Jump to content

Prototype CP-16 motor complete


Michael Collier

Recommended Posts

Introducing the CP-16 Archimedes.

2-64fps crystal sync CP upgrade.

 

 

So the prototype is done, Kodak has graciously sent us some film to evaluate the function (and to get some beauty shots with it), still trying to organize labs for the processing (I have spent all my money getting these boards made, not much left to pay processing and telecine.)

 

But the video of the prototype going through its various tests is online and can be viewed at:

 

Note that this is my down and dirty proto board version (and the video is quite down and dirty...I apologize the LCD is hard to read through the Youtube compression). The through hole CPU seen in this video will eventually be a more rugged version of the same chip in surface mount packaging, making the unit lighter and more durable.

 

Jan 4-Jan 15th is the official first film test date (although I have already run several hundred feet of gash film through the unit without problems.)

 

Takes original batteries, all original film light/sync light/battery light work, it has an EXT input to allow for an after market intervelometer and milliframe controller. The unit itself will fit in the space where the old back panel used to sit.

 

Now I just need a name....I have been leaning towards CP-16 Archimedes. Any suggestions?

 

(note there is a deal available still: the first 10 to sign up and pay a deposit on their boards will receive free installation (a $250 value) email me at cp16 (at) randomacronym (dot) com for more information, or to sign up.

 

The design is at the manufacturer and will be available for install sometime Q2 2010.

 

(if your a camera tech or deal with these cameras on a regular basis, email me. perhaps we can work together.)

Link to comment
Share on other sites

  • Premium Member
The through hole CPU seen in this video will eventually be a more rugged version of the same chip in surface mount packaging, making the unit lighter and more durable.

I strongly recommend against surface mount. Through hole with critical chips in milspec sockets is infinitely more serviceable and easy to diagnose.

 

I personally can handle surface mount diagnosis and repair but it takes much more experience and finesse to say nothing of specialized equpment to work on surface mount. My hot air solder pen, Haako desolder station, and Weller low power thermostatic soldering iron alone were well over $1500. You can work on through hole with a $25 iron and a roll of solder braid.

Link to comment
Share on other sites

Yeah, well its a classic give and take between servicability/cost/ruggedness. While a DIP is easier to service (especially in a socket) they are less durable than their surface mount counterparts (especially when talking about a 44pin package) I will be the only one to service these, and service will likely be more based on parts that could fail, namely the output amplifier, which is through hole and over the shelf. Since the chip contains proprietary software, there is no way a user or tech could service a failed one. It would have to be sent back to me for replacement. And lets be honest, the odds of a solder going bad on either package is very low, more commonly wires and connectors fail long before a chip soldered to a board will.

 

If there is a problem with the chip, I would likely just replace the entire board, it would be cheaper at that point, but the number of incidents would be greatly reduced. In my opinion the first problem experienced by a user should be 15-20 years down the line, not 2 years down line, even if the 2 year fail would be cheaper for me to repair. Also since my design leverages R&D in software, not hardware, a surface mount reduces fraud. Finally the size of the PDIP chip is huge compared to its surface mount counterpart. There is barely room in the camera to fit the surface mount. If I went for a PDIP, there would have to be two boards, which would increase the possibility of a defect.

 

Its all about making it rock solid. That's been my mantra from the beginning.

Link to comment
Share on other sites

Quick Update: after discussing the product with the manufacturers, I can announce a price for the Archimedes electronics:

 

$995 for the board, $250 for the install. The first 10 to sign up get free install.

 

 

Still working on a Q2 date for the first install.

 

Professionally built boards on double sided PCB boards with protective green solder mask. Install includes full functionality test, and a 12 month/12,000' warranty against manufacturing and install defects. Each board personally installed and tested by the inventor.

 

email me at CP16@randomacronym.com for more information or to join the mailing list to be notified when installs will be scheduled.

 

Happy New Year!

