Ticking sound from the Cyton Board ? [resolved]
Hello everyone how are you doing?
I have this Cyton board I got from my college. I tried turning it on a few months ago and it worked perfectly. Then I put it in its box and left it as I was busy with exams. I tried powering it on again using the same battery pack I had since last time. When I put the switch on 'PC' or 'BLE' , a weird tunning soud could be heard from the board. I tried switching it to OFF and to PC multiple times till it worked normally like in the past. But then suddenly it made this ticking sound and stopped working all together. No tunning sound no blue LED indicator nothing. I tried buying new batteries but that still didn't work out.
So is the board dead? Are there any test I can make to make sure if its still working?
Sorry for my bad English and thankyou fro your time and help.
Comments
Amir, hi.
You could try re-flashing he Cyton mainboard firmware.
https://docs.openbci.com/docs/02Cyton/CytonProgram
One test you could try with the GUI, is to use the AUTOSCAN feature. If this works, it means the two RFduinos (dongle and mainboard), are alive and you have a functioning serial COM port. This would enable you to do the re-flash.
https://docs.openbci.com/docs/01GettingStarted/01-Boards/CytonGS#5-optional-settings
Use the arrow on section 5 to expand it.
William
Okay i left for somedays and it somehow works now. But not quietly unfortunately. Iam using the ultracortex IV with the dry comb electrodes. When iam reading data using the OpenBCIGUI, if I read from the headset without wearing it, it's placed on the table, the readings are fine and smooth, and it outputs noise as expected. However, when I wear it, the reading becomes laggy and sluggish and not real time. It speeds up and then slows again and sometimes it totally stops altogether, and there are also these weird impulses across all channels that happens by random. I tried playing with the setting of the GUI, to see if that can fix that but to no avail. And as soon as I remove the headset from my head it works totally normal again.
I tried using the wet electrodes and they work totally normal, be it when they are just placed on the table or placed on my head. Don't know why the dry electrodes makes this weird behaviour. Any suggestions on to how to fix this would be highley appreciated.
Thanl you fro your time and reply and sorry for my bad English.
When did you purchase the Ultracortex? If it belongs to your college, it could have possibly been used in many previous courses / projects. The dry electrodes last a long time, but it is not infinite.
The fact that wet electrodes work normally with your Cyton, points to an issue with the dry electrodes.
The Ultracortex is 1 year old and it was used for another project before ours, so it wasn't used extensively fro too long. But the thing I noticed is that even with dry electrodes, the device wasn't just behaving right. For example I couldn't generate Alpha waves by closing my eyes. I tried to change the bias electrode placement from the earlobe to the forehead and then to Oz but all gave me same results. The FFT response was going up and down in all frequencies but wasn't showing anything significant at 8-12 Hz. Don't know if I did something wrong in my test but I will test again and tell you my results.
As for the dry electrodes and using the headset, I tried changing the electrodes placement and it worked with no lag. But as soon as I put it in the configuration I want, which is electrodes all around the optical lobe, it starts lagging again when it touches my head, then becomes smooth again when I lift it, which is really odd behavor.
I will try making more test and playing with the electrodes placement, maybe I can finally generate the Alpha waves I want. But if there is a guide on the exact placement and what to expect that would be really appreciated.
Thank you again fro your reply and time.
The amount of eyes closed alpha generated is dependent on the individual being measured. Can vary quite a bit, person to person. Also correlated with your state of relaxation, and if coffee, caffeine, other supplements are involved.
Good locations for alpha are in the neighborhood of O1, O2, Pz. You may also find better results with the wet vs dry setup. EEG is measuring differential voltage between reference and channel. So the larger that distance is, the greater the differential.
re: lagging. I assume you have read the material on FTDI buffering settings. If set wrong, this will result in lags. But should not be dependent on electrode setup.
https://docs.openbci.com/docs/10Troubleshooting/FTDI_Fix_Windows
Regarding the eyes close part, I just got coffee a few hours ago so I think the effect of that is still present in my system, so maybe I will retry again tomorrow morning before I get my coffee.
Regarding the lag part, I have already adjusted that the first time I tried using the headset with brainflow on python, as the data was coming in bursts, so I read that adjusting the buffering settings would fix that. But it's to note that i am using windows 10. Would switching to Ubunto give me better results? Or it won't be noticable?
