DW1000 Ranging Jitter

Wireless Connectivity Ultra-Wideband
1

Hello all. I have a product that is using a DW1000 based tag and anchor system for ranging, and we’ve been on the market for 5+ years with this. We need fairly steady readings for the system to work properly, and have a production test system where we position a test anchor and a test tag about 1 meter apart, and run the ranging on this for some amount of time. On most of our boards, we get very stable readings, varying by just a few cm reading to reading. However on some amount of the boards, the ranging readings vary significantly 20 - 30 cm of jumping around which is unusable for us. We have been just setting this “bad” boards aside, but would like to find a solution to the issue since we have quite a few of these “bad” boards now.

We have noticed that during testing, if we heat up the “bad” board a bit using a heat gun, the readings stabilize completely. After a few minutes as it cools back down to room temp, the readings slowly start to become unstable again. In looking more at this, just precisely heating up the DW1000 itself, and no other components on the board, makes this occur.

We have put the unit in a CW test mode and verified with a spectrum analyzer that the signal is right on frequency (less than 1ppm from the test radio that is used for the tag) and the CW doesn’t show any obvious jittering. Also just heating the crystal oscillator and not the DW1000 doesn’t cause the problem to go away.

Any thoughts on what to look at next? Could there be something inside the PLL circuitry inside the DW1000 that is causing some instability on some of these units? Or something with the LDE circuitry?

Appreciate any insight that someone many have.

Thanks,

Ken

2

When you have range jitter does the RSSI also jitter?

The only time I’ve seen similar issues has been with poor solder joints to the antenna, this was also associated with a significantly larger than normal variability in receive signal level.

It doesn’t seem like a particularly good fit for the symptoms you describe but may be worth a check.

I’m assuming this is a TWR system, are you single or double sided? If single sided is there an increase in variability of the clock error value? If double sided can you compare the values for each side and see if the noise increase is symmetrical or only in one direction.
I’m not sure what exactly any of this would tell you but my normal approach when facing an issue like this is to try and narrow down where the issue is coming from.