Integrator Circuits for PI.

[Ferric Toes]

Here are a couple of alternative arrangements that I have used from time to time. The differential version is interesting as it enables you to change the integration time with a single pot. The track/hold arrangement prevents the voltage droop you get with the more common "leaky" integrator where there is a high value resistor directly across the cap.

Eric.

[Altra]

Eric,
Thanks for posting these circuits.

I always wondered what the function of the other amp was for in the second drawing. Any results from the VCA log amp? I received mine, but they are surface mount so they need to be mounted on an adaptor. On the to do list.

Mark

[Orbit]

[Ferric Toes]

Eric,
Thanks for posting these circuits.

I always wondered what the function of the other amp was for in the second drawing. Any results from the VCA log amp? I received mine, but they are surface mount so they need to be mounted on an adaptor. On the to do list.

Mark

Basically the second amp just performs an inversion so that you can feed back to the + input of the first amp to keep things symmetrical. However, by giving it a bit of loop gain you can alter the TC of the integrator.
My VCA610's are still waiting for attention too. At least the 8DIL's are harder to lose.

Eric.

[Midas]

Here are a couple of alternative arrangements that I have used from time to time. The differential version is interesting as it enables you to change the integration time with a single pot. The track/hold arrangement prevents the voltage droop you get with the more common "leaky" integrator where there is a high value resistor directly across the cap.

Hi Eric,

Thanks for posting these. Can you tell us what the advantages are of avoiding 'voltage droop'?

Midas

[Davor]

The track/hold differential configuration resembles an integrator that Douglas Self refers to as "2C integrator".

[Ferric Toes]

Hi Eric,

Thanks for posting these. Can you tell us what the advantages are of avoiding 'voltage droop'?

Midas

Precious little in reality. Just for the purists. The differential circuit's main advantage was the ability to slow up the response in noisy EMI situations. This was used in the Superscan detector.

I don't use either arrangement now as matching the two sets of RC components was always a pain and you get a far better performance by summing + and - signals into a single TC integrator. More on that later.

Eric.

[Ferric Toes]

The track/hold differential configuration resembles an integrator that Douglas Self refers to as "2C integrator".

Where does Douglas Self refer to the 2C integrator? I used to enjoy his articles in Electronics World some years ago on aspects of audio amplifier design.

Eric.

[Davor]

It is in the 5'th edition of his Audio Power Amplifier Design Handbook.

[mikebg]

Let we reinvent the integrator used in classic PI metal detectors.
WHY WE NEED AN INTEGRATOR?
What information delivers it for operator?
How should the integrator work?
Read what is written by John Corbin:

[mikebg]

Now read what is written for integrator in Hammerhead:

For the differential integrator to work properly, both sampling pulses need to have the same pulse width, so both pulse width monostables (IC11b & IC12b) are controlled by the same RC time constant, R46/R46a/C26, by using isolation diodes D6 and D7. This scheme*) works nicely, as long as the two pulses have a sufficient amount of delay time between them, which is easily met in this application.
_____________________
*) Suggested by Eric Foster.

Here is the principle:

[mikebg]

The lower diagram illustrates what an integrator calculates . This is a DC voltage proportional to an area. The calculation of area starts at moment t1 and ends at moment t2.
Let we analyse the circuit diagram.
What is timing of switches S1 and S2?
What calculates the "Track/hold differential" integrator?
What information we need for target?

[deemon]

Let we reinvent the integrator used in classic PI metal detectors.
WHY WE NEED AN INTEGRATOR?
What information delivers it for operator?
How should the integrator work?

You are a good theorist , Mike ... but don't forget that electronics is not only a theory . Electronics is a kind of art

[Ferric Toes]

The lower diagram illustrates what an integrator calculates . This is a DC voltage proportional to an area. The calculation of area starts at moment t1 and ends at moment t2.
Let we analyse the circuit diagram.
What is timing of switches S1 and S2?
What calculates the "Track/hold differential" integrator?
What information we need for target?

A typical example is S1 closes after a delay of 10uS from TX off and stays conducting for 10uS. There is then a further delay of maybe >100uS before S2 conducts for an identical period of 10uS. The repetition rate rate of the TX/sampling cycle may be anything from 200pps to >3000pps, depending on other factors of the detector design. This type of differential sampling cancels any offset from a dc coupled preamp, waving the coil in the earth's field, remanent magnetism in hot rocks, and LF interference from power lines. Differential signals obtained from sampling a decay curve are not cancelled, although the S1-S2 timing may have to be increased to avoid partial cancellation of long TC objects.

Eric.

[deemon]

Differential signals obtained from sampling a decay curve are not cancelled

Of course , the main problem in the classic PI design is the decay curve itself . And the worst thing is that this curve ( and its integral ) can depend on many factors . This fundamental problem which cause DC instability of the whole device ...