A question To Jim Thompson

Jim,

Could you please explain me how this :

formatting link
actually works ?

i'm tired doing maths but i launch LTSpice on your reference and ... how could i put it ... it's seems not to be a Notch Folter to me.

Regards, H.

Reply to
Habib
Loading thread data ...

Don't forget to connect power to the OpAmps!!

TomC

Reply to
tomcees_pc

The gyrator part certainly looks gyrator-ish to me. Double-check the power supplies, and that you're taking VOUT where it says, and anything else you can think of.

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

Ok i missed the Vout node, thanks.

Regards, H.

Reply to
Habib

You can just attach your LTSpice file here, or copy it into your post. Then people can try to massage it and re-post the corrected version, one that (hopefully) runs.

--
Regards, Joerg 

http://www.analogconsultants.com/
Reply to
Joerg

To the right of C2 is a gyrator-created inductance. And Tim is correct, check your nodes. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

So Jim. Is the primary motivation for the gyrator to make the notch filter, or is the primary motivation for the notch filter to show off the gyrator?

'cause it seems like there are better ways to make a notch filter, at least for a lot of circumstances, than to make something that's sensitive to both the input and output impedance, or that requires four op-amps where one op-amp and a twin-T network would work.

(and yes -- your circuit would make an easier-to-achieve deep null).

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

The motivation is that someone asked about how to do variable "neck" notch filters. It's been so long ago I don't remember, but I believe it was a discussion of all the filter types you could make with a gyrator core.

In my world OpAmps are dirt-cheap... I don't even blink an eye at systems requiring 10's of thousands of transistors >:-}

(And the other, "simpler", configurations have all kinds of sensitivity issues. My goal, always, is manufacturing repeatability... and minimizing labor costs.)

Yes. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

What do you need that for? I thought everything was wireless these days?

Jamie

Reply to
Jamie

In chip design, you minimize the number of "floating" nodes, where in this vernacular floating means not being driven by a low impedance circuit, i.e. op amp. The parasitics will ruin the circuit response much more than the phase effects (say Q enhancement) of the opamps.

Generally you have one op amp (or more!) per pole in integrated circuit designs. And leapfrog design rules!

Reply to
miso

Ok i'll remember next time.

Reply to
Habib

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

.....>

For no reason other than curiosity, what is the maximum frequency that a circuit like this could sensibly be used at? Assuming suitable OpAmps and layout.

Reply to
John Smith

[snip]

It's a function of the gain-bandwidth product (GBW) of the OpAmps used. I'd simulate to make sure, but I'd guess a reasonable rule-of-thumb would be around GBW/10.

For large signals, you'd also have to watch out for slew-rate effects. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

I figured it was coming from a chip-design sensibility.

Of course, unless I want to start messing with 0402 discretes, four op- amps in a bitty package and seven or eight discretes may be smaller than a "space saving" twin-T.

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

Hi Jim, I've never done a notch thing, But I assume it's like a bandpass turned upside down. In which case the GBW requirements are going to be something like Fmax * Q ~ GBW/10 George H.

Reply to
George Herold

upside down. In which case the GBW requirements are going to be something like

[snip]

In the gyrator configuration, the emulated inductance doesn't depend on Q.

But an interesting prod. I'll test it out. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

The emulated inductance won't depend on Q, no -- but the impedance of the emulated inductance will depend on the transfer function of the op amps.

So as you go up in frequency the emulated inductance will start to affect the Q of the overall circuit somehow -- probably by pulling it down.

--
Tim Wescott 
Wescott Design Services 
 Click to see the full signature
Reply to
Tim Wescott

Did you miss my comments about GBW?

Actually my first guess at the approach to GBW is that the inductance value will decrease.

I'll test it out. ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
 Click to see the full signature
Reply to
Jim Thompson

Apparently, yes.

If each op-amp has a voltage gain k/s, the impedance looking into the positive input of U1 is

V- (2s + k) k C1 R3 R4 s

--- = ----------------------------------------- I 2 C1 R4 s^3 + (3 k C1 R4) s^2 + k s + k^2

This looks right dimensionally, and it goes to V/I = C1 R3 R4 s as the GBW product (GBW = k / (2 * pi)) goes to infinity. So I think I have my head screwed on straight.

Dinner was recent and rich, and intuition is coming slowly to me, but if you assume that C1 R4

Reply to
Tim Wescott

GBW

ad

,

))

Hi Tim, well I'm not going to try and write down the loop equations. I ass ume you did them correctly. But my 'simple minded' understanding is that e verything to the right of C2 looks like an inductor. (Maybe I have this wro ng?) So it's a series RCL. Now if I was to take the output from the other side of C2, then it would be a bandpass filter... And that's where I got t he factor of Q in the max frequency estimate. Hey, if this type of gyrator circuit can give me a bandpass response without the Q factor that would be nice to know.

George H.

Reply to
George Herold

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.