Hi,

I have attached a circuit which I decoded from a board. I wonder why such slightly complex interface that uses a TVS, RC network, and opamp. I intend to interface a potentiometer directly to the microcontroller input, but that has made me reconsider it. Would it be great disadvantage if I don't use a kind of protection for the potentiometer?

Also, I've trying to find that TVS diode datasheet without sucess. There a lot of KE diodes, but can't find something related to KE09A.

 

Picture unavailable ??

 

Picture unavailable ??

I have uploaded the picture again. I had uploaded the wrong one.

 

It does seem a bit OTT unless the pot is on long leads remote from the rest of the circuit?

 

It does seem a bit OTT unless the pot is on long leads remote from the rest of the circuit?

The wiring between the pot and the circuit is like 10-15 inches.

 

Where did you get the particular circuit from?
Why did they say it was needed?
Max.

 

Where did you get the particular circuit from?
Why did they say it was needed?
Max.

I just drew it by myself following the PCB traces. So, I don't know what they said it was needed. The signal comes from a potentiometer embedded in an elevation motor.

 

I just drew it by myself following the PCB traces. So, I don't know what they said it was needed. The signal comes from a potentiometer embedded in an elevation motor.

What is your application??
Max.

 

What is your application??
Max.

The motor is like the one attached in this post. It's in a treadmill to sense the elevation. So basically is to sense the position, or incline angle of the plataform.

 

What if the floor is sloping downward? Don't you want to know the angle relative to gravity/Earth's surface rather than some arbitrary reference like a floor or hill? For that, I would use an inclinometer/accelerometer to measure g directly. Most of those devices have digital outputs, but some also have analog (i.e., voltage) outputs.

 

It actually isn't horrible.

You don;t know the input impedance of the uP. The OP amp makes that variation go away.
The TVS can be for ESD protection and spike suppression or to lessen the pickup from the brushes.

The 1K resistor probably protects against the uP not being powered and the OP-amp being powered. i.e. The input current when the uP power is at 0V, so I=5V/1K or 5 mA. The small cap definitely filters the noise as the pot is moving.

The ADC's reference should be the 3.3V supply so it's ratiometric.

 

It actually isn't horrible.

You don;t know the input impedance of the uP. The OP amp makes that variation go away.
The TVS can be for ESD protection and spike suppression or to lessen the pickup from the brushes.

The 1K resistor probably protects against the uP not being powered and the OP-amp being powered. i.e. The input current when the uP power is at 0V, so I=5V/1K or 5 mA. The small cap definitely filters the noise as the pot is moving.

The ADC's reference should be the 3.3V supply so it's ratiometric.

I could not figure out the cap value. It's a small smd ceramic capacitor, could be 10, 100nF?

What about that 20K resistor in parallel with the TVS?

 

I could not figure out the cap value. It's a small smd ceramic capacitor, could be 10, 100nF?

It's probably something small. Ceramic caps usually aren't more than 1 uF. It might be 0.1 uF Remember that at 5 * R*C your at 99% the final value. 5*RC is 5 mS if C = 0.1 uF

What about that 20K resistor in parallel with the TVS?

You need a place for Ib to go and it is part of a voltage divider.
1K & 20K and 0K & 20K at the extremes. you need to know the input Z of the processor.

 

I haven't been able to find the datasheet of that TVS to see its specs. All I know that its marking code says KE09A. I found this datasheet about 1.5KE series, but they are through hole. I don't see something related to KE09A as a SMD TVS.

 

What about that 20K resistor in parallel with the TVS?

It may be there so that if he wires to the pot get broken/disconnected it avoids a floating input and gives a defined voltage to the input.