AMZ FX ANALOG ALCHEMY GEOFX MEF ROG MEF DIY WIKI

In-line diodes = noise gate + crossover distortion

Started by WGTP, February 28, 2007, 09:52:54 AM

Previous topic - Next topic
Print
Go Down Pages1
User actions

WGTP

February 28, 2007, 09:52:54 AM
I was at the PV forum and the Bravo which is a tube amp has in-line diodes in the pre-amp section.  I asked about it and the reponse was that the primary function was noise and hiss reduction, with a by-product being the cross-over distortion.  Sort of a crude noise gate.  Something to try anyway.
Stomping Out Sparks & Flames

puretube

#1
February 28, 2007, 11:51:14 AM
right...

electrictabs

#2
February 28, 2007, 12:05:34 PM Last Edit: February 28, 2007, 12:11:17 PM by electrictabs
in order to reduce crossover distortion and extreme gating you should try a resistor across diodes. Try to find a value that will work good for both the effect and by-product. It should probably be in the Megaohm region. . PV does this to recent amp series.

DDD

#3
February 28, 2007, 12:38:57 PM
Due to the logarythmic sensitivity of our ears we can easily detect even the negligible additional hiss. That's why shunting the diodes with the multi-MegaOhm resistor will give the frustrating effect regretfully. At the same time it won't reduce audible crossover distortion too much.
Try Shottky diodes and the capacitor from the "output" terminal of the diode combination to ground.
Too old to rock'n'roll, too young to die

electrictabs

#5
February 28, 2007, 06:08:53 PM
Great links Ton, thanks


IMO the inline diode circuit is just a simple noise gate circuit having its limitations.
Using schottky diodes with a cap to ground might give less crossover but it will alter the frequency response of the whole circuit making it darker – even when the signal is pretty strong – but mostly when the signal is weak . That is really cool if we want to put a gate there but also shave some of the annoying high content of a distortion . But if we want to preserve ALMOST the same freq response to an existing circuit I think it is better to leave the cap out and test which diodes fit. 1N4148 shunted by a 2M2+  resistor gives a nice result, at least to my ears. The most important thing is to find where to place them in the signal path to allow both fast and dynamic playing.

Print
Go Up Pages1
User actions