Multimeter tests of electrodes ?
algmur
UK
Hi everyone,
I bought the UltraCortex IV headset about two years ago and upgraded to the ThinkPulse electrodes about 8 months ago because I didn't really get anything working with the standard electrodes. A lot of this is my own fumbling into an area I'm interested in, but in which I still have much to learn. I have seen wjcroft often suggest using an ohmmeter / multimeter to test the connectivity of the electrodes. I recently bought a digital multimeter for this purpose, but I am a little stuck on what exactly I need to do in order to test for effective functioning of the electrodes. Can anyone provide a full-steps (nothing skipped) set of instructions.
For example, I connect the reference electrodes on the headset to my ears and turn on the board. Then I unplug one of the white cables from the daisy module and follow the path to the corresponding electrode. I then turn on the multimeter to continuity mode and then point one end (does it matter which? red or black?) to the exposed pin and I connect the other lead to the ThinkPulse electrode and nothing happens. I have the blue light of the board showing it is on, but nothing else happens.
Would you expect moving one of these leads around a ThinkPulse electrode should cause the expecting beeping to show the connectivity of the circuit? I feel like I'm doing something wrong.
Any advice much appreciated.
Alex
Comments
It's super important to drop the "Gain" on every channel when using the ThinkPulse active electrodes when they are fully connected.
In the near future, we are looking to add an option to the GUI control panel that will set these values by default:
https://github.com/OpenBCI/OpenBCI_GUI/issues/999
Hi retiutut,
Yes, I do this by resetting the gain commands to the board directly in BrainFlow (or if I'm using the GUI, I have a saved settings file that loads them all to the correct version). However, for the multimeter tests, I am not connected to the computer. Do I need to be? Thanks for the heads-up on that point though, in case other users come here and wonder about it.
You cannot do 'continuity' (zero ohms) checks with active electrodes. Because they have internal electronics. Ohm meter checks only work on passive electrodes.
Can you summarize what signal distortions you are seeing? In past threads there seems to be a lot of mains interference going on.
https://openbci.com/forum/index.php?p=/profile/algmur
Regards, William
https://openbci.com/forum/index.php?p=/discussion/3061/wrong-notch-parameters-in-gui-resolved#latest
Hi wjcroft,
Thanks for your reply. It's a shame about not being able to do the tests. I was just a bit sick of trying things and never being fully confident that my equipment was correct. I liked the idea of going back to basics and fundamentally checking every logical step up the chain, starting with whether the current was correctly passing through the loop and then I would iteratively add in more and more steps closer to a full data recording session. I had seen the idea of the multimeter as a good step to verify the basics.
I haven't got any specific issue at the moment, i.e. I'm not actively trying anything and having problems. I want to start playing around in the next few weeks. Today, I just had a bit of time and thought I would test the fundamentals of all the electrodes are all sound, so I can at least have confidence at the basic level that the equipment isn't faulty.
Is there another sanity check you can recommend, that is compatible with active electrodes, that helps me to verify the functionality of the equipment?
Are you using exclusively Thinkpulse on all 16 channels? That seems like a huge number of actives. If you have some passive electrode channels, you can do the continuity check there.
Generally it has been acknowledged that any type of 'dry' electrodes (the original Ultracortex passives) or the Thinkpulse, will have some slight amount of signal amplitude loss, compared to equivalent 'wet' electrodes. In many applications it is not enough to matter. But as comments on the Forum have shown, certain EEG / BCI paradigms, such as SSVEP, cVEP, P300, will benefit from wet electrodes, because the signal to noise ratio can use all the help it can get. For example with the MindAffect cVEP system, all their examples / tutorials use a simple water or saline based passive wet system. They found that dry passive was just not good enough for the complex cVEP waveforms.
https://mindaffect-bci.readthedocs.io/en/latest/fitting_guide.html
If you say: "... I haven't got any specific issue at the moment, i.e. I'm not actively trying anything and having problems...", then I'm not clear what you are worrying about. Wait until you have an actual issue before you need to take remediation steps.
Another test you might try with neighboring Thinkpulse electrodes, is to moisten the skin underneath one of them with a small syringe or dropper. Then compare that channel with the other nearby channel without the moisture. Theoretically the active electrode and conductive rubber should not need any moisture. But adding a bit puts it closer to the 'wet' electrode situation. You could also do the same test comparing neighboring Thinkpulse and passive electrodes, with or without some moisture.
Regards, William
There IS a Thinkpulse Q&A thread here on the Forum, that the Thinkpulse engineer is monitoring. If you want to pursue checkout of any Thinkpulse issues, you can post there, and Julien will see it. If you use the at-sign notation, he will also get a 2nd email notification.
https://openbci.com/forum/index.php?p=/discussion/2869/thinkpulse-active-electrodes-q-a-resolved
Hi William,
Thanks for the info. You're right that perhaps I should just try to get something working first and try to take remediation steps later. I had seen that link before, because I actually bought the velcro straps mentioned there, but forgot all about them until now. I will set that up and start trying to record again. In my normal EEG work at uni, I've got clear tests in my wet system I can use to determine signal quality, but doing it on my own head without much electronic shielding in a system whose noise I don't know is within acceptable limits or not - is quite another thing.