Jump to content

Making new Crystal Sync electronics for CP16R


aapo lettinen
 Share

Recommended Posts

Slightly off-topic, would there be difficulty in designing and building a controller for a single speed for 24/25FPS geared cameras as stable and accurate as the original system which could be user installable or handed off to a local camera tech? Having the faster and slower frame rates would be sweet. Many owner-operators are probably only interested in the sync frame rates of 24 and 25 and there are plenty of affordable Bolexes and other types which can do the novel frame rates. 

Link to comment
Share on other sites

1 hour ago, Robert Hart said:

Slightly off-topic, would there be difficulty in designing and building a controller for a single speed for 24/25FPS geared cameras as stable and accurate as the original system which could be user installable or handed off to a local camera tech? Having the faster and slower frame rates would be sweet. Many owner-operators are probably only interested in the sync frame rates of 24 and 25 and there are plenty of affordable Bolexes and other types which can do the novel frame rates. 

there is not much difference in complexity between the 1 speed system and a 12 speed system other than the 12 speed system needs some kind of user interface to select the different speeds and the selectors and the back panel design often need much more time and resources to complete. So one would save on the graphics design costs and some of the most expensive components are not needed (the good quality rotary switches tend to be expensive) .

One would still want the film end warning even if having a single speed system. Additionally the circuit board for a system like this always needs to be custom designed which raises the costs (I have some designs which could be used almost as is but the circuit boards need to be of specific shape and size with accurate mounting holes to fit easily into the camera) .

It is fully possible to make this kind of single speed system if you would order at least three pieces. The price per unit would depend on how extensive features it needs to have and how the back panel is designed and made (it is easiest to just replace the back panel with a new one because one needs to scrap the voltage meter and the original speed selector anyway and then it makes no sense to save the switch either so just much easier to make a completely new panel instead) but I would assume something around 400 to 450 usd a piece including some kind of OK quality back panel and the electronics kit and if using the original tachometer assembly so that new sensor assembly does not need to be made

Link to comment
Share on other sites

Any system can be installed by a skilled enough person but at some point it needs too much training to install a system and it does not make sense anymore to let the user or another technician to install it (it costs more to make the training materials and provide the technical support than to just send the camera to the other side of the world for modification) . This is why I wanted to make the CP16R system in a external box so that the installation steps left to the end user would be as simple as possible.

I could maybe downsize a single speed design enough that I could fit most of the 1 speed controller to the backside of the back panel which would ease the installation a lot. One would still need to use lots of time to install it but it should be easy enough for anybody able to solder and use a multimeter. If you are interested in ordering a single speed system like this (at least 3 pieces needs to be ordered for it making sense to design it) I can investigate the options in September

Link to comment
Share on other sites

Aapo.

You can put me down for a single-speed. It will be for a camera which I have gifted to some younger filmies which is in otherwise good condition. It is the second camera with the dummy sound head in the CP16R "how-to" videos I made some years ago.

I gather you may require a deposit. If you do and advice via this forum does not contravene the forum's guidelines, please mention it here. Otherwise send me a private mail on this forum. 

The end of film warning feature would not be needed. If shouldering, one can hear the film run off. If using on a tripod, one can train one'self to watch the outspooling pulley on the magazine, someting I have done for some years.

Edited by Robert Hart
added text
Link to comment
Share on other sites

2 hours ago, Robert Hart said:

Aapo.

You can put me down for a single-speed. It will be for a camera which I have gifted to some younger filmies which is in otherwise good condition. It is the second camera with the dummy sound head in the CP16R "how-to" videos I made some years ago.

I gather you may require a deposit. If you do and advice via this forum does not contravene the forum's guidelines, please mention it here. Otherwise send me a private mail on this forum. 

The end of film warning feature would not be needed. If shouldering, one can hear the film run off. If using on a tripod, one can train one'self to watch the outspooling pulley on the magazine, someting I have done for some years.

great, I will look what I can do 🙂  I will try to figure out if shutter parking can be integrated because it is annoying if one needs to open the camera door after takes to inch the camera manually. Shutter parking needs a separate sensor which needs to be fabricated (the original one is integrated to the main circuit board and thus cannot be used) so I will need to do some testing if it is possible to include to the system.

 

-------------

Generally speaking I am currently working by first doing the plans without anything is charged yet, but when I have to order components for actually building the system and need to get the circuit board bases manufactured etc. which costs me a lot, then I will charge for the parts and materials costs before proceeding further. Usually I already have the software ready at that point so the system should ship in couple of weeks or max a month after the parts costs are charged (and of course the rest of the price is charged only when the system is actually ready for shipping).

