bandpass filter

It is a thread. Learn to follow it. This service does not allow a lot of quoting.

IOW "I meant to do that." ;)

Cheers

Phil Hobbs

-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics

hobbs at electrooptical dot net

Run the Spice filter thing that I posted and explain to me why it can't do what it does, and how it violates some law of thermodynamics, as you said it would.

Like I've been accused of doing. And of vioating causality, an even bigger sin.

Hey, I *can* predict the future: There will be yet more idiotic whining here.

I can predict the future too...

You will again call me "old git"... repeatedly.

You will never post your wondrous 40kHz version of the zero phase BP, showing how it is beneficial. (Just saying it works isn't sufficient, show us.) ...Jim Thompson

Yeah. Exactly right. I hadn't thought about it much before this thread, so it has been very good in that way.

Getting positive phase shift (negative delay) what I was trying to do by adding the shelved PD to the circuit I posted. I didn't optimize that circuit, and I could have added another shelved PD and still had BP characteristics.

It does not matter, because the first step is to get the background idea straight, *and* any final circuit @40kHz would likely be active, but a loaded Q of 1000 for an actual LC tank is not practically realizable under any normal circumstances.

The shelved PD (not a BPF) has almost entirely negative group delay at frequencies below the upper shelf. In retrospect, I think it is the simplist way to illustrate the truth of "Group delay =/= time delay." Here is a shelved PD showing that:

=========================== .asc

Version 4 SHEET 1 880 680 WIRE 96 16 64 16 WIRE 208 16 176 16 WIRE 64 128 64 16 WIRE 64 128 -96 128 WIRE 96 128 64 128 WIRE 208 128 208 16 WIRE 208 128 160 128 WIRE 208 176 208 128 WIRE -96 208 -96 128 WIRE -96 352 -96 288 WIRE 208 352 208 256 WIRE 208 352 -96 352 FLAG -96 352 0 FLAG -96 128 in0 IOPIN -96 128 In FLAG 208 128 out0 IOPIN 208 128 Out SYMBOL res 192 160 R0 SYMATTR InstName R1 SYMATTR Value {R} SYMBOL res 192 0 R90 WINDOW 0 0 56 VBottom 2 WINDOW 3 32 56 VTop 2 SYMATTR InstName R2 SYMATTR Value {k*R} SYMBOL cap 160 112 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 SYMATTR InstName C1 SYMATTR Value {C} SYMBOL voltage -96 192 R0 WINDOW 123 24 124 Left 2 WINDOW 39 0 0 Left 2 SYMATTR InstName V1 SYMATTR Value "" SYMATTR Value2 AC 1 TEXT 184 -160 Left 2 !.param k=10**({dBa}/20) - 1 TEXT 184 -216 Left 2 !.param R=1/(2*pi) TEXT -240 -72 Left 2 !.ac oct 100 .01 100 TEXT 184 -104 Left 2 !.param C=1/(2*pi) TEXT 184 -272 Left 2 !.param dBa=30

=========================== .plt

Unless you are somehow getting younger. Or nicer.

You don't know how to frequency scale a passive filter? Do you need me to post the equations for you?

Nope. Don't need you at all. Does anyone ?:-}

But please explain the benefits. Show a real example of the improvement.

The flat phase region is inversely proportional to Q, so I don't see what the exercise improves over, for example, a two-complex-pole-pair BP. ...Jim Thompson

A conventional bpf has a big phase slope in the passband. The one I posted can have a very small phase change in, roughly, 1/3 of the 3dB bandwidth. And really small in, say, the middle 10%.

I also posted a version where the basic circuit was cascaded with another 2-pole bpf, to improve stopband attenuation. It still had the phase flat in the middle.

If you don't have an application for a thing like this, don't use it. I have an application for it. Two, maybe.

I asked a simple question about I filter that I'm interested in. Several people agreed that it's impossible, and I still thought maybe it was possible, so I tried some things and one works. Interesting follow-up discussion ensued, included causality, negative group delay, network theory. All you have done is whine.

Most bp filters cascade resonators; this one runs them in parallel, summing the outputs. None of this sort of thing seems to interest you.

Bwah bwah bwah, As I said before...

Instead of the "Ernst Guillemin" routine, please show us your final values for your 40kHz implementation... so I can run a square wave thru it >:-}

Don't reply with a retort that I should be able to do it myself. Of course I can, but anything I derive will be labeled by you as "not what you meant".

So show us your grandiose solution already >:-}

But everyone knows you won't... you'll have one of your usual scatological-brained excuses.

S*it-for-brains BS... nothing factual. ...Jim Thompson

The florescent lamp above my bench is flickering. I think I need to replace it.

BTW, tomatoes and celery are very good together.

What is that supposed to mean? You don't like 1 H chokes?

In fairness, amongst the whining, he did suggest a PLL which seems another way to do it? Interesting way of looking at the problem perhaps.

But you got PH to think again which has got to feel good! :)

As I've said more than twice, I plan to use the filter ahead of a PLL.

I'm a simple circuit designer, and Phil is one of the smartest people on the planet, so, yes.

He apparently doesn't know how to scale filters.

But he doesn't actually care about filters. He just wants to squawk and cackle.

Guillemin was noted for his classroom circuit examples that were always 1 Ohm, 1 Henry and 1 Farad. I lucked out and didn't have to endure such useless frivolity... the Honors EE group had Harry B. Lee for passive circuits... where I learned how to cope with the real world.

Of course I do. In the early '70's my primary business at Dickson Electronics hybrid facility was filters. I even developed a "cute" way to avoid problems with dissipation factor at very low frequencies...

What Simon (or you) deleted from the quotation was...

"Don't reply with a retort that I should be able to do it myself. Of course I can, but anything I derive will be labeled by you as "not what you meant"."

And this...

"I can predict the future too...

You will again call me "old git"... repeatedly.

You will never post your wondrous 40kHz version of the zero phase BP, showing how it is beneficial. (Just saying it works isn't sufficient, show us.)"

It IS amusing, this flat spot in phase, but I have yet to see a demonstration that it actually improves performance. Anyone want to show its benefit/value? ...Jim Thompson