November 22, 2024, 11:02:11 PM

EMG-Shields,problems with channel 5+6

Started by Lyra, July 08, 2016, 11:40:04 AM

Previous topic - Next topic

Lyra

Hi,
i have the olimexino 328+ 6 emg shields. I use the demo sketch and as display software brainbay. I try to get a 6 channel setup, but it doesn't work for channel 5 and 6. The first 4 channels work, , but the 2 last have a big offset. On the the picture I don't have electrodes connected to the shields, but you can see, that the last 2 are out of range. I then tried to read out only channel 5 and 6 with a simple read-sketch for the arduino plotter,  here as well they are all time at 1000 and when i connect electrodes, the signal is cut, because it comes out of  range. I changed order of shields, but always the last 2 don't work.
Can you tell me, what could be the problem? Could it be, that the olimexino 328-board is damaged?


Then i have another question, i decided to switch the 16 MHz quartz crystal against 14,7 MHz to get a baudrate of 460 800 bits/s, because i want a sample rate of 2000 Hz. (2k* 10*17 Byte=340k, next baud =460800) I made a new bootloader and flashed it.
Now when i try to load a new sketch, it says upload and unfortunately the uppload never ends.

It would be nice, if anybody has an idea!Thanks

LubOlimex

About the 6 shields and OLIMEXINO-328 - how are the jumpers set? Try to close one REF_E jumper and leave the rest of the shields with REF_E jumper open and test again. Did you also cut the wire between the R6 pads (R6 is not mounted, but there is a wire between the pads by default)?

About the quartz - make sure the crystal has the same load capacitance as the ones we used (20pF). Also you need to change the board entry in the Arduino configuration boards.txt. Refer to this forum topic: http://forum.arduino.cc/index.php?topic=62631.0

Best regards,
Lub/OLIMEX
Technical support and documentation manager at Olimex

Lyra

Yes i had the jumpers, as you described it and I have cut R6 pads. Today I've got another olimexino board, but with the new one there is the same mistake with the 2 last channels, so at least I can say it is not a problem with the first board.

LubOlimex

Hey,

I think I found why this appears - most probably this behavior appears since A5 and A6 are also used for SDA and SCL and are pulled up - that is why they always stay up. Which analog lines you are using for the data of these two shields in your code? It seems logical to use A5 and A6 but when you use OLIMEXINO-328 these are also used for I2C (SDA and SCL) - and they are also connected to 3.3V via 4.7K resistors (R24, R23). If you have revision C of OLIMEXINO-328 there are jumpers provided and they seem disconnected but there is a board wire between the pads of the jumpers - make sure to cut between the pads with a sharp object to disconnect the wire. The SMT jumpers are named R24_E and R23_E. Alternatively, you can directly remove the resistors on SDA and SCL lines (R24, R23).

Note that the above procedures would ruin the hardware I2C of the board on the UEXT port.

Best regards,
Lub/OLIMEX
Technical support and documentation manager at Olimex

Lyra

Hi,thank you very much for the help, now it works pretty well. And I also solved the problem with the quartz crystal, so everything is fine.

LubOlimex

Hey,

Glad you got it going! If it is not much of a trouble for you and if you have the time it would be great if you can publish some instructions about you setup and the software that you used. I'm sure your experience would help a lot of people. I haven't had time to test 6-channel operation myself and I'm also curious on what software did you use!

Best regards,
Lub/OLIMEX
Technical support and documentation manager at Olimex

Lyra

I used Brainbay for the first tests. Now I am trying to display the signals with matlab for further evaluation, but i am still working on code.

Lyra

I have another problem. I want to remove the low pass( fc=40 Hz), but without losing the gain. I've read here in forum,that it is possible to remove r14 and r18, but then i will lose the last amplifier stage. Is there another possibility like just removing capacitors or something else?

LubOlimex

Hey Lyra,

If you want to remove the low pass filter but at the same time use the last operational amplifier you can try the following:

1. Remove capacitors C13 and C14
2. Replace resistors R14 and R15 with 0R (alternatively remove R14 and R15 and solder the pads of each resistor together, e.g remove resistors and restore the line)

I haven't tested it but this should accomplish what you want.

Best regards,
Lub/OLIMEX
Technical support and documentation manager at Olimex