RED LED stuck on

Hi Jeff,

Are you making sure that the coil is left untouched while the LEDs flash?

Thanks,

Hi Charles,

I have a similar problem. For my curiosity I removed the coil and connected my own coil to the J2 connector. I also connected a 2.2nF capacitor. Everything worked well, then I reconnected the original coil with the 100pF capacitor on-board. Now the red LED is stuck ON.

I couldn't find anywhere in the documentation the purpose of the red LED. Can you please describe it's functionality?

Thank you.

Liviu Banuta

Hi Liviu,

When the EVM powers up, it runs through a series of Rpmin and Rpmax codes centered at the average EVM coil value in order to find the most optimum value for coils that have Rp parameters quite similar or identical to that of the EVM coil.  During this time, the red/green LEDs flash.  If the calibration passes, it would show a green LED, if it fails, it would show a red LED.

When the coil is replaced with a Rp that lies beyond the calibration range, it will always show red.

Of course, this does not indicate that it would not work; you can always modify Rpmin and Rpmax after the device is connected.

Initially, the idea was the user could replace a coil with similar Rp without going through manual calibration (e.g. no impedance analyzer needed).  However, since the range is limited to save on calibration time, we will be storing parameters on flash for future firmware revisions.  This would also allow the user to save device parameters in the microcontroller on subsequent PORs.

Stay subscribed to the software thread for updates.

Thanks,

Hello Charles

I restored the values in the register settings back to their defaults and the red LED still stays solid ON. Also, I changed the Rpmin and Rpmax to several other values with no effect. I purchased two EVMs and the other one works well.

Thank you

Hi Liviu,

Are you getting any data out of the red LED board?  The red LED board interfaces an EVM coil, correct?

Thanks,

Hi Charles,

Both, Proximity Data and Inductance are 0 from the startup. As I mentioned before, changing the register settings does not help, neither placing a target next to the coil.

Thank you.

Hi Liviu,

We've seen this symptom before and it turns out it was a cold solder.  Try reflowing a bit and if it doesn't work please email me.

Thanks,

Hi Charles,

Reflow didn't help. I looked at all components with a microscope and couldn't find a soldering defect.

Thank you

Liviu Banuta

Hi Liviu,

Can you ship this defective EVM to us for analysis?  We will send you a new one.  Please email me (see my email in my profile page) for details.

Thanks,

Hi Charles,

I took a few days to reply because I ordered a second development board. My results are similar to those reported by Liviu. For my new board both the coil that comes with the dev. kit and discrete tank circuits work good. Next I went back to the original board which has problems. What I learned was that the board was fine. The coil/capacitor associated with that 1st board does not work. I have rechecked the solder joints and the behavior does not change.

For my application I plan to hand build a coil ~75mm in diameter. Can you provide me an application note for setting up the LDC1000EVM using my impedance measurements (configuration page)? If you have any additional information on building larger air loop coils to work well with the LDC1000 that would also be helpful. My present understanding is that:

a) resonant frequency should stay between 1-5MHz, but it can go as low as 5KHz
b) RP must stay between 800-4MOhm
c) Parallel capacitor 20pF-100nF

I appreciate the help, Jeff

Hi Jeff,

Your understanding is correct.  However, the parallel capacitor can go up to 300pF in some cases, even more (like 10uF) if you decide to significantly lower your resonant frequency.  You will have to measure Rp of your particular coil to determine the proper rpmin and rpmax parameters.  There is an empirical way to do this, see the FAQ.

For a 75mm coil it all depends on the gauge of the wire, and its expected inductance.  You might find for smaller number of turns that the inductance is too small, so an inductor must be put in series to obtain oscillation.  1V pkpk on CFB would be a good indicator that the device should be outputting data.

Thanks,