Gumstix VIC Driver

[Login to see usernames]

I'm kicking around an idea of building an interface for my Gumstix Overo that will allow me to mimic the functionality of the VIC circuit, but also provide realtime feedback using the pick up coil.  I'd bet there could be a finer control over achieving resonance by programatically comparing the feedback to the output pulse.

Items needed:

Step 1.  Gumstix Overo computer with appropriate module with PWM outputs.  (I presently have Chestnut board).  www.gumstix.com
Step 2.  PWM kernel driver for Overo.  One exists, but will require customization.
Step 3.  ADC for return signal from cell connected to SPI bus of Overo
Step 4.  Optocoupler to connect PWM of Overo to driver circuit (to protect the Overo)
Step 5.  A well written program or script to manage the PWM logic with the feedback on the SPI
Step 6.  A linux program to simulate the logic of the VIC PWM circuit.

Please keep in mind, this is presently only in a planning phase.  I cannot say with any certainty that/when I will complete this or if it will even work.  If anyone here is adept at any linux driver programming or circuit design, please speak up as we work along with this as although I have an rudimentary understanding of both, my limited time prevents me from having proficient skills in either area.

The next several posts will detail each step as much as possible and will be updated with additional detail as information is put together.

TS

[Login to see usernames]

Item 1 - Gumstix Computer

I have chosen the gumstix computer as my platform simply because it seems like one of the easier platforms to use.  Its modules contain headers with built in PWM's and an SPI input to which a ADC (analog to digital converter) can return feedback from the VIC for resonance detection.  There are some PWM drivers already written which with some modification could be very suitable for driving the VIC coils.

The Overo EVM pack essentially contains the components I have chosen to use.
https://www.gumstix.com/store/product_info.php?products_id=253

The Overo Summit pack may also do the trick if you are looking for a less expensive model and don't want/need a video display.
https://www.gumstix.com/store/product_info.php?products_id=220

Both of the above have 40 pin headers.  There are other gumstix computers and expansion boards which may also do the trick.  Check out the www.gumstix.com site.  Other small computers that may be usuable are Raspberry Pi's.  But I don't have experience with them.

This product contains the following items:

»USB Cable (mini-B to standard-A)
»5V US Power Adapter
»ethernet cable
»USB Cable (mini-B to mini-A)
»Overo® Fire COM
»Chestnut43
»LG 4.3 inch LCD Touch Screen
»8GB MicroSD Memory Card

[Login to see usernames]

Item 2 - Modified PWM Driver

An open source PWM driver for the Overo exists, written by Scott Ellis.
https://github.com/scottellis/omap3-pwm

A little detail with others experience with the Overo PWM drivers.
http://gallinazo.flightgear.org/technology/gumstix-overo-rc-servos-and-pwm-signal-generation/
http://www.jumpnowtek.com/index.php?option=com_content&view=article&id=56&Itemid=63

Presently, to my knowledge, the driver sets its frequency when the module is loaded.  It needs to be modified so that the frequency is variable after the module is loaded.  Code also needs to be written to gate the frequency with a variable duty cycle.

[Login to see usernames]

Item 3 - Analog to Digital Converter (ADC)

http://www.digikey.ca/product-detail/en/MCP3301-CI%2FP/MCP3301-CI%2FP-ND/440171

[Login to see usernames]

Item 4 - Digital Isolator to Protect the Overo Computer

http://en.wikipedia.org/wiki/Opto-isolator

[Login to see usernames]

Item 5 - Code to Run PWM/Feedback

[Login to see usernames]

Item 6 - Simulator Program (PWM with False Outputs/Inputs)

[Login to see usernames]

Hi Timeshell,

It took me a while to get through all the information you provided but as far as I can see the approach you have might work . I studied the Gumstix and Rasberry PI info you gave. Very interesting! I think that if we want to crack the Stan Meyer secrets we can do it with this setup. It will require some time to hook everything up but it can be done. The programming seems also feasible to me. I drew a little (amateuristic scheme in attachment  ) on how I think the setup of the components can be. Please check it out and give me your comments. I was just wondering would it make sense to link the pulse pickup coil to another interface then the optocoupler? I was thinking of a USB Oscilloscope? Linking the Pulse pick up coile with a USB Oscilloscope to the Gumstix might enable us to tune into the frequency off water more easily. In a next step I think we need to connect a watercell and measure the amounts of gas produced.

I will have to learn the programming of the gumstix but it is feasible for me.

So let's gert started what do you think?

Warerfuel