I have been searching all over for this project. Wishing you the best! 

I would like to do this with my old 35mm. I always wanted to do this on a Nikon F3 w/ the DA2 Sport Finder (just because the viewfinder looks like a screen), and only dreamed about having having a house big enough to mortgage for a MF back, so I will just follow this instead :)

sensors are coming down in price.  it's not just the # of megapixels, it's the photodiode element size in μm, global vs. rolling shutter, RGB response curves.  FPGAs powerful enough to handle the image processing are in the $20 single quantity neighborhood.  pseudo-SRAM organized 4Mx16 is $5.00 in single qtys.  the effort is to see what is possible with commodity hardware first, then either EPIC_FAIL = 1 or take the lessons learned and go big.

Very useful project... but how are you planning on making it medium format?  That sensor is tiny (~8mm wide).  Would you be ganging multiple ones together?

I have wanted to do similar in the past (but for 35mm cameras), but looking around there is no economical source for small quantities of large sensors (and by large I don't even mean medium format, even APS-C is essentially impossible to find, let alone 35mm).

I'm still in the research phase - I just threw that (OmniVision
OV14825-A16A) sensor out there as an example of cheap (and because it
garners helpful feedback like yours).  Seeing as how they want an NDA
before giving you a datasheet...it is a non-stater, but the ON semi 14MP
(MT9F002I12-N4001-PF) is independent developer friendly but as you say,
hard to find (perhaps a group buy or dare I say it, kickstarter
campaign).  The initial thought was (as with every effort) to get
something cobbled together and working first, then worry about stitching
together a ganged array of sensors and working out the semi-complicated
algorithms and hardware.
Speaking of APS-C, my Canon version sits next to my Rollei...
Maybe the answer becomes to cannibalize and reverse engineer other equipment (like Fujifilm S9XXX 9MP cameras) to make something great - this project is to lay down the idea and start the work and conversations.

Ahh, ok.  I will be following your project with interest!

What I am interested in is taking an old 35mm camera and putting a sensor in it instead of film.  I don't care about a display, I don't care about fast shooting, etc; I just want a simple camera.

I know there have been commercial ventures trying to do this in the past, but they used small sensors (this was early on in the digital revolution, and 35mm sensors were very expensive then) and they never really got off the ground.

I realize that a MF back will be at the same time both easier and harder than a 35mm back: you will have more room to play with, but will need a larger sensor (and if I think that 35mm is hard to find, MF is even harder!)  If you are able to figure out the problem of sensor size, I think it would go a long way towards realizing this.

Anyway, feel free to chat if you want to bounce ideas off me or something.

Cheers

"I don't care about a display, I don't care about fast shooting, etc; I just want a simple camera."

That's what I'm aiming for, simple tech...allowing a person to recapture/retain their expenditure in quality photographic gear.  Put the money into a quality sensor and beefy FPGA and forget about the luxuries (dual ported RAM is off the table, and <= 10ns SRAM, etc.)

I'm going down the line to see who'll talk to me in the sensor domain - they (mfgrs, distributors) act like this is anti-gravity, zero point energy, or nuke warhead privileged info...really sick of it already.