I have an LS20031 GPS receiver mounted adjacent (about 1" away) to my micro-controller board. When I power up the micro-controller, I observe significant reduction in signal strength (the SNR value from the GSV sentences) in the GPS module. The reduction is so large that the GPS is often unable to get a fix from a cold start while the micro-controller is running but has no trouble fixing after removing power from the micro-controller.
My question: What can I do to reduce interference from the micro-controller?
I’ve tried connecting the GPS and micro-controller to separate regulated outputs on a bench supply, and the problem persists. I’ve tried disconnecting all peripherals (various sensors and an LCD) from the micro-controller, and the problem persists.
Here are a few untested ideas: I could put a separate voltage regulator about 1/2" away from the GPS. I could add additional bypass and decoupling capacitors next to the GPS. I could put a ground plane under the GPS.
Have you encountered a similar problem? Do you have ideas on the best approach to solving this problem?
Did you design the micro controller board yourself, or did you buy it (If you bought it, which one is it?)
Depending on the PCB design, it could be spewing a lot of EMI which may be contributing to the problems… You could try placing a grounded shield around the MCU board and seeing if that helps…
That’s a good idea. I hadn’t thought of shielding the micro-controller.
It’s a [Mini RoboMind. The board has a fairly fast (25 MHz) bus and is a two-layer design (i.e. no ground plane), so I could see EMI being a problem. I don’t know how to measure EMI, and I don’t have the experience to estimate if a design is likely to produce EMI.
Directly measuring EMI takes a lot of specific gear to do roughly, so isn’t really an option for hobby electronics… You could see if shielding the device helps, however.
Can I suggest that you seperate the Ground Lines of both boards and that they only meet at one point just from the battery terminal or powersupply terminal. this solved problems with GPS for me in the past.
I’m observing the interference even when the GPS and micro-controller have no electrical connection except through a bench supply. I’m using separate regulated outputs from the bench supply for the GPS and controller.
I do have long (24") leads from the supply, and I don’t have a regulator dedicated to the GPS (there’s a 3.3VDC regulator shared with other electronics, and I’ve disconnected that from the GPS for testing).
I’ll experiment with grounding and a separate regulator. Building a shield will be a pain, so that option is a last resort right now.
I put aluminum foil grounded to a bench supply in a plastic bag (as an insulator so I don’t short everything out). I experimented with various orientations for the the foil and observed minimal improvement in signal strength to the GPS. I couldn’t get the foil completely around the electronics due to wiring and mounting, so I might get better results with better shield construction.
I added a dedicated 3.3V voltage regulator for the GPS, mounted less that 1/4" from the module’s power input. The GPS seems to do a much better job of getting a fix although interference is still present.
So, I haven’t yet solved my interference problems, but I think the additional regulator reduced the problem to “tolerable.”
