Simple diode limiting circuits

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Autophase

Well-known member
Joined
Oct 15, 2009
Messages
478
Location
UK - Manchester
Hi guys,
I have recently stumbled up on diode soft clippers / limiters, I have done some reading and i understand the basic principals, however im struggling to find some schematics with specific component values.
To begin with I would just like to make a passive diode limiter that will run of studio line level (1.737v average for +4dbu)
I like the idead that its a simple principal and a simple circuit that can be tweaked, using other diodes such as LED's.
I was hoping someone here would have some experience with this and could point me to some schematics to get me started.

thanks
 
Autophase said:
Hi guys,
I have recently stumbled up on diode soft clippers / limiters, I have done some reading and i understand the basic principals, however im struggling to find some schematics with specific component values.
To begin with I would just like to make a passive diode limiter that will run of studio line level (1.737v average for +4dbu)
I like the idead that its a simple principal and a simple circuit that can be tweaked, using other diodes such as LED's.
I was hoping someone here would have some experience with this and could point me to some schematics to get me started.

thanks

Not sure what a diode "soft" clipper is. Diode clippers are more associated with guitar "fuzz" effects, not very soft clipping.

I guess you could scale down the signal voltage seen by the clamp diodes (need two, one for each polarity) such that the hard clipping is occurring at a higher voltage.

1.7V peak is roughly 3 diode drops, so I would speculate a simple inverting opamp gain stage with gain of -10dB (say 100k input R, 33k feedback R). Put the two diodes in opposite polarity parallel to the 33k R.  Follow this with a another inverting stage using a 33k input R and 100k feedback R to restore nominal unity gain.

#1 tweak the ratio of 100k/33k to get desired clipping threshold.
#2 scale impedance of both R using same ratio to make limiting harder or softer (higher impedance should limit sooner/softer). 

Note: you may still end up with a glorified fuzz tone... Perhaps useful as a special effect.

JR
 
What is your goal, a distortion effect or a relatively clean limiter?

I have designed limiters using variable gain elements with the goal of low distortion.

I actually have a patent that uses a diode clipper (just at bass frequency) for a specific application (to prevent saturation in output transformers). Diode clippers are not low distortion.

JR

 
In order to get even harmonics you should go for asymetrical clipping, (different voltages on both sides) this could be done with 2 options, different diodes (Ge, Si, LEDs, Zenner) or different quantity of diodes... or diodes only for one side. Different types of diodes will change clipping characteristics. Figure 4, 5 and 7 of your link are nice options...

JS
 
Thanks you both for your replaies, I'm definately going to breadboard a few of those circuits next week, ill let you know my findings.

forgive my ignorance here, but in the 2 LED design of figure 4 (the schem is above the actual paragraph)
the author mentions this is an active design, so I assume the power supply I need is dependant upon the OP-amp in the circuit?

I assume because there are 2 diodes the circuit is clipping the full wave?


Cheers
 
For a simple to simulate circuit, it doesn't get much simpler than this one:

Go to: http://audio.kubarth.com/rundfunk/index.cgi

And search for the following entry:

SIEMENS U274
2.1 MBytes, GIF image data, version 89a, 2956 x 1494
Added: Sun Jan 1 00:13:51 2006

I have this one simulating rather nicely, and it does a great job of compressing. It might help you understand yet another one of the multiple diode limiting topologies out there.
 
I've been playing with "soft clipping" for electric guitar, not so much to simulate tube distortion but to see what other, "new" effect can be done. This circuit may not be "simple" but it demonstrates what I'm trying to do.

Below 0.7V (or whatever you claim the forward "on" voltage is) the circuit has a gain of 1, with R1 being a 40k resistor from the signal input to the opemp's negative input, and R3 being a 40k from the opamp's output back to the negative input. When the signal goes over 0.7V, D1 and/or D2 conduct, and the feedback path is R3 in parallel with R4 at 32k, giving 17.7k. Likewise when the signal exceeds 2.1V, D3/D4 and R5 kick in, for even lower resistance in the feedback path.

I've tried to scale the resistors to give a "smooth" transition with mixed success.  I've done this with the guitar input going into an opamp buffer, and that driving a substantially lower value for R1, so it gives enough gain to make most of the diodes conduct at full volume. Full volume gives a loud, highly distorted sound. Lower input volume gives actual lower output volume, but it still has some distortion at a level where any other distortion circuit would sound clean. Looking now, I'm thinking a substantially lower value for R3 (maybe 2k or 4k) would give less distortion at lower level, but still a smoother transition to increasing distortion with increasing volume.

Another possible mod is using only diodes in one direction, and making a parallel circuit with diodes in the other direction so the positive and negative parts of the waveform can be clipped independently at different levels.
 

Attachments

  • diode_soft_clipper.jpg
    diode_soft_clipper.jpg
    128.8 KB · Views: 136
etheory said:
For a simple to simulate circuit, it doesn't get much simpler than this one:
SIEMENS U274

I have this one simulating rather nicely, and it does a great job of compressing. It might help you understand yet another one of the multiple diode limiting topologies out there.
This is not the same as a soft clipper.
It's a bona fide compressor, using the variation of dynamic resistance of diodes as the gain-control element. In order to work well, the audio voltage must be much smaller than the junction voltage. Because of the needed attenuation, S/N ratio is not very good (neither is THD).
 
abbey road d enfer said:
This is not the same as a soft clipper.

I know, it is a limiter or compressor depending on your definition.
The title of the thread is diode limiting circuits, which the U274 definitely is. It's a compressor/limiter.
Though later in the thread the OP mentions soft clipping, which is obviously a different thing, which others have touched upon. I was just providing another option for a limiter based on a diode element.
 

Latest posts

Back
Top