Hello all!
After years and years of just being around, i wanted to collaborate with the forum again.
Bridged T filters are widely used as notch filters and frequency boosters (when used in the feedback path). i wanted a circuit that could do both by mixing the T network in and out of the feedback. At max position, the filter is fully inside the feedback, at mid position it is half at the output and half on the feedback so the notch cancels, and at min, it is after the output, producing a notch.
I don't know if someone has done this before, but i find it very practical as you only need 1 opamp and 4 components to make each boost/cut filter.
it is very extreme..you can get as much boost and cut for each frequency just by tweaking the values (lower resistor to gnd gives more Q, but alters the frequency)
i finally settled on +-20dB and defined 90/600/4K Hz as the three main bands. You could easily add more filter stages but 3 bands are ok for me at least..
this is the schem:
NOTE: The POTS ARE 100K LINEAR
(https://static.wixstatic.com/media/b88141_665b593f406640669e4d06b1bc2a9481~mv2.gif?dn=eq.gif)
if anyone wnats to make a layout for it and/or test it, please do!
Looks mighty interesting! Say, which pot value would you suggest?
Oh i forgot to label them!!
they are 100kB (edited the post)
Interesting! And "welcome back". :icon_biggrin:
Should I assume that replacing the 1000pf+470pf pair with 1500pf won't change anything in a terrible way?
Thx! This will be breadboarded a.s.a.p. :)
Quote from: Mark Hammer on April 26, 2017, 01:13:58 PM
Interesting! And "welcome back". :icon_biggrin:
Should I assume that replacing the 1000pf+470pf pair with 1500pf won't change anything in a terrible way?
yes, it will destroy the nyquist stability coefficient and blow up :P naaa kiddin´
of course, if you have 1.5n caps..
wait a minute i have a new version coming up!!!!
i´ll post it soon!
there we go:
it uses 1 more cap per filter, BUT you can add dual pots to make each band parametric!!
(https://static.wixstatic.com/media/b88141_92bb5288be2e4f25880e1414d09ccc52~mv2.gif?dn=eq2.gif)
Nice one Daniel. Great too see you back too.
I'll have to breadboard at least one stage at some point.
Is there a need for a buffer up front?
Can't wait to see the next incarnation. I love your stuff.
Rich
Quote from: Kipper4 on April 26, 2017, 02:14:55 PM
Nice one Daniel. Great too see you back too.
I'll have to breadboard at least one stage at some point.
Is there a need for a buffer up front?
Can't wait to see the next incarnation. I love your stuff.
Rich
Thanks Kipper!
Try one stage with the dual pot for parametric adjustment :P
it does not need a buffer in front for 1 stage, omit the gain pot and the last gain stage. but it will need a buffer at the output..
Any risk of oscillation at max? Not suggesting there will be.
Quote from: blackieNYC on April 26, 2017, 02:35:19 PM
Any risk of oscillation at max? Not suggesting there will be.
Judging by the Q's , not really..changing the values may get lower Qs to avoid it if it happens. It is probably less noisy than gyrators
> I don't know if someone has done this before
I can not recall seeing quite this circuit.
> only need 1 opamp and 4 components to make each boost/cut filter.
Indeed interesting, and thrifty.
I havent breadboarded it yet but I imagine it has great potential as a tone shaper Pre a distortion.
Also for those projects where you end up with a spare op amp. BANG add some Eq..
nice and interesting. will be monitoring this and potentially build this. maybe somebody can
draw up a pcb hehehehe ;D
Nice. 20db is super powerful! Anyway to reduce the amount of cut and boost?
Is it drawn right cuz I threw it together on the bb. Pot one way lots of gain. Pot tother way no gain. No signal.
Could be my breadboard ingredients skills or lack of.
Only had a few minutes before work.
Should the other end of pot go to output too?
Sorry bout this.
Cool, nice one daniel... 8)
Quote from: Kipper4 on April 27, 2017, 03:53:03 AM
Is it drawn right cuz I threw it together on the bb. Pot one way lots of gain. Pot tother way no gain. No signal.
Could be my breadboard ingredients skills or lack of.
Only had a few minutes before work.
Should the other end of pot go to output too?
Sorry bout this.
If you mean that the gain pot at zero makes no output, then it is right.. the gain is a simple pre volume control. With 20db of boost you need to lower the gain with hot signals or it will cliip
I admit I had no gain pot. It was the clipping that foxed me ill re breadboard tonight . Thank
Yay
I breadboarded the mid eq from post #1.
