Re: [Openeeg-list] DRL / COM port / Wave forms

[Joerg H. <jo...@sd...>]

Hi,

mls@VegInfo.com wrote:
> Joerg Hansmann wrote:
> 
>>Now you should be able to receive openEEG data packets with
>>a 14Hz fullscale (0..1023) calibration signal on 
>>channel 1 with electric guru.
> 
> 
> Success (sort of) - Did "stand alone" testing of digital section
> and got attached waveform!

:-)

> Thanks to everybody who helped in
> this thread (especially Joerg) and did not give up. Now will
> add optical isolation

This would be the next logical step. Good luck!

> and then analog section. May be even
> get the whole thing working before getting PCBs arrive.
> 
> Some incidental things.
> 1) What is the artifact in middle (probably around 512 value)
>    on ElectricGuru waveform?

This results from the line interpolation algorithm that is used to connect 
adjacent samples.

> 2) Is the waveform supposed to come on both channels?

No. It should only be on channel 1.

What circuitry do you have connected to IC102 pin 24? Nothing? Then it could be 
capacitive coupling.

Try to connect IC102 pin 24 (ADC1) to GND. Now the 2nd channel should show 
constantly 0.

Try to connect IC102 pin 24 (ADC1) to +2V (J102 pin 2). Now the 2nd channel 
should show constantly 512 (more or less).


> 3) The perl program still was not able to read the COM port.
>    May be it was meant only for Linux. Will try it on a
>    Linux box or try Win32::SerialPort on Windows...
> 4) Is there a way to make "stand alone" testing of digital
>    board possible using the PCB?

Yes.
A single digital part openEEG pcb can be tested.

Next to check would be the analog reference voltage:
IC102 pin 21 should have 4.0V (WRT GND)

> Or this privilege is only
>    for folks who dabble with breadboards { others get the
>    whole thing working without all these hassles! :-) ]

Also pcbs can fail (e.g. bad vias) or components but I think breadboards are 
more susceptible for problems (e.g. wiring errors, bad wires, oxide on contact 
springs, high coupling C, no shielding layers, contact rails instead of star 
formations, etc.)


Regards,

	Joerg