Hello Everyone,

I'm not sure how to handle the low pass filter characteristics from the voltage divider to get 6V.

 

5546 hjjkjijjk kook

 

OMG Simply remove R3 and R8 (replace them with short-circuit) also remove R9, R10 C7 C8.
Reduce R4 and R5 to 10K and use one voltage divider for bias two op amps.
C3 put in parallel with C4.
As for C4 value

C4 = 0.16/( R4||R5 * Fc) = 0.16/( 5K * 0.1Hz) = 330uF

 

I cropped your schematic so it is no longer bigger than my neighbourhood.
I removed the useless parts.

 

ok thanks again

 

1) Is it likely I need to use a buffer op amp between the bias point and the R2 and R6 resistors?

Why? The existing opamp already has an extremely high input impedance the same as a buffer.
But I think the opamps will have difficulty driving the very low values of your frequency determining resistors. Use resistors 10 times higher and capacitors 1/10th what they are now.

2. I have polarized caps for C8 and C4 in the revised diagram. I am guessing those are the only two that need to be polarized?

C7 also has a fairly high value and will usually be a polarized electrolytic type.

. In the revised schematic, is it acceptable to change C3 to 0.1uF and C4 to 10uF? The cutoff frequency then becomes 3.15 Hz

No.
C9 is not needed. C3 should be 10uF to 100uf. It is a hum filter and does not create a cutoff frequency.

What about if I use 100K resistors for R3 and R4 with C3 = 0.1uF and C4 = 10uF?

If you have hum then the value of C4 should be high. If the biasing resistors are 100k each then C4 can be 33uF. Then it is a good filter to reduce the hum.
If you don't have hum then C4 is not needed and C3 can be 0.22uf.

 

Thanks so much for your help!

 

1. I thought a buffer op amp was important based on what I read here on page 332
and shown in this schematic:
Do you think I need it in my particular case?

The opamp has a low output impedance which is important. But what is the impedance of a 33uF filter capacitor at 35kHz? Only 0.14 ohms which is almost the same as the opamp. So the buffer opamp is not needed.

2. For C7, I looked on Newark.com and they have plenty of 10uF ceramic capacitors rated at 25V. Is it likely I can use one of those instead of a polarized one?

At 35kHz, a 10uF capacitor has a reactance of only 0.46 ohms. But the minimum load for the circuit is only 1k ohms so a 0.022uF film capacitor will be fine. A 0.1uF film capacitor can also be used. Besides, a ceramic capacitor is a microphone that picks up vibration, plus it adds distortion.

3. Does it hurt to leave the C9 1uF capacitor? Someone told me to use those three bypass capacitors off the 12V supply to filter a wider range of noise.

C8 is 0.1uF which is good for high frequencies and C9 is 10uF (or 100uF) which is good for low frequencies. Then C9 is not needed.

4. The circuit is powered by a 12 volt battery. The same 12 V battery powers a preamp circuit that feeds the signal to this circuit. It seems like hum would not be an issue but it doesn't hurt to assume it is? So I should use C4 = 33uF and C3 = 0.1uF?

A battery does not produce hum so C4 is not needed.

5. How come C3 doesn't come into play for the cutoff frequency calculation? It seems to me like 1/(2*R*C) should use the parallel of R3 and R4 (50kohms if both are 100k resistors) to get R and the parallel of C3 and C4 to get the C value.

C3 simply filters the bias voltage at high frequencies. At 35kHz its reactance is only 46 ohms but it is in parallel with C4 that has a lower reactance. Both do not affect the cutoff frequency of 35kHz.

 

Thanks Audioguru!

 

I thought the buffer op amp makes the equations/analysis easier from what I had read. With the R3, R4, C3 and C4 components there is no effect on the resistor and capacitor values C1,C2, R1, and R2 to achieve a 35 kHz high pass cutoff?

In this case buffer doesn't help to much.
But I will also change as Audioguru suggested filter resistor values.
R1 = 3.9K ; R2 = 5.1K ; R6 = 11K ; R7 = 2K and all filter capacitor to 1nF
And then change R3 and R4 to 100K

I'm confused how the high impedance of the op amp pin and the low impedance of a 33uF filtering capacitor of 0.14 ohms at 35kHz allows for the values of the C1,C2,R1, and R2 components not changing.

Simply, all AC current will be shorted to the GND by low impedance of a 33uF filtering capacitor.

I see you mentioned the minimum load for the circuit is 1k ohms. Does that come from the spec sheet for the op amp or how is that value determined? The reactance of the capacitor at 35kHz just needs to be under 1k ohms? Why is that important? Is it just so the load impedance is higher for proper signal transfer?