Inc input gain and output make up.
It works a treat. Great amplitude range. Really effective.
C100k pots (it's what I had in my breadboard stash is all.)
TL074.
Thanks Daniel.
I'll try swapping out some caps and maybe even try a 10k pot to Ground from the cap junction.
See what it does.
yep this
(http://i.imgur.com/Mpt0Izc.png)
I will definitely try this one! Is it noisy like gyrator based op amp EQs?
C taper?
it will work but it will not be 0dB at the center..
True that about the C taper. It was just to try it.
No it's not noisy like a gyrator. The Q pot I added does make it noisy because of the added q in the highs though.
I tried a 10k Q pot it didn't work through all the range. Plus the 5k pot needs a stopper R.
The high gain is a bonus but probably to much for most applications imo.
A very good low parts low pots Eq.
Sterling job Daniel.
Quote from: Kipper4 on April 27, 2017, 04:35:46 PM
True that about the C taper. It was just to try it.
No it's not noisy like a gyrator. The Q pot I added does make it noisy because of the added q in the highs though.
I tried a 10k Q pot it didn't work through all the range. Plus the 5k pot needs a stopper R.
The high gain is a bonus but probably to much for most applications imo.
A very good low parts low pots Eq.
Sterling job Daniel.
nice shem!
you need to add the dc blocking cap at the input
changing the frequency via the ground resistor is a PITA..it changes the Q too much and has limited range..i´m sure there´s a way to do it but i tried and failed..
Yer I'm not sure the extra 5k pot is worth it. Maybe if one where to make a 6x stage maybe one extra pot might make sense.
Oops forgot the input cap. I used 100nf.
A useful addition to the thread would be the calcs for freqauncy.
That's a really neat, elegant little circuit. I like it a lot.
Well done, and nice work Daniel.
Tom
Quote from: Kipper4 on April 27, 2017, 05:29:01 PM
Yer I'm not sure the extra 5k pot is worth it. Maybe if one where to make a 6x stage maybe one extra pot might make sense.
Oops forgot the input cap. I used 100nf.
A useful addition to the thread would be the calcs for freqauncy.
i used this formula to calculate the center freqs:
(http://sim.okawa-denshi.jp/images/CRTwinTf0.gif)
R1 is the big resistor, and R3 the smaller.
where C2=C3 and R2=0 (use the bottom formula for bridged T)
The Q is given by the relation between R3 and R1 i don´t have a formula for that but it is sort of R1/R3 dB´s
So, an updated Purple Peaked, sort of
Sent from my XT1032 using Tapatalk
Quote from: jonathanfdean on April 28, 2017, 01:31:40 PM
So, an updated Purple Peaked, sort of
Sent from my XT1032 using Tapatalk
Well, the use of T filters at the feedback is a common practice for frequency boost. The novelty is that my configuration let's you boost or cut the frequency with the same T filter..similar to a baxandall mid control..but you can get extreme boost or cut.
Thanks for the info Daniel. :)
This looks really neat, Daniel. I want to check it out, but only have TL072s...yes, I could just use them, but what fun would THAT be?
Do you suppose this would work well with TL064's? I still have a few from an Easyvibe build, this might be a good way to use them! I am not sure they are as quiet as the 74's, etc.
Thx.
It should work fine on any opamp.. in this case, the opamp is just making the operation it should do..
But feel free to try different ones while boosting the hell out of frequencies..maybe there's an interesting distortion pedal hiding there :)
I've spent some time playing with EQ with W-shape response.
Well, it's very interesting and can produce a lot of useful tones.
Sorry about the necro, but I've been messing with this circuit a lot and I can't quite get it working correctly. Version 1 with the resistor to ground works great, but the dual pot version is acting weird. I keep rebuilding the circuit and it keeps ending up with the same problem. There is only output when both the gain pot and the dual gang are roughly in the same area of rotation. Example; gain pot at 0-10% then parametric control works from about 0-15% then cuts off the output. If gain pot is around 90-100%, then I only get output when the dual gang is somewhere around 85-100%. I have a 100k dual gang pot and I've tried 500k, 200k, and 100k for the gain pot and I keep getting the same result. I'm using all the values from the 600 Hz section and I have the last buffering stage afterwards as well. Turning the gain pot all the way one way gives a nice peaky Q to sweep, but the other way produces no notch.
Anyone have any ideas? Do I have my dual gang pot wired up wrong? Should I add a buffer before the input?