That same link I gave (in the left column), shows buffering adjustment settings for all three OSes: Windows, macOS, and Linux. So since you've already fixed it for Windows, no need to move to Ubuntu.
Okay then I will continue testing on windows and see if I can get better results. Thank you for your patience. Will keep this post updated with any new thing I try.
Stress level, and the amount of daily sleep you get, also have correlations to your level of relaxation and alpha production.
EEG meditation studies, have shown that those with a regular meditation / relaxation practice, are better at generating alpha. Although there are different meditation styles that generate different patterns.
https://www.google.com/search?q=eeg+correlates+of+meditation
Okay so new updates. I still didnt manage to get it to work, unfortunately. Wich made me question if maybe there was something wrong with the cyton board itself. I disconnected all the wires and left it only connect to the battery pack. I got these results:
https://drive.google.com/file/d/1XB5MD1t5Ea1GMyiX3J2xbzBnQwBGGYRb/view?usp=sharing
As you can see even in rest mode with no wires connected to it, the signal is very noisy. And there are these weird impulses that happen every so often and cause the entire FFT signal to go crazy. I'm conducting the test in a closed room with the only electrical device is my computer. Mobile phone close and light closed. So don't know what might be causing this noise, or if its even the reason why I am not getting the alpha waves. Any help would be appreciated.
You wrote on January 12th:
I'm not sure I understand. It works intermittently? Your video shows expected results from no wires attached: generally very low microvolts. UNTIL the burst of 50 Hz mains noise. That could have come from proximity of the board to a noise source, for example even your hand / body in a room can have substantial mains voltage from proximity to cords, conduits, lights, power supplies, etc.
The clue that it is mains related is in the FFT, 50 Hz peak.
Regards, William
Regarding the left it for somedays part, it didn't work out of the gate, I just left it on PC mode and it kept making this tunning sound. Then it powered up on its own. And since then it just works whenever I turn it on no tunning sound again.
The low level level noise is expected cause yeah I know that its hard to have a room with absolutely no interferences or noise sources. But the sudden impulse of high amplitude is strange as it doesn't happen with certain period it just happens randomly.
One thing I noticed is that whenever I touch the batteries on the battery pack, I can replicate this impulse, or if I just move the battery pack cord that impulse noise happens again. I have read that its possible to use Ion-Li rechargable batteries with the cyton board. Would that improve performance? And get rid of this random noise? Or could the problem be somewhere else?
Again thank you very much for your time and reply.
How does the board perform when following the tutorial suggestions?
https://docs.openbci.com/docs/01GettingStarted/01-Boards/CytonGS
https://docs.openbci.com/docs/01GettingStarted/02-Biosensing-Setups/EEGSetup
For valid EEG, you need all THREE of the connections: channel, reference and ground. Eliminating any of these, or tests with no wires, are not valid tests. All EEG amps are DIFFERENTIAL amplifiers. Require both channel and reference. Plus Ground / Bias. The differential aspect cancels out what is called "common mode" noise: noise present on both channel and reference leads.
The 'ticking' or pulsing noise you are hearing, could simply be coming from the RFduino Bluetooth radio module on your Cyton. The shielding case around the RFduino chip could be resonating with the radio pulses / power supply demands during packet transmission. Just ignore this, it should not affect operation. Even though it is somewhat unusual for this to happen to the radio module.
Any unusual behavior with the dry electrodes / wires, points in the direction of those electrodes or headset as the issue. The fact that wet / paste electrodes work just fine, means that your Cyton is working normally.
William
Well unfortunately, I am back at square zero again. The cyton refuses to power up. Have even tried to change the batteris but still not powering up at all. My last hope would be getting new Lithium Ion batteries and see if that will make it work. We are of campus now due to covid restrictions, and I don't have any method of testing if the kit is outputing VDD and GND from labeled sockets. Any suggestion what I might do next?
Thank you again for your time and reply.
Your initial post stated:
I assume you are using the 4 AA cell pack? With fresh batteries from the store?
Yes exactly, I am using that. I tested the batteries alone on other devices and they work. But I couldn't test them in the battery pack as I don't have any other device using the JST 2-Pin connector for power input.
You definitely need to try with brand new batteries. Devices have different requirements for minimal voltage. Replace ALL FOUR at once with new cells.
