Hi,

I want to simulate my cable tracker receiver before manufacturing. The total gain exceeds 10 000, so it would be good to at least roughly estimate if there might be oscillations.
How do you calculate the capacitance between the sensor and the PCB where the amplifier will be mounted?
The sensor is partially screened, the screen has a gap of about 1 cm, and the sensor is about 10 cm in length. The distance from the sensor to the screen is about 1.5 cm, and the distance from the PCB to the screen is about 1 cm.
If there weren't a screen, it wouldn't be hard to find the capacity.
Can anyone point me to an online calculator for this kind of calculation?

Thanks in advance

 

Accurately will be hard, but I'd calculate (to know the order of magnitude) just by the area of the screen (1 cm x 1 cm?), the epsilon of air, the distance between the coil and the PCB, there you have it. Just apply the equation.

 

Accurately will be hard, but I'd calculate (to know the order of magnitude) just by the area of the screen (1 cm x 1 cm?), the epsilon of air, the distance between the coil and the PCB, there you have it. Just apply the equation.

It won't be that simple. The screen has a great influence, that's the whole point. Calculating without the screen, the result is unsatisfactory, oscillations

 

When you said the screen has great influence, I remembered I forgot that it is tied to GND.
So, if the PCB has lower potential than GND, the field is extended after the screen through the opening, to the PCB.
If it's GND, then the most capacitance is due to coil to the screen.
The mix between the two is what gives uncertainty in a simple calculation.
I'm pretty sure that it needs simulation, otherwise it would be just a big guess.

But my point that knowing the order of magnitude is enough to have an idea of the order of magnitude of the resonance frequency.
Calculating as if it were a coaxial cable:
1.5 cm of distance from the sensor to the screen = 3 cm outer radius
and inputting (I haven't this information, but it's more or less this value) 1.5 cm to the coil diameter
we get 8.026075568832983e-11 F/m
as the length is 10 cm, this gets us 8.02e-12 F. So it should be in the order of 8 pico farad.
Evaluate if 10 pico farad will give you oscillations. If you can determine what capacitance gives you oscillations, then you compare with this 8 pico (compare the order of magnitudes) and you'll conclude you are safe, without ever calculating exactly the capacitance.

Calculator here

 

When you said the screen has great influence, I remembered I forgot that it is tied to GND.
So, if the PCB has lower potential than GND, the field is extended after the screen through the opening, to the PCB.
If it's GND, then the most capacitance is due to coil to the screen.
The mix between the two is what gives uncertainty in a simple calculation.
I'm pretty sure that it needs simulation, otherwise it would be just a big guess.

But my point that knowing the order of magnitude is enough to have an idea of the order of magnitude of the resonance frequency.
Calculating as if it were a coaxial cable:
1.5 cm of distance from the sensor to the screen = 3 cm outer radius
and inputting (I haven't this information, but it's more or less this value) 1.5 cm to the coil diameter
we get 8.026075568832983e-11 F/m
as the length is 10 cm, this gets us 8.02e-12 F. So it should be in the order of 8 pico farad.
Evaluate if 10 pico farad will give you oscillations. If you can determine what capacitance gives you oscillations, then you compare with this 8 pico (compare the order of magnitudes) and you'll conclude you are safe, without ever calculating exactly the capacitance.

Calculator here

For oscillation to occur, a capacitance smaller than 1pF is enough. That's why they came up with the idea of adding a shield.

 

For oscillation to occur, a capacitance smaller than 1pF is enough. That's why they came up with the idea of adding a shield.

Interesting, so it's the other way around. There have to be some capacitance in order to oscillate.
The shield adds area and is closer, thus elevating the capacitance.

 

Interesting, so it's the other way around. There have to be some capacitance in order to oscillate.
The shield adds area and is closer, thus elevating the capacitance.

A screen placed between two conductors greatly reduces the capacitance between those two conductors. Naturally, this is if it’s connected to GND.
In exactly the same way, the unwanted capacitance between two windings in a transformer is reduced. A shield is placed between those two windings and connected to the ground.

 

You may find the following calculators helpful.
https://femm.info/doku/doku.php?id=electrostaticstutorial
https://www.quickfield.info/pcb_impedance_calculator.htm
https://www.emisoftware.com/toolbox/

 

You may find the following calculators helpful.
https://femm.info/doku/doku.php?id=electrostaticstutorial
https://www.quickfield.info/pcb_impedance_calculator.htm
https://www.emisoftware.com/toolbox/

Thank you for your response, but unfortunately, I didn't find an example where the capacitance between two partially shielded conductors would be calculated.

 

A screen placed between two conductors greatly reduces the capacitance between those two conductors. Naturally, this is if it’s connected to GND.
In exactly the same way, the unwanted capacitance between two windings in a transformer is reduced. A shield is placed between those two windings and connected to the ground.

It wasn't specified that the capacitance was only between coil and PCB. Between coil and GND is another one.
It's interesting.
I calculated the capacitance between coil and screen, so ignore all that.
Between coil and PCB must have a simulation, nothing else can be done. It's because you have a narrow window through the screen, and you need to know the capacitance between coil and PCB, the field is deviated (it curves), thus requiring a simulation. It's like the fringing magnetic lines in a magnetic gap.

 

You can, however, make a circuit that resonates with this capacitance, see the frequency of the oscillation and then calculate the capacitance. It's more realistic than just simulating or approximating the value. Then you can adjust the distance between coil and PCB to see if the capacitance reaches a value you need.

 

You can, however, make a circuit that resonates with this capacitance, see the frequency of the oscillation and then calculate the capacitance. It's more realistic than just simulating or approximating the value. Then you can adjust the distance between coil and PCB to see if the capacitance reaches a value you need.

Here’s the design of my device. If this layout shows too much capacity, I’ll have to change it. That’s why I want to at least roughly estimate it before making it

 

Here’s the design of my device. If this layout shows too much capacity, I’ll have to change it. That’s why I want to at least roughly estimate it before making it

Can't you make it so you can slide the system coil+screen from the PCB? Then you can adjust and have a real experiment. Because sometimes there are hidden variables we aren't accounting for, so the fastest route is to experiment with it. I know it's awesome to simulate/calculate, but sometimes we need the project done ASAP. One simulator (not easy to set up) is COMSOL. It's the only one I know that simulates these kind of things.

Also, why can't you put the hole in the screen to the right/left side? It should give way less capacitance between coil and PCB. Does it have to be aimed to the PCB? Can't you make it narrower?

 

Can't you make it so you can slide the system coil+screen from the PCB? Then you can adjust and have a real experiment. Because sometimes there are hidden variables we aren't accounting for, so the fastest route is to experiment with it. I know it's awesome to simulate/calculate, but sometimes we need the project done ASAP. One simulator (not easy to set up) is COMSOL. It's the only one I know that simulates these kind of things.

Also, why can't you put the hole in the screen to the right/left side? It should give way less capacitance between coil and PCB. Does it have to be aimed to the PCB? Can't you make it narrower?

I have a suitable case for this device, so I'm trying to design according to the existing case. Secondly, it's a portable device, and I don't have much freedom for it to be too big. Two edges of the screen are also part of the case

 

I have a suitable case for this device, so I'm trying to design according to the existing case. Secondly, it's a portable device, and I don't have much freedom for it to be too big. Two edges of the screen are also part of the case

I was talking about a testing prototype, so then you can play with the distance. Once you know what distance is safe or needed, you can build the product.