Brushless motor systems are an exception because it may take a very long time to find a suitable bldc driver which works correctly with the selected motor (especially in current times when some special parts like such bldc drivers may have a lead time of up to a year or even more) and the parts can be really expensive so with brushless systems I may need to charge for the parts beforehand and then it takes from half a year to a year to complete the system. With brushless systems I need to first get my hands on the actual driver and motor to be used before being able to write any software and this is why it takes so long to get a brushless system made. (it is also why I prefer brushed systems if they are possible to use with a camera... so much faster and easier to finish the project with a brushed motor)

But the CP16R is a traditional brushed motor system so it is much more straightforward to make a controller for it. I can do some plans and will DM you when having a draft of what kind of 1-speed system can be made, how much it costs and if a parts costs deposit is needed

Link to comment
Share on other sites

Thanks for that. Shutter parking whilst convenient for eliminating flash frames and enabling the reflex viewfinder could be eliminated if the added circuitry causes difficulties as I expect it will. A replacement opto-interrupter would be needed for the small shutter disk which is towards the front of the camera. There is a small LED and detector built onto the existing circuit boards which sandwiches the rotating small shutter much like the interrupter disk on the motor itself. 

Some video cameras with infrared detection or a night vision can see the LEDs glowing.

For setups where focus positions and marks are critical prior to rolling, the mirror could be reset to the viewfinder finder's optical path by simply removing the lens and moving the shutter manually via the open mount. Otherwise buttoning on and focusing during the slate would be fine with little or no wastage of film. The older CP16A I have just starts and stops. There is no shutter parking feature. 

Slightly offtopic, in a recent message exchange with Ken Hale, he mentioned that he has wondered why no one has examined designing daughter circuit boards to patch in and replace the hybrid ICs.  

Edited by Robert Hart
added text
Link to comment
Share on other sites

57 minutes ago, Robert Hart said:

Thanks for that. Shutter parking whilst convenient for eliminating flash frames and enabling the reflex viewfinder could be eliminated if the added circuitry causes difficulties as I expect it will. A replacement opto-interrupter would be needed for the small shutter disk which is towards the front of the camera. There is a small LED and detector built onto the existing circuit boards which sandwiches the rotating small shutter much like the interrupter disk on the motor itself. 

Some video cameras with infrared detection or a night vision can see the LEDs glowing.

For setups where focus positions and marks are critical prior to rolling, the mirror could be reset to the viewfinder finder's optical path by simply removing the lens and moving the shutter manually via the open mount. Otherwise buttoning on and focusing during the slate would be fine with little or no wastage of film. The older CP16A I have just starts and stops. There is no shutter parking feature. 

Slightly offtopic, in a recent message exchange with Ken Hale, he mentioned that he has wondered why no one has examined designing daughter circuit boards to patch in and replace the hybrid ICs.  

my newer crystal sync firmware tries to stop the shutter at approximately the same position than it was when the camera was started but it may not be 100% accurate so one would still need to manually reset the shutter every now and then. It calculates the shutter's rotation angle all the time during the take so it should know when it is close enough to 0° to stop the motor (modern microcontroller based electronics can actually do this kind of calculation on the fly knowing all the time the exact angle the shutter is positioned even if it rotates dozens of revolutions per second. The old 70's electronics had no chance of doing anything similar) . I will test how well it works with the cp16r but can't promise anything yet because it is a bit camera model and motor dependent 🙂    

A separate shutter position sensor is 100% accurate but it needs building a support bracket to hold the sensor in correct place and that would take time and add to the costs so it is a less ideal solution if the system needs to be cost effective. A normal slot type infrared sensor would likely work perfectly for the application but the custom support bracket is the actual issue rather than the sensor itself...   I will test how well my sensorless automatic shutter angle calculation works on the CP16R and can then determine if the separate shutter position sensor is needed or not 🙂

 

Making replacement circuit boards for the original hybrid ic's would be possible as long as one know exactly what is inside the original ones but the replacements would be larger than the original hybrid ic's and thus it would be pretty difficult to fit them to the old system. The manuals have detailed schematics of at least some of the old designs including the insides of the hybrid ic's too. It is just that working with separately packaged ic's takes lots more space (the whole idea of the hybrid ic's was to get rid of the outer packages of the ic's and only connect the actual silicon chips directly together using wires, thus not wasting space for the plastic/ceramic outer casings the normal ic's would have)

Link to comment
Share on other sites

I will make a really simple 1-speed system for the CP16R soon which will be the first complete design for this camera. It is user installable and shipping probably in September 2022.  It will cost 400usd per kit + shipping costs (or slightly cheaper is ordering more at the same time).

