Greer Green Giant

[motterpaul]

Read all the comments for both designs Green Giant Silicon and here).  

Is the Nastified the working version now? I believe I read Rocket says it sounds awesome, Iggy says he built it, but he wants more play in the bias trimmer.

Also - no one has posted a demo?

So, the idea is to get the new guy (me) to play with this and see what happens? I'm Mikey and I'll try anything? (old Life Cereal reference there). Oh, why not?

[Beaker]

I believe that all the versions work.

[rocket88]

I'm pretty sure I removed the ones that had issues, the ones with the tone control, I actually have to mess with my last idea to see if I can get it to work right. Everything that's normal does work. I haven't had a chance to make a demo, plus depending on diode choice you'll get major differences in the clipping and tone. I would build a normal one and socket the diodes and just go to town trying different ones till you find something you really like.

[motterpaul]

Funny - you guys are so literal, by "working" I mean "serving to permit or facilitate continued work:
a working model; a working majority." - it's one of the definitions in the dictionary of "working." (I know you know it, just clarifying that was how I used it.)

I started building it already - but talk about stupid mistakes that one repeats. I have been on this thing lately of cutting my vero with the wrong orientation, with rows going down instead of across. Usually, I notice right when I start soldering and I am I like "fuck. fuck. fuck... did it again" But I am starting a new cocktail of drugs today to treat a condition which has been making me pretty foggy for a few months now. The meds just make it worse. So, today I didn't even notice it until I had several components in. Talk about stupid habits... this one is killing me.

Anyway - didn't take long to get back to where I left off when I saw I had effed up.

[bogey]

I have to agree that this is a great first germanium build(small board/parts count-easy to breadboard).

I also went with Beaker's suggestion of upping the 680r for a bit more "cleanliness"-good call btw.


So to echo what has been said,  for such a small simple circuit this is excellent.

EVERYBODY SHOULD BUILD ONE

Oh forgot to say I built the first one in the thread-no tone pot/trimmer/diode switch

[Silver Blues]

I got around to putting this little monster on the breadboard, and I've settled on the 2SA49 + 1N34A combo as the one I'll be building. In the very end it was a toss-up between four diode pairs (1N69A, D9I, SFD106 and 1N34A) then I shaved it to two (1N34A vs. SFD106) and the 1N34A won out. All four were "almost there" and I'd recommend people try them in case they like them; I tried everything I had, and what I found was

- the 1N69A were nice and aggressive, but lacking a little smoothness. Early breakup.
- The D9I were like the 69A but smoother.
- 1N60 were really trebly. I didn't really like these (probably good for guitar though, nice and sharp tone).
- 1N67 sound okay but they don't break up much in this circuit unless you play really hard.
- SFD106 were really deep but lacked some of the edge of the 1N34A (basically a tamer 1N34A).
- 1N34A were nice and big-sounding with a good amount of hair overall.

I also tried a few asymmetrical pairs but those weren't as good as matched diodes. I think the best one was probably 1N69A and SFD106 but still not as good as 2 1N34A. I also had a few single diodes that I didn't bother to test because I had already established that the asym pairs weren't as good. (Maybe a mistake? It's still on the breadboard, I could always try them anyway...)

As far as transistors go, I can say for sure that AC125 are no good in this circuit. Really bad mis-biased fuzzy tone unless you cut the collector resistor to 10K (and I really mean it, I also tried 12K and 15K and anything higher than 10K causes the fuzz to return), and then it's still not really that great. All the others I tried (GT2306 and 2307, 1T308V, 2SA49 and 2SA53) were pretty similar, I found the 2306 significantly better than the 2307 and the 2SA49 a little better than the -53. In order of preference I'd say 2SA49 > GT2306 > 2SA53 > 1T308V > GT2307. I also tried the basic 2N3906, and it was ok; decent choice if nothing else.

Here's another layout for PNP with an inverter if anyone's interested. This is made to look like exactly what I'll be building. Obviously you'll have to watch the pinout depending on what particular transistor you're using. (EDIT Layout fixed).


Nice little circuit

[motterpaul]

I also tried doing the AC125s and had not much luck. They are very leaky.

Rocket - I thought I read you saying you got a batch of the AC125UZ IV, while what I had was VI (higher hFE overall), so you inspired me to try a batch of the AC125 uz V  - so far I have just tested a few, but they are lower hFE (most below 100) but the leakage is still pretty high (from .2 to .4 ma)- so I don't know how they will work out in different circuits. I like the hFE range, but the leakage has me a little concerned.

I should be getting some NPN Ge's soon, so I can try out the Ge Giant on my own as intended.