Link to comment
Share on other sites

  • 5 weeks later...
  • 4 weeks later...
It's been a month. Any news about the progress?

 

 

February for this project was mostly spent designing the copper traces and solder masks for the board, getting mechanical drawings made of the back plate, and generally getting protocols for install set. I have been looking over various schematics from the different incarnations of the CP to make sure the wiring is the same.

 

I have also been adding a few tweaks here and there to the software menu interface.

 

* Saves all settings and counters when they are modified, and loads them on boot up

* Boot time under 1/8 sec.

* Added an odometer ( 0-65,535 x 100' )

* Finished the battery alarm section, which now features:

* Real voltmeter, measuring battery power (5.0 - 36.0v +/- 0.1v)

* set the battery alarm threshold in volts

* back display now shows ( ##fps ###/400' ) or when you hit a button ( ##fps ##.#Vdc )

* Updates counters and battery LCD while running

 

Its looking pretty good, as far as I'm concerned. I am really happy with the Battery Meter. I haven't done much ADC work before, so I figured I would just cop out and have a bar graph. Once I got up to speed on the ADC system on my CPU, I found it easier to do a voltmeter than a bar graph. I am also happy I have kept everything to one controller. That simplicity ensures reliability. I have done extensive debugging as I developed the software, so it is ridiculously rugged.

 

I am also planning a shoot with the camera. Last week I spent a night doing a fitting with one of the models and my wonderful wardrobe guru. I think in a few weeks I might take the camera out to the Iditarod start and get a bit of footage from the chute.

 

I am working for a release date sometime around the end of April/Beginning of May.

 

I should hopefully follow this up with an intervalometer accessory, and I intend to make that very competitively priced. I can't really announce a price just yet, but it will be the one of the cheapest and best featured intervalometers on the market.

 

Next month will be putting it all together and testing overall system performance and stability. I will test the full range of speeds with gash film loaded, and move forward on getting stocks of LCDs, resistors, caps and all the other things that will populate the boards. Hopefully by the end of the month I should have the first boards in, with the controller CNC placed and soldered. More or less I'm still on my time line.

Link to comment
Share on other sites

Interesting times and a worthy effort.

 

 

The memory battery leak damaged the crystal sync for mine. Ken was able to put it right for me.

 

Do you still get a bit of hunt in the motor speed once the system is pulling film or does it settle?

 

 

CPs came with two interchangeable cog pulleys, one for 24fps, one for 25fps. Which of the pulleys is your system calibrated to? People with cameras geared for PAL TV will have 25fps pulleys. Some of those cams have ended up back in the US. Is your system capable of switching references to permit both gears? The motor spins at the same speed with the old governor board for 24fps or 25fps.

Link to comment
Share on other sites

Hi Robert,

 

Any crystal sync controller will have some sort of standard deviation. The board controls this to a very tight tolerance and well within OEM spec. The two factors in deviation is the degree its able to drift off absolute target, and the frame to frame variability. My system puts focus on keeping each frame exactly 1/24 sec, and allows for small tweaks 600 times a second to keep itself perfectly in beat.

 

Overall it never drifts enough induce visual artifacts in the film under any condition. Flicker and strobing are eliminated. If they get too far out of spec, the sync light will comes on. The timing is done with absolute time (like a stop watch taking split timings) so the chip always knows exactly where the target is, ensuring perfect lip-sync, even with a take spanning an entire 400' mag.

 

The Archimedes can be post-calibrated for either pulley. There is a jumper setting to select 24/25fps pulley. There is no true "reference" frequency generated on chip like other controllers, instead it has tables of intervals for every frame rate. When you change the jumper the chip will load the appropriate table (this also changes the calibration of the film counter and odometer to match the different ratio) If the spare is still available I would recommend using the 24fps pulley. That gives the camera exactly 600 feedback pulses per second @24fps, whereas its 576 with the other pulley. The faster the feedback frequency the more accurate the control will be, and the faster the LCD will be refreshed.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...