RangeMaster Temp Comp Diode

[Phend]

Hello:
I see Fuzz Central has a modification on the Dallas RangeMaster.
It is simply adding a "temperature" compensating diode from the Ge Transistors Base to the Emitter.
The diode is a 1N34.
Any comments on this modification?
Is it as simple as that or does the diode need to be "selected" for some value?
Is it good to add?

[PRR]

diode is a 1N34.
Any comments on this modification?

It is 2023. What is a "1N34" these days? Unless you got it from a stash your daddy sealed in 1963, it probably isn't a leaky Germ.

[bluelagoon]

Further Discussion pertaining to, Found Here -

RangeMaster Temp Comp Diode

[Phend]

Sounds simple enough, 2 solder connections, lol. I do have a few old (~1980) RS 1n34s. More than likely won't was em here.

[mozz]

IF you had another transistor the same you could use that instead. It would be a closer match.

[amptramp]

In answer to one of the OP's questions, the diode leakage current would have to be selected to match the  collector-to-base leakage of the transistor.  I like the idea mozz has of using a reverse-biased collector-to-base junction of a transistor that had the same characteristics as the transistor you are using as an amplifier but leakage currents are all over the place even within a transistor batch, so you would still have to match them.

[mzy12]

Tonemaster MKIII does this as well, also allegedly to account for temp variation. How effective  is this with modern Ge diodes like the 1N34A? Like PRR said, they're not very leaky nowadays. It would definitely be a terrible waste to put in another identical germanium diode to act as a leaky Ge diode though, even if it would be theoretically maximally effective due to the (hopefully) very similar BE leakage.

[mozz]

I didn't know they had leakage more than silicon. "In general the leakage current at a given temperature will be higher for a lower voltage rated Schottky diode".

So if you wanted more leakage get the lowest rated voltage ones, probably cheaper too.

[Rob Strand]

In this day an age

Real germaniums have the wiggley wire inside:




Whereas small signal Schottky's look more like silicon diodes,




If you have bought a "1N34" and it looks like a Schottky it's going to be a Shottky, simple as that.

[PRR]

Real germaniums have the wiggley wire inside:

I'm sure you know, but for the youth: early Silicon 1N914 were also cat-whisker "S"-wire. Even into the 1970s when I was sorting reject 914-types to find 127V Zeners.

[Rob Strand]

I'm sure you know, but for the youth: early Silicon 1N914 were also cat-whisker "S"-wire. Even into the 1970s when I was sorting reject 914-types to find 127V Zeners.

Not off hand.   I do remember a whole lot of different packages which were common in the day (maybe 60's spilling over to early 70's) but long gone now.   The very early diodes had a vast array of packages - not recognizable by modern day eyes.

[Phend]

Off topic, tried this with an AA battery, got no reading, used a 9v and get 0.20ma with old RS 1n34 and 0.22 ma with D9J Russian diode.
Short video.

[Rob Strand]

Off topic, tried this with an AA battery, got no reading, used a 9v and get 0.20ma with old RS 1n34 and 0.22 ma with D9J Russian diode.

Those results don't look right.

Try a 1M resistor first to make sure your test set-up is correct.
You should end-up with about 1.5V/1M = 1.5uA = 0.0015mA.

The 1uA to 5uA zone sounds right for the germanium.
And < 1uA for a *small signal* Schottky diode would be about right.

For me the thing that is brain-dead about all those type of videos is you can see which diode is germanium and which diode is Schottky.  You don't need a meter!
Another similar video

Results:
Real 1N34A:   1.7uA to 1.8uA
Schottky:      < 0.1uA
Diodes around my desk:
Green band ge:   3.5uA, 4.0uA   ; likely 1N34A
Schottky BAT46   0.2uA

There's a hidden trap measuring leakage:
If leakage doubles every 10 degC  then that means someone performing the leakage measurement at 25C will get double the leakage value of someone measuring the same diode at 15degC.    So casual measuring at home can introduce a significant variation in the reported leakage measurement.

What you really need to do is record the temperature, then it is possible to normalize the leakage to a temperature like 25 degC.

[Phend]

Those results don't look right.

Thanks Rob, that was the reason I posted those results, wasn't sure what the readings should be.
Remeasuring, (my DMM seems to be a magnitude, ie decimal is in the wrong place, off on the mA setting)
Using the uA setting, I get 1.9 uA with 1M resistor and 2.2 uA with D9J diode.
(I haven't been measuring current in my DIY effect life, learning something today, thanks)
The Russian diode is the same physical size, call it large, as the 1n34.
Question, are there Ge diodes the same size as a 1N914 for instance.
Small, probably not for construction purposes.

[Rob Strand]

Question, are there Ge diodes the same size as a 1N914 for instance, small, probably not for construction purposes

I don't remember any, definitely not like the clear 1N914 package.

Some *silicon* diodes are painted, like the Schottky pic I posted earlier, but the package is very small like the 1N914.  Some are even smaller.

The older painted germaniums are larger than the 1N914's.  They are the same size as the germaniums with the wiggley wire, except someone has painted over it.
https://www.eonet.ne.jp/~k-matsu/germanium_diodes.html

Using the uA setting, I get 1.9 uA with 1M resistor and 2.2 uA with D9J diode.
The Russian diode is the same physical size, call it large, as the 1n34.

All info so far pointing to real units!
EDIT: clarified something

[mozz]

My cheap MK328 component tester gives me a few ua's leakage when testing old germaniums. I think the highest i saw was 6ua or 7ua.

[Chillums]

Real germaniums have the wiggley wire inside:

I'm sure you know, but for the youth: early Silicon 1N914 were also cat-whisker "S"-wire. Even into the 1970s when I was sorting reject 914-types to find 127V Zeners.

Yup, I've got a few of the 1n916 with the s-whisker and a few with colored bands like you usually see on resistors. 

[Rob Strand]

Yup, I've got a few of the 1n916 with the s-whisker and a few with colored bands like you usually see on resistors. 

Somehow your post jogged my memory.  I have a "brand new" black OA202, date code F411 or F414.  Exactly like this one,
https://www.fabian.com.mt/en/products/webshop/9333/--rectifier---------diode-160ma-150v-oa202.htm

I can't see inside though.   I know it is silicon.

I threw out all my funky old parts years ago but I started gathering new ones after I started playing guitar.  I'm pretty sure someone gave me the OA202 with a small bundle of other parts and thought those parts were from around 1985.