Op amp distort much heavier when output load is increased.
So you should not use load smaller then 1K.

I am going to use batteries to power the preamp and this filter but there is a chance in the future I may use a DC-DC converter if I use a single battery to power other components that are not related to the preamp or filter. Is C3 = 0.1uF and C4 = 10uF better than just using C3 with 0.1uF in case of any noise effects from the DC-DC converter or components powered by it?

You can leave C4 = 10uF if you want.

 

thanks again

 

I have a couple of questions still:

 

Why you change component value ? Why you not use a standard E24 resistor values ?
Also don't forget about component tolerance. So you real Fc will be slightly off from that you calculated.

I had read somewhere for unity gain amplifiers sometimes it is better practice to use a resistor for feedback instead of none. Is it likely needed with the op amps I am using for this circuit at unity gain? If so what resistor size makes sense for this op amp?

Normally we don't use any extra resistor.
http://www.analog.com/library/analogDialogue/archives/41-05/input_protection.html
And sometimes we add the extra resistor to compensate offset current.

Should I worry about an input resistor, like in the preamp circuit I had posted on the other forum? Not sure what value to use if so.

Input resistance should be at least 10 times greater then source output resistance.

Should I worry about an output resistor after C7, like in the preamp circuit I had posted on the other forum? Not sure what value to use if so.

No,
C7 = 0.16/ ( F * Rload)
To reduce the C7 effect on the circuit we should choose F = Fsignal_min/10

4. I'm guessing a pull down resistor after C7 is a good idea?

But maybe the next stage has a pull down resistor.

I have a general question for C7 not related to the filter circuit. If I want to use a capacitor to block a DC offset but still want to work at low frequencies such as a 50 Hz or higher, then a larger value capacitor such as 10uF is a better idea. Should I use a polarized tantalum capacitor instead of a ceramic if I need a value that large and am dealing with a signal in the 50 Hz to 150kHz range? The reason I ask is that perhaps I want to bypass the filter circuit in some instances and just use the preamp at frequencies below the 35kHz. This means I need to keep a 10uF capacitor on the preamp output to block the DC offset.

Use tantalum capacitor or electrolytic. Or use high load resistance then you can use film capacitor.

 

A final dsiscussion about using a buffer opamp for the bias:
1) The impedance of the bias source should be 1% or less than the frequency-determining resistors (R2 is 500 and R6 is 1.1k but they should be higher) because the impedance of the bias source is in series with the frequency-determining resistors. The two resistors in parallel are 344 ohms and 1% of them is 3.4 ohms.
2) The TL081 opamp has Jfet inputs that need no bias current so they are an extremely high resistance. Two 100k resistors can easily provide a 6V reference voltage for them.
3) The output resistance of a TL081 buffer is about 200 ohms which is reduced by its negative feedback voltage gain. At 35kHz its negative feedback voltage gain is 100 so its closed-loop output impedance is 2 ohms.
4) A 33uf capacitor has a reactance of only 0.14 ohms at 35kHz which is much better than a buffer opamp.
The TL081 has a fairly high frequency response. Stray capacitance to ground on its output (maybe caused by a shielded audio cable) combines with its 200 ohms output resistance to cause phase shift which causes the opamp to oscillate at a high frequency. The oscillation does not occur if the circuit has a series 100 ohms resistor.

The OP talks about using a 10uf output capacitor to pass 50Hz. Then its load can be as low as 320 ohms to 1600 ohms which is very low. Most opamp loads are 10k ohms or higher then the output capacitor can be a 0.22uF unpolarized film capacitor.

A tantalum capacitor is not reliable and it causes distortion. A ceramic capacitor is a microphone since it picks up vibration and it causes distortion. They can be used for power supply filters but should not be used in a signal path.

I am confused about the function of this highpass filter. What is the ultrasonic input to the piezo-ceramic sensor?

 

Hi Audioguru,

Thanks for making those points. I will try to think about them and understand them.

 

Hi Jony130,

I chose the resistor values

 

Does anything communicate at ultrasonic frequencies beneath the surface of the ocean? Dolfins use sound frequencies but some scientists "guess" that they hear 30kHz to 40kHz. Whales use very low sound frequencies.

Submarines use very low radio frequencies.

 

Hi Audioguru,




What is the best solution?

 

An opamp can drive a low power voltage divider that feeds a low signal voltage to 50 ohms.
An opamp can drive a transistor that can feed high power into 50 ohms.

Take your pick deterrmined by how much power is needed into 50 ohms.

 

Thanks audioguru,

I'm having a tough time knowing which of the two solutions I need to go with, and I'm not even sure how I would implement it.