Taming hiss in high-gain boost circuit?

[ch1naski]

I'm doing my own take on the classic rangemaster circuit, using a very high gain ( hfe~ 550 @room temp) germanium transistor. It's like Tony Iommi in a box.

But the thing hisses like a snake through a bullhorn. Is there a simple way to tame the hiss, without removing the useable high frequencies?

I'm not educated, electronics-wise, so forgive the noob question.

[brett]

Hi
there are a couple of potential sources of noise.
I'm no expert, but in any Ge transistor circuit, we'd immediately suspect the Ge substrate.
How did you measure the hFE? 550 is quite a high number for a true hFE. If you measured it with a multimeter, you've measured (hFE+leakage). Leakage is current that flows from the collector to the emitter without being controlled by the base. I suspect that a lot of your 'hFE' is leakage. Leakage is closely associated with noise, because it reflects manufacturing defects in the device that give rise to noise.
There a circuit for measuring 'true' hFE and leakage at geofex.com.
On the other hand, if that 550 is a true hFE, then the transistor would be a modern Ge device, and these are known for low noise. But...the gain of 550 amplifies the modest noise to very high levels. Chances are that any Ge device with high gain will be noisey in a Rangemaster. To a large degree, it's just a fact of of the semiconductor physics of Ge combined with high gain.
A low-noise alternative would be to use a Jfet or Si BJT booster (with gain of 2 to 10) ahead of a Rangemaster with gain of 100 to 200.
cheers

[ch1naski]

Hi
there are a couple of potential sources of noise.
I'm no expert, but in any Ge transistor circuit, we'd immediately suspect the Ge substrate.
How did you measure the hFE? 550 is quite a high number for a true hFE. If you measured it with a multimeter, you've measured (hFE+leakage). Leakage is current that flows from the collector to the emitter without being controlled by the base. I suspect that a lot of your 'hFE' is leakage. Leakage is closely associated with noise, because it reflects manufacturing defects in the device that give rise to noise.
There a circuit for measuring 'true' hFE and leakage at geofex.com.
On the other hand, if that 550 is a true hFE, then the transistor would be a modern Ge device, and these are known for low noise. But...the gain of 550 amplifies the modest noise to very high levels. Chances are that any Ge device with high gain will be noisey in a Rangemaster. To a large degree, it's just a fact of of the semiconductor physics of Ge combined with high gain.
A low-noise alternative would be to use a Jfet or Si BJT booster (with gain of 2 to 10) ahead of a Rangemaster with gain of 100 to 200.
cheers

Yes, it's a modern germanium, ac128. I did use a dmm for the measurement. Also noticed that the hfe shot up drastically if I kept my fingers on it and warmed it up. (Definitely going to need a bias pot, )

Guess I should go ahead and build the test circuit since I plan on building a few germanium-based circuits.

I really do hope to use a high gain germanium in this circuit.... I've read about Darlington pairs, I wonder if that could be done with this?

[R.G.]

It is unlikely that the real gain is that high. Almost all DMMS are set up to measure silicon devices, and cannot account for the leakage of a germanium. The leakage looks like gain to the meter, but isn't.

Brett is right, it may well be that you have to sort through devices to find a quiet one. Germanium darlingtons are chancy. They're really only good for low gain devices. The problem is that the second device amplifies not only signal but also the leakage of the first one, so it can get out of hand quickly with real high gain devices.

The rangemaster circuit is not necessarily going to give you a lot more signal/voltage gain with a high *current* gain germanium device in it, as the circuit has other things that limit voltage gain, which is what we listen to.

[ch1naski]

I actually had a decent b175a in there, and it acted like a normal rangemaster circuit. When I added this ac128, the bass frequencies are prominent, and the only real treble boost is the transistor background hiss. it's almost a fuzz unto itself now, kind of a happy mistake. I like what it's done to the circuit, I'm wondering if I can somehow turn this into something useable. You know, make a wrong circuit, right.

