Simple DiodeBridge Compressor

[wlinart]

This is why groupdiy is so amazing! More than just a paint by numbers set, where 2 great minds of the forum contribute to something new and exciting! Thanks Richiyobs and Heikki!

 

[richiyobs]

Thanks... Is a pleasure to be here

 

[richiyobs]

here is the meter driver, always thanks to super Heikki
just change values according to your meter sensitivity.

 

[richiyobs]

the signal path....
in the meanwhile it can be builded, now you get all parts, from meter to SC and SumAmp.
for the last one, please let me know if is better to leave as it is, or going for mods
in the SC part can be ease added the Auto set with a jumper and switch, for the moment... even for this let me know your suggestions for this implementation especially for longer release time
best
richi

 

[richiyobs]

40Hz, fast att and rel... Combined with 1uF cap

 

[richiyobs]

this is the idea about MS matrix....
really don t know if i m missing something or to complex to achieve
i d like to realize something simple to implement
thanks in advance for your suggestions
cheers

 

[richiyobs]

meter added to main board

 

[richiyobs]

quasi floating balanced out
i m going to test

 

[richiyobs]

has someone already tried this design?
thanks for your attention

 

[abbey road d enfer]

this is the idea about MS matrix....
really don t know if i m missing something or to complex to achieve
i d like to realize something simple to implement

Those 1uF caps should really be 100uF.
And the trimmers won't do anything, except increasing noise.

 

[abbey road d enfer]

has someone already tried this design?

Of course! It was standard on thousands of mixers in the 80's and 90's.
They have been ditched in favor of chips like THAT1646 or SSM2142, which are way better, and don't need tuning.

 

[richiyobs]

Great! Thank you so much for your quick reply and explanation

 

[dogears]

As for the diodes I´d suggest 1N4153. They are less noisy than 1N4148 in a diode bridge compressor.

Would it be worth it to look at using transistors as diodes in those positions?

 

[richiyobs]

Would it be worth it to look at using transistors as diodes in those positions?

https://toshiba.semicon-storage.com...ors are,tenth the required collector current.Something like this?

 

[richiyobs]

I m not so knoleged but replacing diodes means having a different Vca, and so a different machine... Isn t it?

 

[abbey road d enfer]

As for the diodes I´d suggest 1N4153. They are less noisy than 1N4148 in a diode bridge compressor.

What do you reckon in their parameters make them less noisy?

 

[abbey road d enfer]

I m not so knoleged but replacing diodes means having a different Vca, and so a different machine... Isn t it?

The intrinsic dynamic resistance of a PN junction is 26.10-3/Id, so all diodes should be more or less equal. The differences are essentially in the spreading resistance, which is the sum of all the parasitic resistances, and capacitance. This resistance is usually quite low. For a1N4148, the effect of it appears at about 100mA, where the dynamic resistance is about 0.25 ohm., very far away from the typical operation area in a diode compressor.

 

[richiyobs]

This forum is Great! each day i m learning something new...
Thank you all for your support and knowledge.
i d like to better understand this concept but to hard for a beginner like me...
so wich could be the proper dynamic resistance for this kind of application?
best
richi

 

[dogears]

I was just thinking maybe easier to find matching parts.

 

[richiyobs]

The intrinsic dynamic resistance of a PN junction is 26.10-3/Id, so all diodes should be more or less equal. The differences are essentially in the spreading resistance, which is the sum of all the parasitic resistances, and capacitance. This resistance is usually quite low. For a1N4148, the effect of it appears at about 100mA, where the dynamic resistance is about 0.25 ohm., very far away from the typical operation area in a diode compressor.

hi Abbey road,
the sum of all the parasitic resistances, and capacitance is refferred to this:
since we put both ac and dc to diodes, where dc part is the rectified one from sidechain. varying dc we change the impedance of diodes, so attenuation....
ac is well attenuated before going to vca, to avoid distortion...
instead to avoid noise we run balanced, right?
my console have unbal send and ret , and didn t noticed noise, but still have to perform noise and thd measurment.
on old devices , for example u273 b, noticed that it s using both silicon and germanium, is it a way to lower resistance slope and so distortion?
thanks in advance for your support
best
richi