gain/atten control on passive EQ

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

warpie

Well-known member
Joined
Feb 7, 2009
Messages
1,566
ok, so I'm breadboarding a basic LC Resonant circuit with a pot for gain/attenuation.

The problem is that I can't gat a linear behaviour of the pot although the pot is linear.
In more detail, the pot will increase/decrease much more radically when is rotating to the ends of it's travel.
If the pot is somewher near the middle the gain/att is smoother. Of course this gives me better control in the middle position but I was wondering if this is normal and if there're other eq designs that have the same behaviour.

I have tried other techniques like gyraf's and manley's eqs where there's a switch for cut/boost and the pot is 0 to 15/20 dbs. In this case an antilog pot seems to give the best results but again the behaviour is not so smooth.

I believe that Neve and SSL used to have some similar behaviour in their pan pots but is it normal for EQs and if not are there any workarounds?

well, I hope my description is not very confusing and it makes some sense  :)

thanks
Michael

 
 
can you post a schematic of exactly the circuit you are making? it might make it easier to troubleshoot...

You could try strapping resistors across the pot to modify it's "law" - depending on how the pot is connected - it doesn't work for every application. There's lots of info on the web/forum about this - a quick google search for "modify pot law with resistor" (no quotes) brought up:

http://sound.westhost.com/pots.htm#chg-law

http://web.archive.org/web/20050307111237/www.wavefront.mcmail.com/pot.htm

http://www.geofex.com/article_folders/potsecrets/potscret.htm

BUT that might not solve your problem.. show us a schem and we'll be able to figure it out, hopefully!
 
thanks for the links haima  :)

I have the schematic only on paper at the moment. I'll try to post it if I don't give up on the idea  :)
 
OK, I haven't given up yet  ;D

here's a rough schematic of what I'm talking about. The top section is the one where I have my gain/atten control and it's linear.
In fact, if you notice R6, R7 are used to create a 'fake' center tapped pot.

The problem is that I don't get a smooth resolution. I think is very similar (if not the same) problem that PRR is mentioning here
http://www.groupdiy.com/index.php?topic=42735.msg531605#msg531605
 

Attachments

  • simple_lc_ex2.gif
    simple_lc_ex2.gif
    7.2 KB · Views: 72
> I have the schematic only on paper at the moment.
> OK, I haven't given up yet   here's a rough schematic


Um...... what a tangled web.

Go back to hand-drawing, skip the details, emphasize the architecture.
 

Attachments

  • simple_lc_ex2-hand.gif
    simple_lc_ex2-hand.gif
    3.8 KB · Views: 42
warpie said:
The problem is that I can't gat a linear behaviour of the pot although the pot is linear.
In more detail, the pot will increase/decrease much more radically when is rotating to the ends of it's travel.
This is a very common problem. Some manufacturers offer pots with a specific law (the law is the resistance vs. percentage of rotation); it's generally called W or S. I have never seen these available off-the-shelf; one has to order them in several thousands quantity. The most common workaround is using rotatry switches.
 
thank you gentlemen!

PRR> Could you please elaborate a bit more on the sketch? Being an idiot I'm having hard time to understand a couple of things  ::)
i) the component that goes to gnd is a resistor?
ii) I can skip the 2k2s but I will need a 'middle point' for my LC filter (?)

I'm looking forward to find some time to go back to breadboard and try your suggestion.

abbey road d enfer> It is very hard to find W-law pots indeed. Can I get the same result with something like this?
pots-f10.gif


If everything fails I'll use a rotary switch. The only reason I prefer a pot is for better resolution (not stepped dbs). Not great for stereo matching though...
 
warpie said:
i) the component that goes to gnd is a resistor?

Pretty sure that's the LC (inductor & capacitor in series).

warpie said:
ii) I can skip the 2k2s but I will need a 'middle point' for my LC filter (?)

Not sure I understand you - wouldn't the wiper of the pot form the "middle point"?
 
Pretty sure that's the LC (inductor & capacitor in series).
thanks! of course... ::)

Not sure I understand you - wouldn't the wiper of the pot form the "middle point"?
in the schematic I posted before (very messy indeed) the input of LC is connected to  some kind of virtual earth point (consisting by the two 2K2 resistors) and the output goes to the wiper (i.e the input of the LC is not connected to the ground).  

With rotary I think it will look something like this...

(unfortunately I don't have a scanner at the moment)
 

Attachments

  • simple_lc_rotary.gif
    simple_lc_rotary.gif
    4.4 KB · Views: 26
I wasn't aware of this schematic but it seems very similar topology. I think I'll give it a try too.
thanks haima! all your links have been very useful so far  :)
 
I believe it's the midrange or "presence" control from the 1073 & 1084 - so have a look at the full schematics for these to see how it's implemented and also the schematic for the "amp" section shown above.

The current AMS/Neve company has the full schematics available on their site i think - might take some googling to find it. There's plenty of info on these out there.
 
Unfortunately, you want exactly the opposite!
Unfortunately, you are right...  :)

...which leads us to the rotary solution again. I might skip the whole idea and try to use a rotary instead.
Shame because I really got into the 'pot solution' mood...  :)
 
Back
Top