ampage
Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

 

Re: DIY Parametric EQ at GEO


 :
10/7/1999 1:00 PM
R.G.
Re: DIY Parametric EQ at GEO
quote:
"you mention Resonant Fr as 1/6.28*SQRT(L*C)  
SQRT, is that square root? Is the series  
resistor ignored?"
 
The "SQRT" is indeed "square root". That particular  
nomenclature is a leftover from FORTRAN (ugh!).  
In an LC circuit, resistance changes the "Q" or  
resonance/bandwidth, but does not shift the center  
frequency, so the resistor makes no difference and  
is ignored.  
 
quote:
"Later you mention Center Fr as 1/6.28*L*C  
I suppose to an EE, this is childs play,  
but I'm stumped. "
 
Well, actually, how about I just think faster than  
I type? Should be the same formula each place, and  
should be 1/2*pi*SQRT(L*C).  
 
For a .22uF and 56mH, you get  
Fr = 1/(2*3.14*SQRT(0.056*.22E-6))  
or 1.43E+3, 1430Hz if I didn't drop the decimal point.  
I think you meant 1.2kHz, not unreasonably far from the  
computed value.  
 
10/7/1999 4:41 PM
Ed Rembold

Thankyou R.G.  
That clears it up for me.  
Yes, I made a typo too- 1.2Khz is what I  
meant to type. Truthfully, I did not think  
you made a formula boo-boo, I just didn't  
quite get it, but, I got it now.  
Ed Rembold
 
10/7/1999 5:59 PM
R.G.

Never assume that I haven't made an error - I make lots of them 8-)  
 
I like to think that with the aid of modern computer hardware and software to amplify my efforts, I can make and correct more errors per day than I ever could by hand!  
 
 
 
10/8/1999 12:56 PM
R.G.
Update to DIY Parametric EQ at GEO
I've just uploaded an update to the parametric EQ article at GEO. It contains some info on Twin T and Bridged T filters, bootstrapping them for higher Q, and adding variable Q and frequency to these versatile filters.
 
10/8/1999 11:40 PM
Ed Rembold

Dear R.G.  
Jack Ormans notch filter article at his  
site gave a good intro , But , your newest  
update is the bomb!  
Its embarrassing to always be needing  
Math help- Oh well-  
In the example for the Bootstrapped Twin T  
w/Variable Q-  
R is given as 100K  
C is given as .0027  
My Question, given the example values, is-  
2C = .0054?  
0.5R = 50K?  
Please help, Ed Rembold
 
10/9/1999 1:18 AM
R.G.

quote:
"Its embarrassing to always be needing Math help-"
 
Not to me. I've asked for help with the math on lots of stuff. Don't worry about that. If you keep asking questions, I may learn where I'm not being clear enough - and the examples looks like one place for sure.  
 
Anyway, you have it right. In those schemos, "R" is taken to be some specified value of resistance, like 100K. All the others that are specified as multiples are the multiple times the value of R, as you did with 0.5R being 50K. Likewise, 2C is two times 0.0027, or 0.0054. The simplest thing to do with resistors and caps for 1/2 R or 2C is to put two equal values together in parallel to avoid the standard values not being quite exact if you're trying for a close match.  
 
By the way, I did another update tonight. I got some questions about a graphic EQ as opposed to parametric or the notch filters, so I put a seven band graphic EQ circuit schematic in the same article; I also put a jump table at the top of the article so you can go directly to certain topics.  
 
The graphic uses the same capacitor/gyrator technique as in the parametric, but in this case uses transistors for the voltage followers as GFR mentioned, to illustrate that what's needed is just a voltage follower.  
 
I didn't do any great design work there - the circuit is substantially the same as the one at the web page noted in the schematic. All I did was adapt it to a single 0V supply from a bipolar supply setup. Note that the circuit only shows the essential stuff - I didn't draw in the resistor-divider Vbias generator that is referred to by "Vb" nor stuff like power supply decoupling caps, bypass switching, input buffering, etc. To really do this thing up right, you'd use a dual opamp. The first opamp would buffer the input to prevent treble losses, and the second would function as shown in the schematic.  
 
In my mind I group every effect into a "wrapper" of standard stuff to make it work, and the actual goodies inside. The wrapper is stuff like the box, the jacks, the bypass switching, the power supply, power decoupling and conditioning, and so on. The goodies are just the stuff that does the effect.
 
10/10/1999 4:13 AM
hyschip

R.G. I love ya man ! this is one thing I have wanted to build for a long long time Thanks...hys chip
 

<<First Page<PrevPage 2 of 2