Noise Gate - Buffered Loop Problem/Decimator

Started by lars-musik, December 12, 2019, 05:13:06 AM

Previous topic - Next topic

lars-musik

Dear Forum,

I am working on a small 1590a decimatorish Noise Gate. There a number of different schematics out there and basically two different versions exit: a standard gate and one with a loop for the signal-to-be-gated. I successfully build one standard version and now I am trying to utilise the same schematic but with an added buffered send/return loop.  The schematic I am following for this is this one:

https://guitar-gear.ru/forum/index.php?app=core&module=attach&section=attach&attach_rel_module=post&attach_id=31065



The section with the red overlay is the newly added loop section (the blue line is the connection of the first circuit I built without loop).


The circuit works somehow, but not perfectly. In the standard version, the only pot (Threshold) is adjusted and as the name suggests it simply adjusts the point on a volume curve where the gate sets in. Now with the new PCB and Send&Return connected I expected the same behaviour but somehow on near the end of the threshold pot travel, the volume goes up and to the very end down again. Just like some kind of out-of-phase problems.

The schematic seems to have a redundant non-inverting voltage follower (DA1.1) in the send path (why would that be?). But all in in all these two added op-amps (compared to the non-loop version)  are non-inverting there shouldn't arise any phasing problems, right?
So: Can you spot anything that would explain the strange threshold behaviour?

Cheers, Lars

Kipper4

Might just be the pot you chose Lars.
Damaged track at the end of the travel? Did you try replacing it with another of the same value?
Rich
Ma throats as dry as an overcooked kipper.


Smoke me a Kipper. I'll be back for breakfast.

Grey Paper.
http://www.aronnelson.com/DIYFiles/up/

lars-musik

Hi Rich,

yes, unfortunately I already tried another pot - different make & model. I just plugged the suspicious buffer from the SMD board and am on my to testing it. Let's see. The board was SMD populated by JLCPCB and I just went through all my pick & place lists again, verified solder pads and resistor codes under a binocular microsope. Looks all good. Maybe I'll just try another board in the new year and see if it behaves differently.

Kipper4

And If you put a 1k stopper resistance between thresh and gnd?
i suspect it wont make much difference but might be worth a punt.
Ma throats as dry as an overcooked kipper.


Smoke me a Kipper. I'll be back for breakfast.

Grey Paper.
http://www.aronnelson.com/DIYFiles/up/

lars-musik

Now I took out DA1.1, R4 and C7 (because this one misses in another schematic I found oin the net) and now it works. Really I pain, because I had these boards already fabricated and I cannot figure out why that is so, but I am mostly happy anyway.

Kipper4

Ma throats as dry as an overcooked kipper.


Smoke me a Kipper. I'll be back for breakfast.

Grey Paper.
http://www.aronnelson.com/DIYFiles/up/

lars-musik

Gosh, seems like I didn't nail after all. I built another one with the same problem. And somehow I overlooked that this issue was already adressed some years ago.

Quote from: Kennt82 on August 31, 2017, 05:47:29 PM
"max threshold is at the half of the pot range"   That's my experience using a PN2222A.   I guess I should try a MPS2222A (with a few spares).  What are the differences I should be paying attention to on the spec sheets?

So maybe the specs of T3 have anything to do with that. A big problem for me is, I really do not understand that circuit and all it intracasies.
I will socket that transistor and go through my stack of 2n2222a, PN2222a and MPS2222a. And some 2n5088 and 3904 for good measure.

If there's understanding why and what values of T3 might cause this behaviour: please share!

duck_arse

from the datasheet:

QuoteThe voltage at V- (pin 5) is four diode drops below
ground, which, for the 2181, is approximately -2.85 V.
Since this pin connects to a (high impedance) current
supply, not a voltage supply, bypassing at pin 5 is not
normally necessary.

what's your voltage at that pin? [way out of my paygrade.] as far as transistor type/number/prefix, it just sits there and provides a static current determined by the diodes referencing the base, which probably any transistor will do the same. way different types might show variations, but a 2222 is a 2222 by any name smelling as sweet.
" I will say no more "

gryf_folker

What's up people, I hope you're all well

Here I leave some rar of Decimator projects that I found on the web

The second file is a modification that I'm doing to make it as similar to the original as I want to add the link in and link out, I have those link ports on another board so it was just a matter of adding some pads, I added them but I don't know how to use Kicad 7 much and I would like someone to review it and see what is wrong.  :icon_eek:

Greetings!

https://mega.nz/file/aE5FQa7I#KHZKmM_3ywIK-R-JLsJTB7tgQ1JqSHQxz8zpheBk-rI

Mod

https://mega.nz/file/7VZ31T5Q#RIv2tGtplKfZokZEZb9rJ9Y-5YL8iLT5oX1AQL3OTZA