these for a science class as their inexpensive, very safe and teach so much. The hot side does get extremely hot (not scolding though with a heat sink just about 150(f)). I used a dedicated PC heat sink from TigerDirect (the $15 model worked fine but a stock free AMD FX heatsink worked even better). With a good heat sink it just gets warm not hot. I mounted a heat sink to the cold side as well and the condensation in the air builds up very very quickly. My LAB power eveuntally was maxed out and I used a dedicated 400 watt PC power supply and used the CPU/Motherboard 12v power connector for the high amperage and it worked great.

My original intent was to use the peltier for my two GTX 670 SLi video cards to cool them. I thoroughly tested this peltier on old video cards to monitor their reaction on fail safe equipment, and to be honest they built up so much condensation that when the computer would be shut off the water will begin to drip off the GPU's processor (I live in Florida so this does not help either). So not good. But now Im having a blast getting the efficiency as high as possible testing different cooling methods. If your interested still in using this on your CPU just be forewarned that the thickness of the peltier (about 5mm) might be too much to be able to mount your CPU so be conscious of this.

[duck_arse]

https://www.taydaelectronics.com/catalogsearch/result/?q=peltier

3$ and not 150W.

[Frank_NH]

There is also the John Hollis Rock Face:

http://www.hollis.co.uk/john/circuits.html

"Rock Face:
I've taken the classic Fuzz Face circuit and added optical temperature compensation. This ensures that the output stage is biased at half the supply voltage regardless of variations in the transistors, particularly those caused by temperature. Tweaks: Obviously, one could make the bias point adjustable, but it sounds fine set to halfway. One could add a resistor across the LDR to set the maximum value which would also cut current consumption slightly as the LED would go darker in response. This achieves lower minimum values too. Try 47K."


I'd like to experiment with this on a breadboard, but then I'm sure it would sound like a fuzz face most of the time...

[mattypalm]

I dunno what you need a bias light for. Generous design leads to very self-stable circuits. The BBC did this for remote amplifiers. Battery like 10X the signal swing and base networks carrying like 10X the worst-case base current.

Here is 48V (a standard switcher voltage) and 75uA bias bleeder for 2uA base current (hardly optimized but it is late). Over -40C to 80C the collector point varies much less than a Volt. Over a player-practical (if uncomfortable) 20F (air over Phend's sauna pond) to 104F (Mojave desert in the shade) less than a half-Volt on about 25V idle point.


The transistor model has a way-high Is value so it leaks like Germanium.

Yes, you could rip-off a verse of "Auld Lang Syne" at -20F but not a lot more. Very few players could get through "Summertime" at 104F without fainting. (At 80C your skin will blister in minutes.)

I can't say I've looked at this long enough to understand it fully yet (is it directly applicable to me just using a couple volts? How do I incorporate generous design in my case?). I intend to understand it though. I'm a slow but thorough processor of information, haha. My experience was this: I built many germanium fuzzes (usually fuzz face style circuits) - played them, sold them etc. and didn't have any issues really. But recently a buyer of one of my pedals complained about the bias and returned it and it kinda made me a little more paranoid than I was before (that, plus crazy changing Texas temperatures and an AC that wasn't coping well). But after looking at the returned pedal I concluded it might have been user error, or he had a super hot house, because the pedal seemed fine to me. The last 20 or so pedals I've built have all had an external bias control (for Q1...I found a Q1 bias change did more interesting things than Q2). But I did find that when the first transistor in the fuzz face circuit is biased at 0.7V, temp can cause it to go past 1V and this could be a serious change to the sound. But anyways I have an external bias control for it - it's easy to bias it to maximum saturation just by ear (provided Q2 is right). Q2 can also change but it's less dramatic. Now I want to be absolutely sure nothing can go wrong, and that people can bias Q2 without a multimeter, just a turn of a trim pot and an LED that changes color. Maybe it's unnecessary. There have been a lot of people talking about this problem on the internets which also could've amplified the problem in my mind, ha.

[mattypalm]

Benson Germanium Fuzz Teardown

https://www.youtube.com/watch?v=UfqdW3_kuZw

Schematic:

http://guitar-fx-layouts.238.s1.nabble.com/file/n49945/868551CB-EECF-416D-9A07-37FBCCA1FA27.png

Thanks for these! Super interesting. Am I right in saying they don't really have anything special to "tune" it to the higher temperatures? Maybe it is just a case of heating the transistor to the higher temperature, and then biasing it. Now I am enlightened and know that germanium transistors can technically be heated to 176F with no issues.

Thanks to everyone here for your wise thoughts! It's clear that I posted this question in the right place.

[bluelagoon]

I think you may be on the right track there saying " Maybe it is just a case of heating the transistor to the higher temperature, and then biasing it." That's likely all there is to getting it setup right to a specified pre set running temperature.
Good luck with it,. Also gave me a lot to contemplate, since I have been looking to delve into eventually the germanium fuzz type pedals.
Cheers.

[bluelagoon]

Probably a good way to go is putting in a digital volt meter as per the video in previous post above by Phend.
Check them over at Aliexpress out of China, there is a plethora of various shapes, sizes, and styles, all very well priced.
That's the way I will likely be implementing when and if, one day Lord willing I get to putting the germ fuzz boxes together.

https://vi.aliexpress.com/w/wholesale-digital-Voltmeter.html?SearchText=digital+Voltmeter&catId=0&initiative_id=SB_20230618173440&origin=y&spm=a2g0o.detail.1000002.0&trafficChannel=main&g=y&page=2

https://vi.aliexpress.com/item/1005005686039338.html?spm=a2g0o.productlist.main.9.711e521a3NHXnx&algo_pvid=a032d167-3abe-40c6-8b81-b77adc90aca2&aem_p4p_detail=202306181839228034105410573400014746207&algo_exp_id=a032d167-3abe-40c6-8b81-b77adc90aca2-4&pdp_npi=3%40dis%21AUD%211.59%211.59%21%21%21%21%21%40211bf48d16871387624395503d07bc%2112000034007744664%21sea%21AU%21712531695&curPageLogUid=oBOVra25PlhY&search_p4p_id=202306181839228034105410573400014746207_1

[Gus]

In 2002 Joe shared this circuit with a voltmeter
https://sounds.ampage.org/files/nitroburner_schem2.jpg

[Phend]

Work in process



Get fancy with one of these
Stick on the box, fish not included

[Electric Warrior]

The last 20 or so pedals I've built have all had an external bias control (for Q1...I found a Q1 bias change did more interesting things than Q2). But I did find that when the first transistor in the fuzz face circuit is biased at 0.7V, temp can cause it to go past 1V and this could be a serious change to the sound. But anyways I have an external bias control for it - it's easy to bias it to maximum saturation just by ear (provided Q2 is right). Q2 can also change but it's less dramatic. Now I want to be absolutely sure nothing can go wrong, and that people can bias Q2 without a multimeter, just a turn of a trim pot and an LED that changes color. Maybe it's unnecessary. There have been a lot of people talking about this problem on the internets which also could've amplified the problem in my mind, ha.

The interesting thing about Q1 is that it sets the bias on Q2. Q2's specs make little difference for the collector voltages. But by tweaking Q1's collector resistor one will set the voltage on Q2's collector as well.

Another thing that is often underestimated is the fuzz pot's tolerance. It can make a great difference and I believe this is why the tone of fuzz faces can vary that much.