Friday, June 30, 2017

Et Erit Lux - Let There be Light

I took one year of Latin in high school... no clue why. Et erit lux (translation - let there be light) and a few nouns are all I can remember. Oh, and semper ubi sub ubi which isn't even a proper Latin phrase but rather an inside joke to those that took time to learn the language. Tut tut.

"Let there be light" is a good title for this heap of Pola-geekery blog fodder, and we're gonna get down and dirty on this one with lots of cross-eyed tech talk, and a smattering of pre-school physics. If that wasn't enough, there's a bunch of obscure pics that will both confuse and astound you. It's the only way I know how to explain it.


Just about every Model 1 SX70 I get in the shop to fix up has some sort of overexposure problem. Images are faded, overexposed, highlights are heavily washed out, there's zero contrast, and some objects are barely visible in bright scenes. Some shots even have a slight blur to them. There's many variables that can affect exposure metering on the SX70 (sticky shutter blades, sluggish S1 solenoid, incorrect faceplate ND filter, LTO - long time out), but the most common is a crunchy-crusty photodiode filter.

Overexposed indoor/outdoor shots caused by a heavily blocked photodiode filter.

The SX-70 was designed to "see" ambient light by means of a photodiode that is built into the shutter's circuit board (ECM - electronic control module) which is mounted right behind the shutter frame. What exactly is this "the photodiode"? Well... simply put, a photodiode is simply an electronic component that takes light and converts it into electricity.

Symbol for photodiode. Ever see Stargate? Well... those symbols were constellations but this looks like one of em... whatever. Read on.

When the shutter button is pressed and the internal mirror flips up, the exposure sequence begins. The shutter blades open to reveal light to the photodiode through secondary aperture holes located right behind the faceplate's "electric eye" (which is nothing more than an ND filter held into place by a pretty chrome ring bezel). Light is converted into electricity that runs into a capacitor on the ECM. When that capacitor is full, it pukes out it's charge in one huge burst which is read by the shutter's IC "brain" and tells the shutter blades that it's time to close the doors, ending the exposure part of the cycle. The remainder of the cycle then mechanically spits out the exposed print.

Shutter assembly cutaway showing the shutter blades. Secondary aperture holes are shown to the right of the blades. Located behind these openings is the photodiode.

A good way to understand how the automatic light metering system of the SX70 works is this:

- The less light that hits the photodiode, the longer for the capacitor to charge and the shutter blades remain open for a longer amount of time.
- The more light that hits the photodiode, the faster for the capacitor to charge and the shutter blades remain open for a shorter amount of time.

Keeping the above comparison in mind, because of low levels of ambient light, a lot of indoor shots without a flash or tripod are blurry. The shutter blades remain wide open for a few seconds and any camera movement results in blur. I'm gonna pat myself on the back for that explanation... ready for that second cup of coffee and a valium?

So back to the crusty photodiode filter. The photodiode itself is protected by an opaque housing and only light can get in through a little glass window. This window, filtering out any type of unwanted and interfering light that may cause inaccurate readings, is a tiny green glass gel filter. Without this filter in place, the photodiode lets in tons of nasty light that can't even be seen, throws off the reading, and results in way too fast of a shutter speed. Shots then emerge very underexposed and dark.

A very clean photodiode filter on an SLR680 (note the conformal coating only surrounds the housing and traces leading to the capacitor).

The SLR690 also has a similar photodiode but the filter is built directly into the shutter frame.

As far as what the residue is that grows/leeches out of/accumulates on the filter... I haven't identified that yet. Anyone that has studied optics, chemistry, or even internal medicine, please feel free to jump in on this one. I'm thinking of sending a sample to a lab (a sweet cop-show crime lab) to see what it actually is, but it's most prominent on Model 1s that seem to have been stored in wet conditions. The moldier the camera, the crustier the filter. One of my theories is that it's a reaction between the adhesive used to bond the filter into the photodiode housing and moisture, much like the corrosion on a car battery. Or it could be a reaction to moisture and the glass itself? This corrosion is white, very opaque, and can drastically reduce the amount of light that reaches the photodiode. Remember what I wrote above about the less light that hits the photodiode? Yep... shutter stays open longer. Overexposure.

Crusty crunchy corrosion on a car battery... yuk.

Crusty crunchy corrosion on an Alpha 1 ECM photodiode filter which will cause overexposure... More yuk. You can tell it's unfavorable since the word "crap" is written by the area circled.

Wonderful! So what now? Easy!... remove the corrosion. Often times the corrosion is found on the outside of the filter and it's as simple as using a blade and then some al-kee-hol to remove the gunk. Later model ECMs with an epoxy conformal coating (Texas Instruments would dip the edges of their circuit boards in to a black epoxy) are usually easiest to clean possibly because the coating gave better protection from moisture. Except for extremely moldy cameras, most corrosion found on these models is only limited to the outside of the filter. Alpha 1s, Sonars, 680s, and some late model Model 1 circuit boards have this black epoxy coating. But Model 1s from 73-75 had a different urethane/lacquer type coating that didn't offer as much protection and often times the corrosion would occur on the inside of the filter as well. Early 72-73 Fairchild ECMs didn't have any coating at all.

White crusty deposits on the inside of the filter adds a whole additional level of "meter cleaning" and the photodiode housing needs to be rebuilt. Essentially the filter needs to be cut out, the gunk removed, the filter bonded back into place, and the housing resealed. It would be easy if the filter was nice and flexy but it's a fraction of a mm, thin, tiny piece of glass that if not removed properly, can crumble. Ask me how I know this... that little sucker is like an eggshell.

Nice and clean photodiode filters. The one on the right chipped during removal. Oops.

