Clipping stage question

mlabbee · August 31, 2004, 09:30:06 PM

I'm building a TS clone and have been playing around with the clipping diode options to find a sound I like. One that I tried is two opposed 1N914s and a red led in series with one of the 1N914's. The LED seems to round out the sound nicely, but I notice that the volume jumps up when the red LED is in. I'd like to be able to switch the LED in and out of the circuit to get the two sounds, but the volume shift is kind of a pain. Any idea what causes this and any possible cures?

Paul Marossy · August 31, 2004, 10:11:56 PM

I believe that the difference is that the 1N914s and red LEDs will clip the signal at different voltages. I believe that it takes roughly twice the voltage for the LED to clip the signal as it does for a silicon diode. Not sure there is really a fix for that. Maybe two LEDs in series?

cd · August 31, 2004, 10:36:02 PM

First of all, if you haven't already, read Technology of the Tubescreamer over at GEO:

http://www.geofex.com/Article_Folders/TStech/tsxfram.htm

Second, two LEDs in series would make the problem WORSE - it would raise the clipping threshold to double that of a single LED.

What you need to do is switch in another gain setting resistor, like on the old first series Fuldrives. Use a DPDT switch, one half to switch the LED, the other to switch the 4k7/.047uf string to something like 10k/.022uf. Just remember, if you double one value, halve the other (and so on) if you want to keep the same frequency response.

Paul Marossy · August 31, 2004, 11:41:23 PM

Oh. :oops:
That's always kind of confused me. It's kinda counterituitive that adding another diode would increase the volume. It seems like it should do the opposite...

aron · September 01, 2004, 04:04:41 AM

Yes, it's in my simple mods page, but yes. More diodes increase the clipping threshold and level.

Paul Marossy · September 01, 2004, 10:18:57 AM

OK, I understand the clipping part, but I don't understand why it would increase the output level. It seems to me that your signal is only so big, and then you whack it with a clipping diode, so the amplitude would decrease, not increase.

This is the part I don't understand. Can anyone tell me why this is so?

Mike Burgundy · September 01, 2004, 10:32:49 AM

Here's the trick.
Firts off, there are two back-to-back clipping diodes. These are regular Si, so lets assume a threshold of 0.7V. Below this, a diode does not conduct! It needs a voltage of 0.7V to start conducting!
Any signal below 0.7V at the output will not put the diodes in conduction, and so will not be fed back. Anything above that will be fed back. Feedback above a certain value results in the waveform being clipped *at* that certain value. Get that one?
Now, add a series LED to BOTH diodes. Let's assume a 1.2V threshold for the LEDs. Now the feedback threshold is 0.7+1.2=1.9V: anything below this will not be clipped. This means there's MORE signal coming through before clipping occurs: the pedal gets LOUDER, for the same amount of clipping, or clips a lot less at the same volume.
If you only add one LED, this applies to only one half of the waveform, but the idea is the same.
hih

cd · September 01, 2004, 10:37:23 AM

Quote from: Paul MarossyOK, I understand the clipping part, but I don't understand why it would increase the output level. It seems to me that your signal is only so big, and then you whack it with a clipping diode, so the amplitude would decrease, not increase.

This is the part I don't understand. Can anyone tell me why this is so?

You're halfway right. If you put two diodes in series, their combined clipping threshold INCREASES. If you take a single 1N4148 with a .6V threshold and put two in series, the threshold becomes 1.2V. With one diode, any signal above the .6V threshold will be clipped, so if you have two 1N4148s opposed (like on a RAT) the top and bottom of the wave will be clipped, and your maximum peak/peak output will be 1.2V. With two diodes in series, you've increased in the clipping threshold so now signals above .6V will not be clipped, only signals above 1.2V will be clipped. So your maximum peak/peak output has increasd to 2.4V. Bigger output = louder sound.

RDV · September 01, 2004, 10:54:42 AM

How bout this. Put the LED on a footswitch and have a footswitchable boost. :twisted:

RDV

cd · September 01, 2004, 11:04:15 AM

Quote from: RDVHow bout this. Put the LED on a footswitch and have a footswitchable boost.

Doesn't answer the problem about the volume boost.

Paul Marossy · September 01, 2004, 11:50:26 AM

OK, I think I follow that. It still seems counterintuitive to me, though.

mlabbee · September 01, 2004, 12:58:04 PM

Thanks for all the input - I didn't really have a clear idea of how the diode threshold affected the signal and how the values added up. Now it makes much more sense. I'll play around with some of the approaches suggested and see what works.

Thanks again!

RDV · September 01, 2004, 01:32:30 PM

Quote from: cd
Quote from: RDVHow bout this. Put the LED on a footswitch and have a footswitchable boost.

Doesn't answer the problem about the volume boost.

Oh, but it does. A footswitchable boost for a TS-808 style pedal would b cool IMO. You say "volume boost" like it's a bad thing. Are you a singer? :wink:

RDV

cd · September 01, 2004, 01:38:20 PM

Quote from: RDV
Quote from: cd
Quote from: RDVHow bout this. Put the LED on a footswitch and have a footswitchable boost.

Doesn't answer the problem about the volume boost.
Oh, but it does. A footswitchable boost for a TS-808 style pedal would b cool IMO. You say "volume boost" like it's a bad thing. Are you a singer? :wink:

RDV

Did you bother to read the first post completely? The poster did NOT want the volume boost.

News:

Clipping stage question