tube screamer diodes

[saxtim]

I just built my second tubescreamer and I used NOS germanium diodes from small bear.  I did this because it mentioned at GGG that the original used 1N34A diodes.  Anyway, with the ones from small bear in, I got 0 distortion, even with max drive.  I swapped them for 1N914s and it worked fine (distortion wise).  Any reason why the Germaniums didn't work?  Did I miss something?

tim

[jimbob]

I did the same thing. Got mine from smallbear, but i did get some distortion..I wanted omre though.. I used one germ. and 1 1n914 and got a pleasant sound-At the end i liked it better w the 2 1n914's--I understand that using the gernm's cut out distortion and add smoothness--But thats not my thing--I want a little more umph.

[george]

you would probably need 3 Ge diodes in series back to back to get the equivalent resistance to 1 Si diode.

This is because the Vf (minimum voltage across the diode for it to conduct) of a Ge is approx 3x that of an Si.

Mind you I have a TS-9 and I like it just fine with the Si diodes ....

hth

[RobB]

I recently finished assembling two ts808 copies on GGG PCBs.  One for myself and the other for a friend.  Leaving one with the standard 1N914s and JRC4558D I’ve been substituting diodes and opamps into the other and making comparisons.  
The first thing I tried was substituting two Ge diodes for each 1N914.  Each 1N60P has a Vf of about 0.3V.  As a series pair they have a Vf of about 0.59V compared to a Vf of 0.73 (measured just a moment ago) for a single 1N914.  
When I A/Bed these the unit with the Ge’s clipped up sooner as expected.  It produced more distortion than the one with 1N914s.
If you used only one Ge diode per side then it should have clipped up much sooner.
Saxtim, have you measured the Vf of the Ge diodes you used?

[Nasse]

Did you remeber to do some preventive things to not blow your Ge´s with heat when soldering? Squeeze the leads between component and solder joint with some flat-nose pliers or something while you do quick soldering.

Or is there something like unconnected or cold joint in your circuit?

[Joep]

Check the diodes with a DMM or something. I think one of the Ge diodes is shorted. This explains why you don't have distortion.

Remember that added more diodes in series will decrease the level of distortion, since the signal is clipped at a higher level.

regards,

Joep

[Mike Nichting]

Try this: put 2 1N4001's on one side of the clipping section and 1 on the other. It gives great tone and plenty of distortion. try a switch for the 1N4001's and the 914's and see which ones you like best. Or you can try this: run 3 germaniums in series with a 914 at the end. so you would have 4 diodes in a row. That yields some good tone also.

 Try an LED on one side too. The sky is the limit with this circuit. it seems like you can mod about every part of the Tube screamer circuit with good results.

Best of luck~!!!!

Mike N.

[javacody]

I was happy simply swapping a 914 for a 1n4001.

[Gringo]

The 1n4001 are my favorite non germanium diodes for clipping too.
Maybe because of their lower Vf ?? The ones i have measure under 0.6...

[saxtim]

So I should actually be getting more distortion from the Ge diodes?

I just checked both of them - between .3-.4 Vf on my DMM.  I used sockets for them when I originally put them in, so no heat damage.  

I swapped them back in and 0 distortion again.  I tried one Ge and on Si, some distortion, bth Si - much more distortion.

So to reiterate - I should get more distortion from the Ge diodes right?  If so, any idea what's going on here?

tim

[RobB]

Yes, more distortion from Ge diodes.  The lower the Vf the sooner your signal will clip.  

Perhaps the tinned legs on your NOS Ge diodes have oxidised so they are not making good contact with the socket.  Try sanding or scraping the grey stuff (lead tin alloy) untill the ends of the legs are shiny again.

[jimbob]

I just took out a 914 and replaced it w a 1n4001-- i like it better too..Btw-- for a few, i had no dist either--same problm as you..It was a bad 914..haha

[javacody]

I'm glad to see you guys share my same good taste in diode clipping.   :lol:

[Mark Hammer]

In one of the TS clones I built, I stuck a 100k variable resistor in series with one of the 1N914 diodes, as a "warp" control, following Jack Orman's terrific paper.  I recommend it, and also recommend reading the warp paper.  Rather than dickering with diode types endlessly, pick something in the ballpark and add a variable control.  You'll get a wider palette of tones and responsiveness.

[Boofhead]

Like George said, you need more diodes if you want to use Ge's.  At *least* two in each direction, more like three in each direction and you can add more to taste.

The usual one silicon in each direction, or two silicons in each direction sounds better to me on the TS-9.

[Mike Nichting]

so Mark,
do I wire one end of the pot to the leg of the diode that I pull out of the PCB and connect the wiper to the open hole??

Thanks
Mike N.

[Ansil]

have you tried just putting two germaniums inline between the two opamps like the hm2  i did this on my clone and have never been happier..

[saxtim]

Boofhead:

Like George said, you need more diodes if you want to use Ge's. At *least* two in each direction, more like three in each direction and you can add more to taste.

RobB:

Yes, more distortion from Ge diodes. The lower the Vf the sooner your signal will clip.

Sorry guys, unless I'm reading what you're saying wrong, I'm getting conflicting opinions here as to whether you get more or less distortion from Ge diodes!  

I'm not really interested in modding/experimenting too much with diodes (I'm reasonably happy with the 914's), I'm just curious as to why I'm getting 0 distortion with Ge diodes?  If the original used Ge diodes it obviously created distortion - which means I should also get distortion with them no?

Oh, on a side note I retired the Ge diodes after sanding the leads lightly with wet/dry to make sure I got good electrical connection.  I check continuity too with my DMM from the diodes to what they connect to - all was ok.

I'm actually sorta wondering if this means maybe I have another problem in the circuit somewhere? (though all my voltages on the transistors and IC's are fine and it works as expected w/silicons)

Tim

[RobB]

I don't know about the original using Ge's.  I have not tried a single Ge in each direction in my clone but have tried this with a Boss SD1 which is a simmilar cct.  It produced considerably more distortion but lacked output level.

If you want your drive control to work the same with Ge's then you will need 2-3 Ge's in each direction to match the clipping threshold of silicon diodes.  

I ended up with the following in mine.  1N914 on one side and two 1N60s and a 100k "warp" control on the other side.  Still sounds like a TS but with a bit more tonal flexability.

[saxtim]

I don't know about the original using Ge's.

I really don't know - I got the info from that parts lists at GGG where it had various diodes options, one was the n34a and it had 'orignal' in brackets next to it.

If you want your drive control to work the same with Ge's then you will need 2-3 Ge's in each direction to match the clipping threshold of silicon diodes.

I was hoping the drive control would actually work different. Ie, I'd like to find a pair of diodes that give me more distortion at the top end of the drive control, more than what I'm getting from single n914's in each direction.  As I said I tried single Ge's in each direction which produced no distortion whats so ever.  

I decided to revist this thead because today, out of interest, I tried two 5mm red Leds in there, each measuring a Vf of around 1.8V.  I got more signal with this, but still no distortion.

I don't understand whats happening here.  I'd thought the relationship between forward voltage and clipping (distortion) would be linear no?  At the moment it seems I'm only getting distortion from the silicon diodes and I don't understand why.

To clarify - I've only ever used one diode in each direction, regardless of the type.  I've seen some variation in output level (not a whole lot), but as mentioned above I can't figure out why only one type gives me noticable distortion.[/quote]