Okay I will try that and come back to you with the results. Thank for your time and reply.
I tied new batteries and still the same, no turning on. It doesn't even make the tunning sound it used to make. So I am kinda confused now if its even working at this stage.
Thank you again fro your reply.
Can you obtain a digital voltmeter from a friend, or visit a location that has one? This is to check that the battery voltage is reaching the Cyton, and that the voltage regulator chips are functional.
I managed to test the battery pack by connecting some wire and testing it using a LED, and it wasn't outputting any power. After further tinkering, I found some of the metal contacts in the battery pack were getting eroded, and maybe there was no full contact thus creating and open circuit and no power flowing. So I guess now I am kind of certain the fault is with the battery pack. So I will try to find a new battery pack with the same connector and see how that goes.
Thank you again for your help and time.
Great. If you find a 4 cell AA holder with red and black output wires, you can just CUT the original OpenBCI holder wires off (close to the holder as possible, so it is maximum length.) Then strip the wire ends about 1 cm to expose the inner copper wire. Do that with the old and new cables. Then twist red to red and black to black, to splice them. Cover ends with some protective insulating tape.
Okay so now I changed the batteries and it works again, the blue LED isn't blinking and its stable. But the thing I noticed about it there are still very large amount of noise present in the signal I record. And also what ever disturbance happens in one of the electrodes, it appears on other channels as well. For example if I move my finger across electrode connected to channel one, I should expect some sort on signal in channel one only. Instead I get disturbance in all channels, like if the signal recorded by one electrode is caught by other electrodes as well which of course creates huge noise and wrong data. Don't know if this is the expected signal I should get, or if I something is wrong with my kit, any help would be appreciated.
Thank you again for your replies.
Generally, noise that looks similar on all channels, is a sign that your reference electrode is involved. Since EEG is differential, and all channels share the reference as the other polarity.
In terms of moving fingers over electrodes: just tap very lightly (one at a time) as skin movements are widely distributed.
Have you done these tutorials to check your setup?
https://docs.openbci.com/docs/01GettingStarted/01-Boards/CytonGS
https://docs.openbci.com/docs/01GettingStarted/02-Biosensing-Setups/EEGSetup
Regarding the tutorials you mentioned, yes I setup the Cyton board the way its explained in the document. And I aslo tried to generate Alpha waves the same way that was explained in the EEG tutorial using same wet electrodes placement, but I get no alpha waves at all. But when I blink or clench my teeth I can see corresponding artifacts in both time domain and FFT graph, if I blink fast enough I can generate a beak in the FFT graph at the rate I blink with. However, I dont seem to be able to generate any EEG signal, as nor closing my eyes for alpha waves, or using SSVEP patterns seem to work at all.
As I mentioned previously, EEG alpha generation amplitude is highly individual. Suggest trying a test with a friend. Possibly someone you may know who is a meditator, or in general seems to be a 'relaxed' sort of person.
Its kind of hard to meet in these days cause of covid restrictions here. But I will try to see if that is possible in the coming few days. But beside alpha waves, aren't there any other methods that I can use to check if the heads properly read EEG signal from my brain?
SSVEP generation is very timing dependent on your monitor and how it is generated. In other words, if you are generating the flashes with a Python or other program looping on a timer, cpu / OS scheduling delays may cause jitter in that timing. Another approach is to generate a video file and play that back. Video output timing is much more precise.
https://www.google.com/search?as_q=ssvep+timing&as_sitesearch=openbci.com
There is a free cVEP based speller from MindAffect, quite a lot has been posted on it.
https://www.google.com/search?as_q=mindaffect&as_sitesearch=openbci.com
For SSVEP, I used the code provided by chip for the video file generation, as I know real time program won't be very accurate and tried a LED array, but was very uncomfortable. But then again it didn't generate any correct data which had peaks at the intended frequencies.
I will read about MindAffectBCI links you provided, and also I read there was a motor imagery demo available in OpenVibe, maybe I will have better luck with MI than I do with SSVEP. If that also fails then I will try with a friend and tell you the results.
Thank you a lot for your help, really appreciated.
Also one more thing, could this be a bad firmware problem? Like can reflashing the cyton board with the firmware make perform any better? Or that isn't effective in my case?