Thanks for the Power Inverter version, Silver. I just might try that one as well.

[Silver Blues]

No problem man, but mine wasn't the first power inverter version in this thread, I just felt I'd take a crack at it to see if I could better fit the layout to my personal parts.

[Travis]

I haven't had time to look very carefully at your layout but at a glance I don't think it's going to work. The transistor symbol is backwards and you have shorted out the output cap. I think you probably intended to put a cut to the left of the top leg of the output cap. I'll look again in a bit and see if I find any other problems

[Silver Blues]

Oh yeah you're right about the output cap. I'll fix it.

EDIT fixed cut and symbol.

[rocket88]

Correct. I have more then one batch of AC125uZ IV, probably have gotten somewhere between 4-6 batches of 150. Mine are not super leaky like the ones you have. I test all of mine with my DCA75 at room temp. I'm not sure if you're using a DCA or doing the calculations, but I still don't get why you're are mostly very leaky and high hfe. What I can say is, do you touch them and test them right away? If you're handling them even for a second you're going to mess up your results since the hear from your body will raise the gain and leakage. They very temperature sensitive, which is part of the reason that Ge transistors were abandoned in favor of Si.

In any event, yea I posted a layout with an inverter earlier. Never gave the PNP's a try, so glad you did. I'm wondering why they're acting more fuzzy, since they should be more stable then the NPN. Maybe it could be due to the higher hfe? Idk, but since PNP's are so much cheaper and readily available it would be nice to crack it and get it sounding good.

[Silver Blues]

It probably is a product of the higher hFE. Like I said they sound okay with a 10K collector resistor instead of the 47K, feel free to try that and see if you get anything you like out of it. I didn't.

[motterpaul]

Yes, I do now have a DCA75 and I do not touch the bodies at all, or the legs much (as little as possible). I have only tested a few of the V batch, but I have gotten lower hFEs but fairly high leakage, with the VI batch I got higher HFE and about the same leakage as V.

I think that is just the difference in batches, though I will know more when I have tested more Vs.

[Coxy]

Looking for some tips on biasing this please. I built it with the 5K trimpot and D9D diodes (tried lots of D9 diodes in it with much the same results) to get any real sort of gain with the clean pot turned all the way CCW I need to set the trimpot all the way CW. Should I try a different transistor in it ( just got some MP38A's in) and do I adjust the 5K trimpot for best volume or gain?
Thanks in advance!

[rocket88]

Depends on the transistor you're using, you may need a larger trimmer. Keep in mind that the total output and gain(distortion in this case) is determined by the amount you have the diodes in the circuit. With mine, I use an MP38A. When the diodes are not in the circuit I get a massive maybe around 15-20db relatively clean boost with volume full CW, and with the diodes fully in the circuit I'm just above unitiy with volume full CW. Also, if you use Ge diodes you'll get less output then with Si, LED, or BAT diodes.

[Coxy]

Ok thanks I will have a further play about with it and try an MP38A in it. I also had to use a 2M resistor in it instead of the 2M2 would that make much difference ? It still sounds great just its the first time i've used a trimer on anything and wasn't really sure what to aim for. Think I will build another and try lots of different transistors and diodes before boxing.
Many thanks to everyone involved in making this site, I discovered it six months ago and am now totally hooked. I now have five boxed pedals on my board and am loving them.
Cheers

[motterpaul]

I finally got some NPN Ge  MP38a so I want to build the latest most official Germanium Giant - can you please verify if that best build is posted and identify it? Is it the on the first page - or else a new one - please just make sure I have the right build.

I assume it is this one:



Thanks! Looking forward to it.

[rocket88]

That shouldn't make a huge difference, using a 2M instead of 2.2M. If you're really worried about it put a 200k resistor parallel the the 2M to make a 2.2M.

This is a circuit that is super tweakable. What most of us have found and done is use a Ge transistor with Ge diodes. Basically what you should get is a fairly clean Ge booster with the diode blend full CW, so they're not in the circuit. As you start to turn the diode blend CCW, putting them into the circuit you're going to start to hear the signal clip. The less the diodes are in the circuit the less clipping you'll hear, so basically you'll hear the clean signal with a little bit of distortion in the background. As you add the diodes the less clean signal you hear and more distorted signal you'll get.

You've got the right idea. Get the transistor biased, then go to town with different diodes till you get something you like.

[rocket88]

Hi paul. There is no best build for this effect, but you want the regular layout without the tone control. There is still development on creating a layout with a proper working and effective tone control.

[Silver Blues]

I think you mean in series. Putting a 2M and 200K in parallel will net 182K.