Signal internference on the JRK 21v3?

I’m using the JRK 21V3 and have no problems using it with feedback and using a step signal as the input signal (0 to 5V). Now I’m trying to have the input signal be the voltage off capacitor of a simple series RC circuit. The voltage of the capacitor is directly tied to the Rx input pin and power to the RC is switched between 0 and 5V. Normally the RC circuit has its power tied to 0v and then it switches to 5V charging up the Capacitor over a few seconds.

I’ve noticed two issues. First is the RC time constant is not expected for the measured values of the resistor and capacitor. Second, there is a DC bias on the capacitor even though it is suppose to be shorted and no voltage present. Furthermore, the value of this voltage is dependent on the resistor value. I’m deducing that something on the Pololu JrK 21V3 may be interfering with the RC circuit.

Is there a pull up or pull down resistor on the input line of the JRK? Is there a schematic available for the Pololu JRK 21V3? Thanks,


You sent an email similar to this to us yesterday; it looks like you might have posted this before I sent you a response. As I mentioned in my email, the RX user pin on the jrk is directly connected to a pin on the on-board microcontroller. That pin has an internal pull-up resistor that cannot be disabled. It is probably making a voltage divider with your RC circuit.

In your email, you suggested using an isolation circuit, which might work, but I do not have any specific recommendations for such a design. If you do use one, I would be interested to hear how it turns out.

We do not make the schematic for the jrk available.

- Jeremy

There were two issues but each one required a different fix. The issue of the DC bias is corrected with an isolation amplifier. I used a LMV321 set for unity gain and that allowed the voltage on the capacitor to drop to near zero as designed. However, I was still getting incorrect time constants as expected. I was using the built in 5V supply on the JRK 21V3 and it turns out there is nearly 8uf capacitance on its output affecting the capacitance in my circuit. Easy remedy, add a separate 5V source. Here is the schematic for the interface board to the JRK 21V3:

I am glad you got it working. Thanks for letting us know.

- Jeremy

I’ll ask about another input question that I’ve been wondering about. The users manual states under Input Scaling:[quote]By default, the scaling is linear, but you can change the Degree parameter to use a higher -degree polynomial function, which gives you better control near the neutral point.[/quote]
Does this mean that the Target Value = Input Value raised to the 2,3,4, or 5th power centered around the neutral point? I’ve observed that is drastically changes input characteristics but its unclear to me exactly what happens when the users selects a degree other than linear. Could you please clarify the degree function? Thanks,

Yes. For example, when the degree is 2, the relationship between the input and the target become parabolic, where the
center of the parabola (the point where the derivative is zero) is located at the point where the neutral input and neutral target meet and the other point of the parabola is located at where the maximum input and maximum target meet. You still get the full mapped range, but target changes more slowly as a function of input when the input is near the neutral point and more quickly when the input is near the maximum (or minimum).

Could you try this with your jrk and see if this is what you observe?

- Jeremy