It is a modular system and the basic kit does not include footage counter (can be ordered separately for about 150usd + shipping and installed to the camera afterwards if needed) . It is possible to use either internal crystal sync (single speed of either 24fps or 25fps, set by the user when installing the system and possible to change if opening the camera) or external speed source (for example a variable speed controller or a more advanced multi-speed crystal sync reference. These would be dedicated controllers sold by me, costing typically from 150 to 300usd depending on the features needed).

Please let me know in August 2022 how many persons are interested in ordering the basic 1-speed installation kit (400usd + shipping) so that I can determine the batch size and can build all the kits at the same time.

Link to comment
Share on other sites

I have lots of pressure to raise the price of the CP16R controller because of parts availability and thus I was thinking of replacing the 12-speed rotary switch controller with the display based 16-speed graphical user interface controller, almost exactly similar than the Eclair ACL design I am finishing at the moment.

It is possible to redesign the graphical user interface design so that it can be installed inside the camera by the user. A system like this would cost 1100usd + shipping and I would need to have at least two orders + a deposit of about 30% of the purchase price to be able to redesign the system. The controllers would be built in November 2022 and shipped in December 2022 with probably only single batch made this year ( I will be extremely busy in Spring2023 and can't build any controllers then so there may not be another batch until 2024) .

How many of you would be interested in ordering the graphical user interface based 16-speed internally installed CP16R controller?

The features will be approximately the same than on the Eclair ACL controller I am finishing at the moment though the top and low speeds will be different https://cinematography.com/index.php?/forums/topic/93357-new-alcs-crystal-sync-motor-for-eclair-acl/

52204326077_5c5f45202e_h.jpg

  • Upvote 1
Link to comment
Share on other sites

In other words, there would be the 1-speed low cost model of the CP16R controller costing about 400usd, and then the "high end" 16-speed model costing about 1100usd. Both could be installed inside the camera by the user.

The "mid range" external box 12-speed rotary switch model which would had costed about 750 or 800usd would be discontinued. (It is common that the mid range products take the hardest hit from economical issues like corona crisis and the Russian war so this was a fully expected end result)

I will finish the 1-speed model in any case because two of those kits are already ordered. Probably will only make a single batch of those too so let me know ASAP if wanting to order the 1-speed kit.

The fate of the "more advanced cp16r controller" will depend on if there will be enough orders to complete the design (as I mentioned in the previous post, I won't be able to build controllers in Spring 2023 so everything ordered needs to be built in 2022 and the next batch will probably be in late Autumn 2023 earliest)

Link to comment
Share on other sites

5 hours ago, aapo lettinen said:

The fate of the "more advanced cp16r controller" will depend on if there will be enough orders to complete the design (as I mentioned in the previous post, I won't be able to build controllers in Spring 2023 so everything ordered needs to be built in 2022 and the next batch will probably be in late Autumn 2023 earliest)

 

the planned speed presets of the 16-speed controller. Remember that the two User Presets can be set to anything within the range the camera can handle so only the most used speeds need a separate preset (adding too many presets into the menu makes the camera slow to use so the 16 presets is about the maximum practical amount) :

 

User1   (can be set by the user to anything within the range the camera can handle in 1/1000fps increments)

User2  (can be set by the user to anything within the range the camera can handle in 1/1000fps increments)

3.00 fps or 4.00fps  (depending on the lowest accurate sync speed the camera can handle)

6.00 fps

12.00 fps

16.00 fps

20.00 fps

22.00 fps

23.976 fps

24.00 fps

25.00 fps

29.976 fps

33.333 fps

40.000 fps

44.000 fps      (the approximate top speed these cameras can handle without damages as far as I know)

Ext Speed  (enables external speed input. one can use a specially made external controller for additional speeds or speed ramps)

Link to comment
Share on other sites

Aapo.

Have you let Paul Hillman at Visual Products know about your latest intentions for the CP and Eclair controllers. You might also need to register your designs in the US to protect them from reverse-engineering by folk who might get them assembled on the cheap in China.

Link to comment
Share on other sites

1 hour ago, Robert Hart said:

Aapo.

Have you let Paul Hillman at Visual Products know about your latest intentions for the CP and Eclair controllers. You might also need to register your designs in the US to protect them from reverse-engineering by folk who might get them assembled on the cheap in China.

Will definitely send Paul some photos when having the finished designs 🙂

-------

People generally pay for the reliability, customization and support when purchasing a crystal sync controller. A counterfeit product could not provide any of those and a person purchasing them would be screwed in every stage of installing and using the counterfeit.