This circuit is producing a HUGE sound, but flabby, over the top bass.
And no treble.


I'm sure devi ever could do something with it, she manages to sell just about any kind of fuzz sound.

And by the way, thank you guys for answering my question. I love this place.

[ch1naski]

Update:
Swapped out that 128 for a b176 with a purported hfe of around 350 (still way over the top for a rangemaster circuit) and got some treble back, still noisy as hell, but retaining the fuzziness. But now it sounds like a bit of ground hum has creeped in.

[chromesphere]

Hey ch1naski,
Dont know if you've seen these, but thought they might be helpful to you.  
First ones a quick 'noise' test of 2 rangemasters with an AC125 vs GT108V.  
Second video is a sound comparison between the two.

Noise check:

Sound Comparison:

Hope it helps & good luck!
Paul

[ch1naski]

Thank you, Mr. Sphere. Going to watch that now.

[ch1naski]

The noisy ac125 sounds better, to my ears. Is there no simple tone control circuit that can attenuate the hiss, while leaving the shimmering high frequencies useful to a guitar sound intact?

[R.G.]

Would that it were so.

You can try cutting off everything above about 8kHz or so; most wound guitar pickups have essentially no output above 6kHz-7kHz. That at least would cut off hiss that the guitar isn't making.

But there are only ways to make it more palatable. There is no good way to separate hiss and guitar signals of overlapping frequencies in real time.

[LucifersTrip]

the hiss is most likely caused by leakage, and killing the hiss with some type of tone control will kill some high end (as RG noted, I believe with "overlapping frequencies")

It is unlikely that you have the transistor biased correctly at ~ 7V with those super high hfe ones...you should be able to get a great trebly tone if you bias a low leakage, normal hfe transistor properly.   Did you tweak to around 7V with any of the transistors you were using?

http://www.geofex.com/Article_Folders/Rangemaster/atboost.pdf

[ch1naski]

actually, I used that article to bias it with the first transistor I tried, . But not the second.  Smh.  I guess a potentiometer is in order if I'm going to be trying different transistors in the circuit.

[ch1naski]

So....boxing it up with a bias pot, and a resistor in series so it will only go as low as 2k2. That should enable me to try a few different transistors....until I get off my butt and put together that leakage tester. I've got a bunch of old germanium transistors and am hoping to put at least one one two to good use.

One thing I've noticed.....I suck at boxing it all up.

[ch1naski]

Just wanted to report success with the rangemaster. Thank you for your help, guys. And RG's article on rangemaster? Printed that sucker onto paper. Indispensable info.  Quite a finicky circuit, for how simple it is.

[pinkjimiphoton]

hey bro, there's always the cap fudges. sometimes they help, sometimes they don't.

experiment putting small caps between b and c, or c and ground sometimes.

a bigger cap can be used right on the volume control, go from input or wiper to ground. i've used caps up to .022 for noise reduction and still not ruined the guitar tone.

sometimes it's the only game in town.

[ch1naski]

I'll try that. It's still a little hiss prone, even after putting a low leakage transistor in there....

[kaycee]

You can just put in a relevent polarity silicon transistor that will have no leakage and it will work for this circuit. Or you can look up the schematic for various other silicon based treble boosters such as the Brian May one, the Vox and the Hornby Skewes and try those, they aren't that different. Check that you have good joints with your other parts connections, sometime a bad socket joint can cause an amount of noise.

[ch1naski]

You can just put in a relevent polarity silicon transistor that will have no leakage and it will work for this circuit. Or you can look up the schematic for various other silicon based treble boosters such as the Brian May one, the Vox and the Hornby Skewes and try those, they aren't that different. Check that you have good joints with your other parts connections, sometime a bad socket joint can cause an amount of noise.

Using those single inline sockets for transistors...they don't work too good with done old transistors, legs are too thin...i might have to build a new board with room for a normal transistor socket.