spikes & railed signals
ncnaret
no
Hi,
1. when I stack Cyton+Wifi+Daisy, I get spikes in regular intervals. It looks something like that:
https://imgur.com/a/pjJuWwQ
At first I thought there might be some source of interference, but I've tried it at different locations at different times and yet I saw these spikes. At some point I've realized the intervals get way shorter when the batteries are newly charged, intervals of < 3s. (Yes, I only use the provided battery holder with 4 x 1.2V chargeable batteries).
2.
after a while, the signals get often "RAILED", I don't even have a cap connected (only the y-cable is attached):
https://imgur.com/a/TFqg7un
I've been reading in some posts of this forum, that the "RAILED" signals should disappear after a while, but for me they rather appear. If I wait for like 5min, they sometimes disappear e.g.:
https://imgur.com/a/C3Wv9OR (yet you still see these spikes on some channels).
Yet, sometimes the "RAILED" signals appear again.
Is there anything I can do to un-RAIL these? or to prevent these to appear in the first place? What is the reason for this "RAIL".
I've tried to run without the WIFI-board, only Cyton+Daisy (via the COM emulation dongle). I don't have these spikes, yet the same "RAILED" behaviour. I've written my own software (for COM and UDP captures) and in both cases I get signals that are 0xff on many (if not all) channels. In my first (and only) run wearing the cap, I didn't have the "RAILED" issue, but I got weird signals:
https://imgur.com/a/AbJlNRz
Nowadays I get all channels "RAILED" when I try to connect the cap. I've tried to only connect "REF" and "BIAS", to ground the device, in this case it takes some time, but after a while all channels still end up "RAILED".
(Sorry if these questions sound very "beginner"; I'm a software engineer, no real electronics or medic-research experience, this is my first experience with EEG. All I've done (connecting the boards, cap cleaning, etc.) is based on the "Documentation"-area, but that's already where I've reached my limits of resolving these issues myself. I've contacted the mail support before, but they said I might get better answers asking here.)
Comments
NC, hi. What 'cap' are you using?
https://www.google.com/search?as_q=railed&as_sitesearch=openbci.com
re: wifi. Best recommended practices is to use TWO AA battery packs, one each for Cyton and shield. And to use the UDPx3 mode.
Regards, William
The cap was part of this bundle I bought: https://shop.openbci.com/collections/frontpage/products/openbci-eeg-electrocap-kit
That was also the thread I found by searching for "RAILED": https://openbci.com/forum/index.php?p=/discussion/1481/how-to-become-de-railed
and that's why I tried to connect Ref + Bias to the cap (without wearing it, do I need to wear the cap to de-RAIL it, because "...your electrodes on the scalp" ? )
I'm not sure what that means: "Check board function: Connect SRB, BIAS, and railed Channel using jumpers / y-cables / ... This should give you roughly 0.00 uV. See picture (1)." should I connect the railed channel to the cap (while wearing it?) or somehow bridge the railed channel directly to the Ref or Bias (possibly using some resistor?).
It would be nice if a de-RAILing guide was part of the Documentation, it's mysterious for someone who just starts with this all (and I don't dare to try random things as I don't want to break the boards).
re: wifi with two AA packs:
Is this the correct order/way :
1. set the "ext pwr" to off
2. power on the cyton board
3. power on the wifi shield
?
(Haven't really found anything in the documentation).
Yes, electrodes must be connected to scalp for valid readings.