LEd Clipp and Germanium Clip Volume change

[served]

Hi.

I am having a problem. And I am afraid that I cant figure it out by my self.
I would like that LED clipping and germanium diode clipping has the same volume. So that when I switch from on to another volume will not change.

Is there a good way to make it work. The output Vol rises significally and I can not tolerate it.

I hope you can help me.

As for schematic. Lets say I have a TubeScreamer or Rat or similar circuit. If I gen any ideas I will apply it to all of these pedals I have moded in my effect collection.

[Brymus]

You can try putting the Ge diodes in series until the forward voltage equals that of a single LED.
Works for different diode types.

[paulyy]

Iv had this problem berfore. If you do what Brymus says. It will work. I had to do this to one of my pedals and works fine.

[served]

Oh Really? And it will not affect the sound? Well in this case its Awesome solution! I would not have figured it out.

[anchovie]

Adding extra diodes in series will raise the clipping threshold, meaning that you get less distortion than with one pair. You could use a switch with an extra pole that adds a resistor in series with the volume pot when you switch to LED.

[Joe Hart]

Adding extra diodes in series will raise the clipping threshold, meaning that you get less distortion than with one pair. You could use a switch with an extra pole that adds a resistor in series with the volume pot when you switch to LED.

+1 on this idea. Raising the clipping threshold will give you a totally different sound. They will then actually sound quite similar! If you "adjust" the volume with a resistor, then you will still have the same distortion characteristics for both pairs.
-Joe Hart

[Mark Hammer]

Unfortunately, while the effective forward voltage of germanium diodes can be arranged to be equivalent to LEDs, what people associate as the tone of Ge diodes is actually the difference between a higher and lower forward voltage.  In other words, if you want the "tone" of Ge diodes, then you have to accept the forward voltage of Ge diodes.

If a person wishes to be able to flick back and forth between the two "tones" and have negligible change in volume, then you have to figure out a means of compensating for the resulting difference in volume, while preserving the difference in forward voltage.  Typically, this involves inserting what is called a "gain recovery stage" after any clipping diodes.  It should be called a "level-recovery gain stage" since it applies gain for the purpose of bringing the level back up.  No actuall "gain" is lost or "recovered" at any point, only signal amplitude.

This can be an op-amp or a single transistor stage.  The poster child for this is the last transistor stage of the venerable Big Muff Pi.  Since the passive tone control loses a lot of signal via selective bleeding to ground, the transistor stage brings the level back up by applying gain to the signal.

If one uses a level-recovery gain stage like that, there are two choices available.  One is to set it for a larger amount of gain for the Ge diodes, and lesser gain for the LEDs, such that the result is similar volume level.  The other is to give the stage a fixed gain, and then subtract signal level from the output for the LED setting, generally by use of attenuating resistors .

[served]

But Mark. I like Germanium diodes Tone more than LED-s, but I wish to use LEDs also.
That way I think Adding a resistor to the Volume control I can remove Volume jump/drop.
Or should I Add a resistor some where else? In the op amp section? I have 1pair free swich leg so that I could add a resistor if I want.

[Brymus]

I think what you would do is add the resistor before your volume pot, and then have it so that when you switch to Ge diodes the resistor is jumpered,removing it from the circuit so to speak.
Just use a pot wired as a resistor(actually as a divider would work too) to find what value you need then when the volume change isnt noticable measure and use the closest standard value resistor instead.
Or you could leave it in as a Lead boost adjustment.
And make your switch a footswitch (I did this with my BMP build using diodes to achieve the boost)
I just thought of 3 other ways to do this as well.But I wont complicate/confuse it any more than I have .

EDIT - I said after the volume pot instead of before,I dont think it matters in series (a quick check with my DMM shows it doesnt)
But I changed it to before so as not to confuse you when compared to Mark's answer.

Another way would be to use a seperate volume pot for each clipping set and not have a main one,
then just switch between the two volume pots/clipping circuits so either one could be boosted /cut for lead purposes.

[Mark Hammer]

But Mark. I like Germanium diodes Tone more than LED-s, but I wish to use LEDs also.

I still don't know what specific circuit we are talking about, but in a generic sense IF you have sufficient volume for your liking with the Ge diodes, then all you need to do to drop the volume, using LEDs, down to a comparable level, is to stick a single fixed resistor in series with the input to the volume pot.

