THAT4301 compressor

[L´Andratté]

Hello Ungifted,
I appreciate and follow your work.

I was reading somewhere, that with RMS it is inherent, that attack and release time are interdependent.
So when you implement those (like done in the app note 115) you´re leaving true rms behind?
There´s no problem in that, as you´re just adding a possibility, that could also be switched out.

I´m just trying to understand the nature of these things...

 

[Nescafe]

Hi ungifted,

my question about RMS is like what L'Andratté said. Your work progress is really great for this community, really appreciate it.

Thank You.

 

[ungifted]

Thank you for your support and questions!
We feed signal to RMS detector, at the pin 5 is circuit (or just timing capacitor) which determines how fast or slow its reaction (or it's adaptive in auto), and then we feed control signal to attack/release and control them separately...
As I can understand in the case when RMS detector is in Auto attack and release are simply non adjustable. In Auto, RMS timing is set according to the Non-Linear Capacitor circuit (DN114).

A non-linear capacitor ... acting as a large timing capacitor for slow
moving signals and as a smaller timing capacitor for fast moving signals.

When we turn Attack/Release on we can adjust attack and release independently. DN115 says:

Two timing modes are available -- auto and manual. In auto mode, the output of
VR7 is connected to R16 via SW3. Attack and release rates are linked, and depend
on the design of the non-linear capacitor circuit (see Design Note DN114 for more
design information on the non-linear capacitor circuit).
In manual mode, variable, independent adjustment of the attack and release
dynamics are achieved by setting SW3 to steer the side-chain signal through the
discrete transistor circuitry on the right side of the accompanying schematic.

Maybe someone can help, and describe if there is any conflict and we need turn Auto on and at the same time turn off variable attack/release circuit. I just follow DN115.

 

[elskardio]

Made rough tests. TL072 based schema is working (attached). Here is a pic:



So I see no problems with adding this feature to Pico boards if you have one.
BTW can anybody comment on the 500ms for release? Is it usable?

Next is DN115 schema...

Hi Ungifted,

I'm currently testing the TL072 based schematic. The Attack works properly but the Release don't... Turning the pot doesn't affect the signal at all !
Did you experienced any issue with the release?

I also have the DN115 schematic on breadboard. It works but with longer release time, the signal gets weak (low gain)... Still trying to figure that out  ???

Thanks

 

[ungifted]

elskardio
Thank you for your note. Tested again, yes you're right, I'm afraid we have an issue with the release with TL072 schema...
Going to test DN115 version...

 

[elskardio]

Please keep me posted 
Thanks

 

[druu]

How did the test go ungifted?

 

[ungifted]

I've build a prototype. It doesn't work as must when I use LM393 (it sounds like generator not compressor)
But I was lucky with TL072 instead of LM393 (schema from DN115).
Please, folks, take a look at pics. This seems to be true?

Release. Min and max. Attack is in the center.


Attack. Min and max.


PCB attached if someone will decide to make tests. (the non-linear capacitor circuit in not tested!)
PS: I've tried LM393 with pin 4 at ground and with Vcc- at pin 4 perhaps I don't know how to cook it...

 

[RuudNL]

What is the reason the positive part behaves differently than the negative part?

 

[ungifted]

Rough measurements
Android tone generator -> device -> PC.
I think everything will be ok in right conditions, but to be sure, it will be great to see if someone can test the DN115 variable attack/release schema.
PS: I hope I can make more precise tests soon, after my apartment renovation.
Now everything upside down

 

[druu]

Any progress on this?

 

[druu]

Bump!

 

[shot]

Hey, I was just wondering...
When looking at the v1.5 schematic, would that be possible to connect output of the VCA (pin 13) to the input of the RMS detector (probably before or after R11), while lifting the path from input opamp (lifting out C6) and thave this compressor work in feedback mode?

I know FCS comp has it, and I'm just wondering how it's done...

 

[3nity]

Guys i dont know what you think.
I purchased 2 pcbs from this guy: www.diy-tubes.com
http://diy-tubes.com/that4301-compressor-pcb.html

im about to start this week see how it goes..

 

[Tekay]

LM393, isn't that a dual comparators   absolutely not eqvivalent to a dual OP!

 

[ungifted]

3nity
That guy is me. Please let me know how it goes. Thank you!

Tekay
Yes LM393 is not dual opamp, but I've seen both versions (THAT DN with LM393, and redrawn design with TL072).

I'll try to find some time to return to process...

 

[Nescafe]

Guys i dont know what you think.
I purchased 2 pcbs from this guy: www.diy-tubes.com
http://diy-tubes.com/that4301-compressor-pcb.html

im about to start this week see how it goes..

3nity
That guy is me. Please let me know how it goes. Thank you!

;D ;D ;D

That's a Really Nice web store ungifted!!!!!

 

[druu]

I think there's some info in DN113 re: LM393 and it's function in the attack/release circuit?

 

[shot]

Has someone tried attack/release board layout posted in THAT4301-AR.PDF attached a page earlier?
Is this a valid layout?

 

[druu]

Bump!