hello!

my idea to reduce supply voltage is to use germanium transistors. they have a knee-voltage of ca. 0.2 v, so supply may be about 0.4v without saturation.
an idea to prevent saturation ist to use valve diodes with a reduced heating voltage so the knee voltage is less (diode valves have a negative knee voltage). so i think its possible to further reduce supply voltage to 0.2..0.3v.

shurely, ge-transis are not welcome today, because they have more parasitic current loss (this is because device death temperature is about 90°C). but this problem can be solved by freezing down the device with forced cooling to eg. 0°C.

you may probably think my ideas are a bit funny or esoteric, but i like to discuss about unconventional things, often they have a solution to provide standard technolgies with new features.

walter nussbaum, dipl. electronics ing. eth,

 

What do you mean exactly?
Reduce the supply voltage for what?
What are the improvements compared to a current design?

 

Well the trend in recent years is to continuously lower the operating voltage of high density devices such as Intel and AMD microprocessors. They run at less then 2 volts now I believe. This mostly to lower the operating temperatures as the transistor density and clock speed continues to increase.

So far silicon has been able to keep up with this trend but there other materials being researched all the time. So have fun researching this topic but be aware that there are millions and millions of dollars actively working this topic daily.

Lefty

 

ps my idea is to use low supply tech for devices who need a battery or solar cell that must be small and deliver power for long time.
another idea is to use sub-mini-tubes with very low current an cold cathodes or esaki=tunnel-diodes, a very old idea which i think is interesting to revive today also in computer applications. walter, dipl. ing. eth,

 

What do you mean exactly?
Reduce the supply voltage for what?
What are the improvements compared to a current design?

hello mik3!
well, i think low supply is intersting in small devices needing batteries or solar cells or such things. if it's possible to also use my idea to speed up eg. switcher-devices as digital logic i don't can say.
what du you mean with 'current design'?
thank u for ur answer

 

Hello Walter, welcome to AAC.

Keep having the ideas.

Unfortunately germanium has had its day as a useful material for most purposes. The higher voltage required by silicon is no longer an issue and even better semiconductor materials are being developed from a variety of substances.

Just compare my old Vidor or Roberts portable radio with a modern MP3/Radio/pendrive. The old radio had germanium transistors, with their lower Vce, but had more total leakage and quiescent current than the MP3 uses at full throttle. Further in order to set up bias chains etc a 12 volt battery was needed. It looked like a small car battery!
In contrast the MP3 works off a single 1.5 volt AAA cell. Necessary higher voltages are internally generated from this and the total leakage of the silicon inside is so small that the overall power drain is miniscule compared to the Roberts.

Remember also that germanium is a highly toxic substance, silicon is not.

 

A lot of devices, even though they use silicon, don't have any dropping voltage as such. They don't use doping as it was originally practiced with diodes and bipolar transistors. Think MOSFETs, which primarily depend on the thinness of the layers, and JFETs, which do have a junction, but use it permanently back biased.

I've had the same thoughts though, it's fun to experiment. You want low voltage parts, check out Schottky diodes sometime.