If you had a 100k volume pot, and placed a 150k fixed resistor in series with the input, then the resistor+pot would behave as a 250k pot that never makes it past 40% of maximum volume.  Stick a 220k fixed resistor in ther instead of 150k and its like a 320k pot that never gets above 31% of maximum volume (or rather is never attenuated any LESS than 69%).  

How to switch it?  Use a DPDT switch, whether toggle or stomp.  One set of contacts connects the free end of the one set or the other to ground or wherever they are supposed to be connected.  The other set of contacts is used to either bypass, or unbypass the added series resistor.  Since Ge is quieter, you bypass in the Ge position, and  unbypass the added resistor (whatever the value) in the LED position.

[served]

Thanks! I will add a resistor and will let you know! Many thanks for the answers!
First I am going to do it with a Tubescreamer.

[WGTP]

Since GE's have a threshold of around .3v it will take around 6 or 7 of them to equal an LED at around 1.8v.  Interesting experiment to hear the difference in tone.   

[teemuk]

If you "equalise" the signal so that each device (LED or Ge diode) clips the signal at about the same relative level the Ge diode will clip a bit softer than a LED but overall there won't be much audible difference between the two. The main reason why folks get so exited about this is because the different forward voltages provide different amounts of overdrive when the level of input signal driving the clipping stage stays constant.

If you equalise that level (attenuate the signal before it gets clipped by Ge diode so that it begins to clip at about the same threshold as a LED) you pretty much loose all the magic of using different diode types.

Here's an illustration: [Top graph] See how, because of higher forward voltage, the LEDs clip the signal much less than the Ge diodes when both clipping circuits are driven by a signal having the same amplitude. [Middle graph] Now the signal driving the Ge diode clipping stage is attenuated and after clipping some gain is introduced to boost the signal to the same level as that going through LED clipping circuit. What this shows is that the output waveform is about the same regardless of clipping diode type. [Bottom graph] Now the clipping circuits are driven with two times higher input signal amplitude than in the previous examples. The Ge diode clipping is a bit softer than that of LED diode (you can see how the amount of overdrive is a bit lesser and the clipped signal doesn't have as flat top) but all in all, your ears can't likely even tell that there's such a small difference.

[served]

I added a resistor. Now I cant see much differentce between Led and Diode. They sound pretty much the same. LED-s doesnt have this hard clipp sound, maybe I should apply the resistor after vol control. Then I will not change the resistance between the signal and ground.

OK so there is no difference. OK, thats a shame. I was hoping to hear a miracle

[WGTP]

Nice illustration.  Do Silicon diodes look the same or are they sharper? The next thing to try (if you haven't) is using different diodes in pairs.  For example LED/GE  LED/SI  GE/SI  2GE/3GE  SI/2SI, etc.  This produces asymmetrical distortion which includes more even order harmonics and will some different tones.   

[teemuk]

Yep, silicon diodes are a bit sharper and Schottky diodes a bit softer. But, it's an illustration of the principle so don't take it as a Gospel. The graphs are from a SPICE sim so reality with different devices might be a bit different. Also, even the "slope" isn't a too big of an issue because you can control it's steepnes (or softness) simply by introducing some series resistance to the diodes.

For example, this illustrates how the clipping LEDs alone can't really capture the softness of a pair of Schottky diodes (which, also have a noticeably lower forward voltage than LEDs) but with the introduction of a 220-ohm series resistor the clipping characteristics become much softer and closer to that of the Schottky diodes. Infact, a higher amount of series resistance (e.g. 520 ohms) would make the LEDs to have more softer clipping characteristics than what the Schottky diodes have.


So, you can pretty much play around with the amount of series resistance  and with the amplitude of the drive signal and achieve various types of clipping characteristics. All this while using the very same pair of diodes.

[served]

Wow. This is extremely interesting and you are being really helpful. I had never seen such graphs before.

[frank_p]

I love graphs, thanks Teemuk. Self explanatory.

[WGTP]

Yes, the graphs are great.  There was a thread within the last few months about clipping threshold and using resistance and R.G. indicated that it produced a result that was different than what the ear was used to hearing, IIRC.  I may have totally misunderstood, but I'll see if I can find it.  I had been advocating using caps between the diodes and to boost the bass by clip it less which would also make it less distorted than the highs.   

[teemuk]

Here it is:
http://www.diystompboxes.com/smfforum/index.php?topic=82764.msg687158

I think he means that it sounds a bit "odd" that the signal actually gets louder while it also gets cleaner, in oppose to getting louder while distorting more and more (which is what happens in the typical setup where you turn up the gain control).