Germanium Qs and heat

[Brymus]

I read over at D*A*M while researching my Tonebender build that Ge Qs are worthless past 70degrees F.
Is that true?
Here in the desert its never under 78 in my house except in winter,summer in the garage temps reach 130+
This is the main reason I havent bought any Ge's for my builds (after reading that touching them throws the leakage into high gear awhile back)
Any opinoins?
One guy had the same idea as me putting Ge pedals in a cooler,but that doesnt seem worth the trouble.

[mac]

I live in a town by the sea where variations in temp are big in minutes, say 15-20C. And gradients through the year are in the same range.
I have all my Ge TBs and FF boxes isolated with thermal carpets I use in my construction sites, and transistors encapsulated in epoxi. Kind of cave fx, but still have an external bias pot.
You can try the rock face.
Or stick to Si. When fine tuned they sound very good.

mac

[PRR]

> Ge Qs are worthless past 70degrees F.

Oh, bosh. We used them at the beach. My dad kept a Ge radio on the dash of his Rambler. 100+F was fairly routine in the day.

It is harder to design with Ge than Si. It isn't that hard to design to run at any temperature you would want to stay in.

If they must run at "uncomfortable" temperature, there are books and books about the subject. Try the Annex at the University library.... most of those books have not been checked-out in decades.

> summer in the garage temps reach 130+

I'd worry more about ME wilting than the transistors.

Keep base-bias impedances down. (Unfortunately, in cheap audio small bias Rs will screw with the intended audio impedances.) Have a 130F hot-box at your bench (or a 70F cool-box in your summer garage), and compromise bias for both conditions.

If you can keep base-bias impedance VERY low, and Vbe-tracking, Ge power transistors will run as hot as Silicon. See any 1959 car radio. The real limit is package seal leakage and long-term reliability. However those very-hot class A car radios used transformers to drive the base, to keep base circuit impedance very low... not suitable for most designs.

[Electric Warrior]

I read over at D*A*M while researching my Tonebender build that Ge Qs are worthless past 70degrees F.
Is that true?
Here in the desert its never under 78 in my house except in winter,summer in the garage temps reach 130+
This is the main reason I havent bought any Ge's for my builds (after reading that touching them throws the leakage into high gear awhile back)
Any opinoins?
One guy had the same idea as me putting Ge pedals in a cooler,but that doesnt seem worth the trouble.

Don't believe a word! That guy was misinterpretating something Analog Mike said about the room temperature in his workshop. Temperature influences the hfe of germanium transistors, so a pedal will sound best at whatever temperature it was biased at.
Some datasheets for germanium transistors represent the specs at 77°F/25°C.

I knew this was going to become a myth..  

[PRR]

> ...78 in my house  ... garage temps reach 130+

Ico leakage doubles every 10 degrees C. Or at your house, it rises by a factor of 8 from 77 F in the house to 130 F in the garage.

We normally calculate bias currents "much larger" than Ico leakage. But if we get too generous, gain suffers (the bias components eat our signal). "Household gear" might be designed so that "much larger" is a factor of 10. Your temperatures give factor of 8 rise of leakage, which eats almost all of the "much larger" safety-factor.

It goes mis-biased.

Since distortion effects are about overwhelming the signal swing the bias allows, the mis-bias leads to different distortion. And easy distortion goes with "reckless" bias, so typically it stops working at all (transistor jammed full-on).

While Delco built radio power stages to work well over 120 F, your 130 F garage may need major re-design of the common Ge pedals. (Unlike GM/Delco, most gitar-toy designers did not care about users in the Mojave.)

There is another path. Play until it gets hot and punks-out. Put a cold beer on it, it will come back. If the beer is too cold, it may go crummy; take the beer off. This is bias by temperature control.

Don't laugh. When Ampex demoed their first tape deck, the exhibition hall was HOT. It stopped working. Someone ran to the bar for a bucket of ice, set it on the heads. It worked again. They sold a bunch of tape machines that day (each one cost as much as a small house).

[petemoore]

One guy had the same idea as me putting Ge pedals in a cooler,but that doesnt seem worth the trouble.
  It isn't worth the trouble if the transistor goes outside operational temperatures.
  All the bias/temperature test measurement and sound quality scrutiny does is show the relationship by providing exact bias/temperature information by which resistors can help compensate for bias-drift for these' transistors in 'that circuit, and also provides info about what temperature to expect failure, have your opamp distorter handy.
 

[Brymus]

Thank You
Everyone gave excellent and very useful answers,very much appreciated 
Just what I wanted to know.