Hello dear forum 
I am helping a friend with his Elephant-project to prevent them from getting killed my humans. So the aim is to record Elephant rumbles (which is mostly infrasound (<20Hz)) with a network of microphones, which calculates their location. (GPS collars are not really animal friendly and costly).
Sooo. I got this circuit which supposed to work. It's a preamp and low pass filter which attenuates frequencies above 150Hz.
I build it and i get this high pitched noise. (Soundsample)
Unfortunately i am not smart enough with electronics to find out the reason. I used audio-grade capacitors and checked all the connections and wirings twice (but of course never 100% sure).
Where does this high pitched noise come from? What can i do to find out about it?
That would be great if someone can help me out. I am sure the elephants will thank you for it
ED
I am helping a friend with his Elephant-project to prevent them from getting killed my humans. So the aim is to record Elephant rumbles (which is mostly infrasound (<20Hz)) with a network of microphones, which calculates their location. (GPS collars are not really animal friendly and costly).
Sooo. I got this circuit which supposed to work. It's a preamp and low pass filter which attenuates frequencies above 150Hz.
I build it and i get this high pitched noise. (Soundsample)
Unfortunately i am not smart enough with electronics to find out the reason. I used audio-grade capacitors and checked all the connections and wirings twice (but of course never 100% sure).
Where does this high pitched noise come from? What can i do to find out about it?
That would be great if someone can help me out. I am sure the elephants will thank you for it
ED
Attachments
The tone seems to be 4096Hz and its harmonics, a suspiciously power-of-two frequency suggesting it is interference from some digital circuitry near the microphone or amplifier? (Audacity is a handy tool to have!)
Mic amps need to be in a screened box, and mic cables need to be screened, as they are very sensitive to interference.
Actually given the low-pass filtering perhaps the interference is getting in after the amp, or to the signal ground line?
Mic amps need to be in a screened box, and mic cables need to be screened, as they are very sensitive to interference.
Actually given the low-pass filtering perhaps the interference is getting in after the amp, or to the signal ground line?
I think the schematic is correct (standard inverting AOP). But you're right there is maybe an oscillation. You should populate R5 (10 to 50 ohms) to drive capacitive load. Could you reduced C2 (then increase RV1 to have same fc) ? You could try to add a small cap in // with R3 (100pF for example).
OK. I think best way to correct the circuit is:
- remove C2
- add a cap in // with R3 (low pass filter). This cap in // with R3 is the same think than C2 in // with VR1. As VR1 is < 10xR3 -> C in // with R3 can have a value = C2/10 (100pF)
- Add a cap in series with R5 (between R5 and VR1) because there is a DC voltage at output. DC block is missing.
- remove C2
- add a cap in // with R3 (low pass filter). This cap in // with R3 is the same think than C2 in // with VR1. As VR1 is < 10xR3 -> C in // with R3 can have a value = C2/10 (100pF)
- Add a cap in series with R5 (between R5 and VR1) because there is a DC voltage at output. DC block is missing.
I looked at the frequency with Audacity and it really is around 4100Hz.
The mic is connected with a 10cm wire. But even i pack it all up in a metal box. Or place it further away from other electric devices. There is no change of the output signal.
Nice. I learned about feedback in amps. Thanks for that hint. "Sounds" like this can be the problem. I checked the wiring again. Specifically around R3. But can't see a mistake.
Great. Something to try. I will look into that tomorrow. Already after 2AM here.Kartapus said:
But just to be sure i understood correctly (I am a novice):
- R5 is just a wire (zero Ohm) right now. But i will change to 50-100Ohm
- I only have elcos here in different capacities. So must order first (any values in mind?)
- the 100pF in // to R3 should be foil or ceramic? (Need to buy or find on broken PCBs first)
a last important remark because it is not done on my simulation. Please be aware to correctly bias the micro. See datasheet but value of R4 depends on current and voltage necessary for the microphone. A value of 10k under 9V could meet most of micro requirements but maybe it should be verified. If micro needs at least 5V with 1mA then R4 max = 9 - R4*I = 5 --> R4 = 4 kohms max.
All right. Cool.
Novice recap:
- Remove C2
- Add 100pF // to R3
- Add R5 (50 - 100 Ohm)
- The cap in series with R5 has to be which value? Foil, ceramic, electrolytic?
About the simulation software. Looks too sophisticated to me. But i will check it tomorrow too
Can i just change to a better one? You have one in mind?
OPA1678 is pretty good and not too expensive, but there is no version in a DIP package.
Kartapus simulates with an OPA1642, which is also far better than an LM358 and not available in a DIP package.
By the way, cool project!
Hello dear forum
I am helping a friend with his Elephant-project to prevent them from getting killed my humans. So the aim is to record Elephant rumbles (which is mostly infrasound (<20Hz)) with a network of microphones, which calculates their location. (GPS collars are not really animal friendly and costly).
Sooo. I got this circuit which supposed to work. It's a preamp and low pass filter which attenuates frequencies above 150Hz.
I build it and i get this high pitched noise. (Soundsample)
Unfortunately i am not smart enough with electronics to find out the reason. I used audio-grade capacitors and checked all the connections and wirings twice (but of course never 100% sure).
Where does this high pitched noise come from? What can i do to find out about it?
That would be great if someone can help me out. I am sure the elephants will thank you for it
ED
Where is the output DC blocking cap? The DC value on the opamp should be 1/2 B1. What is with the second
opamp? Are you using it or is it just 'dangling'. It could be your problem. Tie pins 6&7, tie pins 5&3.
G²
How low is the lowest frequency of interest? That's important for calculating the required coupling capacitor values.
I really appreciate all your help. That is awesome. I will try the solutions which needs extra parts after i got them shipped here which might take some days (i am on Sumatra far from any city...)
They were in "dangling" mode... but after connecting them. Still same problem.
Just talked to our Elephant guy and he said: in general 10Hz but some go down to 5Hz
They were in "dangling" mode... but after connecting them. Still same problem.
MarcelvdG said:
Just talked to our Elephant guy and he said: in general 10Hz but some go down to 5Hz