Additionally the crystal sync controllers are very low volume products which are very difficult to sell in the first place whether being cheap or not... so it is highly unlikely that anyone would ever get their money back from spending time and effort for reverse engineering.

Personally I would want to sell couple of pieces of a design to get my own money back, and then sometime later make a updated version of it and start to sell the new one instead. If anyone would counterfeit stuff they would need to purchase at least one of each original version whenever one comes out and try to copy it immediately to keep the counterfeits relatively up to date, but maybe selling none in the end and just losing money in the counterfeiting process (of which money some already came to me as they needed to purchase one "real" item to try to copy from. So I am getting money anyway but they may get none in the end so it is extremely risky for them and would only work as a hobby, not a business) .

 

There isn't that much CP16R's in circulation anyway and the ones needing a update will quickly be spent. I think I realistically could maybe sell max. 10 of these CP16R 16-speed boards in the next 2 years if everything goes perfectly, after which mostly spares can be sold at best (or maybe a single new board sold for every two years or so).

Reverse engineering a complicated product like this if the worldwide market is approx. 10 boards total does not make much sense to me though it is still nice to have the designs copyrighted in any case 🙂

  • Like 1
Link to comment
Share on other sites

Hi Aapo;

I've been following your cool CP-16 crystal sync project with great interest, and I wanted to share my own camera controller project.

I'm kind of working at the other end of the scale, though. I got a project rebuilding the electronics for one of the old Fries VistaVision conversions. This thing is monstrous, It weighs in at 40 pounds without lens, mag, film or even finder.

image.jpeg.a67b4d43f1e67b7b69323f8849e9a4d7.jpeg

  • Like 3
Link to comment
Share on other sites

2/3

A few years ago it coughed and sputtered out the magic smoke and decided it would run no more. It sat on a shelf for several years till the owner asked me if I could return it to operation.

After about 30 minutes of tracing my way through the original circuits I realized that it wasn't going to make any sense trying to repair the mid-80's electronics. Fries had closed years ago and there was no information at all available, and even if there were, the design was built around old analog circuits and it didn't seem like a good basis for a camera controller in 2022.

I figured simply rebuilding it from the bottom up was the best way forward.

I designed two new boards for the camera. The first was the main control board, this included the user interface, external interface, and an LCD display.

image.thumb.jpeg.b7ba0e1d6b848db0be956238ed07be3d.jpeg

 

Oddly, the new boards kind of look older than the originals. The original boards used surface-mount chips but the new boards use through-hole chips so that anything programmed or exposed to an external interface could be socketed for easy replacement.

Truth be told, I would have liked to use a better display. This old-school yellow/green LCD kinda screams “1990's vending machine” to me, but with supply chains being what they are, and this being a one-off, it didn't make sense spending the the time to do anything more complicated, especially since our user base only cares about the data, not the look.

The electronics are otherwise straightforward. At the center is a PIC24EP family chip at 70MHz that coordinates everything. This chip has a built-in quadrature encoder interface and three PWM generators, so it makes motor control easy - once you decode the truly byzantine Microchip datasheet for these modules.

There are 3 motor drivers, one each for the camera motor, takeup hub and supply hub (for backwinding). The motor drives are all unidirectional (the camera motor has a hardware reverse switch) so the 3 drivers are all 1-quadrant, simply a big MOSFET driving the low leg. The switching frequency is 20KHz to make it inaudible, which I soon realized was pointless since the camera sounds like a washing machine when it runs - it was obviously not designed for sync-sound work.

The speed is variable from 1.000 to 36.000 FPS in .001fps intervals. The control system is a standard PID loop updated at 1KHz. Between the fact that a camera motor is commanded to run a constant speed, and the large rotating mass involved, the derivative term is basically useless, so after a little testing I just dropped it.

  • Like 1
Link to comment
Share on other sites

3/3

The final user controls closely follow the pattern of those on the Arriflex 435 - I figured that any professional operator would be familiar with that.

image.jpeg.3f93ac9d88024003af42071501062eac.jpeg

One thing that I felt was a big improvement was that I replaced the original 25-hole speed sensor disk with a modern 800 count quadrature encoder. Having 32 times more resolution dramatically improves speed control, but more importantly it allowed me to derive the signals I needed to update the accessory port from the original Fries type to an industry-standard 11-pin Arri port.

This camera will be used in complicated VFX settings, so it has to interface with all the equipment that modern cameras are expected to work with.

  • Like 2
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...