Opamp values calculation

[lilimike]

Hi,
Based on Nigel's post of Audioguru's opamp in the sticky of this forum.
I am trying to figure out how to calculate the values of C1-C4 and R1 (see attached)

My input is a 40 Khz transducer 400R160.
Also if anyone can comment on the values I have indicated and if I should add or remove caps or resistors.

Thanks

Mike

 

[MikeMl]

R1 is the load impedance seen by your sensor. What does the sensor's data sheet say?
X(C1) should be small compared to R1 at the op. freq.
X(C2) should be small compared to R1 or 50K (which ever is smaller) at the op. freq.
X(c3) should be small compared to 4.7K at the op. freq
X(C4) should be small compared to the input impedance of the LM833 at the op. freq.

X = 1/2ΠfC

"small compared to" should be ~1/100

 

[crutschow]

There is no path for the bias current for the op amp (B). You need a resistor from the + input to the pseudo ground at C2.

 

[lilimike]

Thank you Carl,
I fixed this.

Mike,
I am not quite sure how to interpret the datasheet, this is what I have
also, what is this character " Π " in your formula?

Sorry, my understanding will get better with time.

Thanks

Mike

 

[MikeMl]

According to the data sheet, the receiver should be terminated with 3.9K to get the stated resonance frequency. Π=3.14159 (Pi)

 

[lilimike]

3.9K should be the value of R1 ?

X = 1/2 ΠFC
X = 1/2 x 3.14159 x 40000
X = 62831.8

After this I am not sure where to go!
If you can help me with C1 I think I will manage for the rest.

Thank you

Mike

 

[audioguru]

Terminating the receiver with only 3.9k causes its frequency to shift only 0.4% (160Hz) which is nothing but causes its valuable output level to drop one-third.

 

[MikeMl]

3.9K should be the value of R1 ?

X = 1/2 ΠFC
X = 1/2 x 3.14159 x 40000
X = 62831.8

After this I am not sure where to go!
If you can help me with C1 I think I will manage for the rest.

Thank you

Mike

Should be 100x3900 = 390000=1/(2*3.14159*40e3*C)

C=1/(2*3.14159*40000*390000) = 100e-9 = 100nF = 0.1uF

 

[audioguru]

The resistor R1 should be 10k so that the output level is max and the frequency is changed only a tiny insignificant amount.
The capacitor C1 should be about 0.002uF (2nF) to pass 40kHz into 10k ohms very well.

 

[Hero999]

I'd be tempted to omit R1 and C2 and make the 100k resistors as higher (up to 10M), if possible.

 

[lilimike]

I honestly be tempted to remove all resistors and caps but I know I would run into problems wouldn't I ?
I will defenitely try your suggestion. My first thaught is won't this make noize? from browsing I saw that the higher the resistors the more noize it makes but I will try and see the results on the scope.

Thanks


Mike,

Thank you, I am trying to digest and understand all this. The value 390000 is this from the 3.9K?
(3900Ω)?

 

[MikeMl]

...Mike,

Thank you, I am trying to digest and understand all this. The value 390000 is this from the 3.9K?
(3900Ω)?

Crap, I did it backwards! In order for the capacitor not to reduce the signal, make its impedance (1/(2ΠfC)) less than 1/10 of 3900Ω, so I should have used 3900/10 = 390Ω in the calculation.

390 =1/(2*3.14159*40e3*C)

or C=1/(2*3.14159*40000*390) = 1e-8 = 10e-9 = 10nF.

 

[lilimike]

Based on this C1,C2 and C4 should be=10nF and C3=8nF
Looking at the datasheet I have a feeling 9V is not enough for this opamp so I raised to 12 V

Unfortunately I am not seeing what I expected on the scope, maybe tomorrow things will go better!

Thanks for your help

Mike

 

[audioguru]

The datasheet for the LM833 shows that it works perfectly with a single supply as low as 9V to 10V. So when a 9V battery drops to 6V then it won't work.

Did you add resistors to bias the second opamp?

What doesn't work?

 

[lilimike]

I took everything apart and rebuilt again with same results.
I went back to 9V.
This is my drawing:

Values are:
C1,C2,C5 = .01uF
C3 = .1uF
C4,C6 = .047uF

R1 = 3.3k
R2,R3,R5,R7 = 100k
R4 = 4.7k
R6 = 2.2k

The transmitted signal from the Tx paired driven by a PIC is at steady 40.15KHz
I have built the Rx on a separate breadboard to isolate possible noize from the Tx circuit.

When an object (peace of wood 12 in sq.) is placed approx 2 feet away from the Rx/Tx (which are pointing towards the object) I see a change in the signal at pin 7 but within a few seconds the signal goes back to its original state (like before I added the object) without moving the object away.

On top of this the change in the signal is very small for beeing just 2 feet away.

any ideas / changes I could make?

Mike

 

[lilimike]

Ok I have just re-read...

Did you add resistors to bias the second opamp?

I guess I need a resistor between pin 5 and floating ground instead of direct!
Would using same value as R1 be ok?

Sorry about that.

Micke

 

[audioguru]

Yes of course your circuit is missing a resistor.
C2 is supposed to be a filter so 1uf to 10uf should be used.

 

[lilimike]

Wow!
This is awsome, it is working now and I can actually detect the sealling from my workbench!
Thanks for your help.

Mike