Most reliable WiFi Shield configuration?
I'm trying to get a higher sample rate than the Cyton provides. WiFi shield can purportedly raise this limit but is very buggy, with, in my experience, great difficulty successfully initializing sessions among other issues, as the purchase page states:
IMPORTANT: the WiFi shield is known to have reliability issues (cyclical noise spikes and dropped data packets) when paired with Cyton and CytonDaisy boards.
Question 1) Does this mean the WiFi shield is much more reliable when paired with Ganglion? (i.e. No cyclical noise spikes or dropped data packets.) I don't mind sacrificing channels if this is the case.
Question 2) I've also been told that some of the issues with the WiFi Shield can be resolved by avoiding the 3.7V LiPo battery and instead using one, or even two, AA battery packs.
Comments
Easy answer from the person who pushed for this disclaimer:
Yes others have reported better luck with two separate AA packs.
@retiutut Thanks for getting that disclaimer up there. I only wish I had taken it more seriously! I definitely need the data as reliable, consistent, and real-time as possible. And though I'm guessing you tried everything before falling back to the Ganglion, I'll ask anyway — what are the chances that two AA battery packs will fix the noise spikes and dropped data packets with the Cyton?
Other threads on the forum here, have suggested that the power supply regulator on the Wifi Shield, has poor voltage regulation. Not sure why it is so much better with Ganglion. Perhaps Cyton makes more demands on the ESP processor, or causes it to work harder.
http://openbci.com/forum/index.php?p=/discussion/2108/measured-shield-regulated-3-3v-supply-has-6v-dip-at-1-second-intervals-with-resonant-ringing
Now he was initially seeing noise when powered from lithium cell at 3.7V. 4 AA alkaline cells will give you 6V, which can overcome the power drain spikes. It's unclear whether you need separate batteries on Cyton and Shield, but with a separate AA pack on the shield, that will have the maximum likelihood of overcoming power drain spikes. Cyton could be powered by just a lithium cell.
Additionally Jason (alwayswearshats), applied a mod in terms of new capacitor on the regulator output. Most of his tests were done originally with the lithium cell powering the Shield. In later tests (after he had already switched to the AA pack), the AA higher voltage corrected the packet drops. But I believe others have reported that the AA pack alone for the Shield will make a marked improvement and reduction of noise / drops.
Regards, WIlliam