1. A blocked/corroded photodiode filter on a standard Model 1.
2. Simply cleaning the front of the filter reveals there is still blockage on the reverse side.
3. The filter is carefully cut out and fully cleaned.
4. The filter is bonded back into position and the housing edge is resealed.

Cleaning or rebuilding the photodiode filter allows for proper light readings and accurately exposed prints. There's ways to "kinda" fix it, like cranking the LD trim wheel all the way to dark or even removing the faceplate ND filter (which by the way was used to fine tune final exposure and varies from camera to camera) but these are simply band-aid methods that don't allow the camera to operate properly like it was designed and may work in one type of lighting but not in others. There are many additional variables like film chemistry that can affect results as well. One company has gone as far as designing and manufacturing their own ECMs to avoid restoring old technology. I'm straight up green with envy on that one but I don't even have a slight fraction of the capital they do to put toward R&D and manufacturing for an SX70 circuit board so I had to learn how to restore the old stuff... which I find more fun anyway.

Accurate light metering = proper exposure. These were taken with a '73 Model 1 with a rebuilt photodiode housing.

So there you have it regarding the SX70 photodiode/filter, but I have to end this screed with a little 2nd Shot plug - A primary part of my repair process involves checking/cleaning/rebuilding this filter on every model camera that comes into the shop. This falls within my mission statement that if the camera "kinda" works, it's not working properly. This camera relies on all sequences working 100% for optimal results. I like optimal results.

-queue music "Live to Win"-

-strong upward fist pump-

-fade out-

-end scene-

Thursday, June 8, 2017

SX70 Instax - Film Adapter Thingy (I-FAT) Update

Hahahaha... I just wrote that title and did a spit take. That's really not what I'm calling it. Aren't we having fun? Pull up a chair, this might be a long one and I'm on allergy meds.

May saw a tidal wave of camera repair jobs and freelance design projects and June-July is looking to be the same. Needless to say I'm a bit overwhelmed so I've been updating customers that there is currently about a 6-8 week backlog and I'm trying to get to them as fast as possible. One thing I can promise is that I won't rush a camera. It's best to do it right the first time, each time, rather than have it shipped back to me, I then cuss at myself for the rush job I did, and then send it back hoping I fixed it... it's a bad habit to get into and I want each camera shipped back looking and working like new.

One of the many side projects I've been trying to cram in to my schedule is the whole SX70/Instax film pack idea. For those of you on Instagram, you've seen a few pics but I never really had a chance to explain it entirely because Instagram is all about pretty pictures and I like to post pretty pictures. So I'll try to explain what exactly this whole project is and why I'm beating myself up trying to do it.

My first reconnection with film photography was when I bought my first pack of Impossible film in 2011. To this day I'm a brand-loyal Impossible Project (now Polaroid again) customer hands-down. When ordering film, I order a three pack of 600 film first. Whatever money I have left over, it's HP5, Portra400, and then maybe, just maybe, I'll snag a pack of Instax Mini if I bump into it at Rite-Aid. I absolutely love what Impossible has done with introducing an entirely new market, as well as reintroducing to the previous one, to instant photography and they firmly believe in preservation of what many consider to be their favorite camera, the SX70. So when reading about this project, keep in mind that my intent has never been to replace TIP film in my SX-70 with Instax. The intent is to be able to use multiple formats in my SX70 while performing no modifications to the camera whatsoever.

This whole concept started when a few folks on IG showed off their Instax mini film used in different cameras. Very cool. I tried a few and it was fun but (permission to speak freely) a total pain in the ass. Only one shot could be exposed at a time, there was a lot of fumbling around trying to tape an unexposed piece of film to a darkslide, and the film had to be processed in a different camera. And all this had to be done in a changing bag. Pleh!

Here's a few pics I took with Square film in my SX70. I did a temporary capacitor swap and converted the metering to 800. All have been processed in an Instax 210.

So what would it be like if my SX70 was a multi-format camera, much like the Yashi 635 (120 and 35mm). What if I could use Instax Mini, Impossible Project film, and now Instax Sqaure in my SX70 without any modification to the camera at all? That would be cool... I could do some more delicate portrait work or fine art influenced compositions with my Impossible project film. Or I could pop in some Instax and take random everyday shots or vice versa. There's pros and cons (all subjective) to both formats. Impossible film is a bit more expensive, but the resulting image is unbeatable. Instax film is cheaper and more readily available but lacks the soul that you can get from Impossible film. Impossible film is expressive while Instax is convenient. There certainly is a common middle-ground but wouldn't it be great to use both films in the same camera?! Hell yes it would!

Here's a few early sketches for a mini-pack concept (before Square was officially announced) showing the adapter pack, film door, and daylight loading mechanism to transfer Instax film to the adapter pack.

Here's my thought process for the project. If I can't really explain this that well in words, then I'll be posting some drawings and 3D development soon. Plus I suck at staying on-topic and lack organized thinking.


Shoot three different instant film formats, Impossible, Instax mini, and Instax Square in an SX-70 camera. 


- The mechanics and dimensions of the SX70 are not compatible with Instax film packs. 
- Instax film is rated ISO800 while the SX70 is designed for 125. 
- Instax film is exposed from the back of the print. Final images will be reversed.
- Instax cameras are designed with incredibly tight roller tolerances and the gearing is greatly reduced   to allow enough torque from a standard DC motor to pull the film through the rollers.
- Only one sheet of Instax film can be exposed at a time in an SX70 and a separate Instax camera is required for processing (see next bulletpoint). All film must be transfered from camera to camera in the dark.
- SX70 roller tolerances and roller spring strength are not tight enough to apply the correct pressure and evenly spread the Instax developer throughout the entire print. Undeveloped patches at the upper corners would be seen.
- Instax film packs don't have the integrated battery packs like Impossible packs. An external battery pack would be needed.

There's more but I can't think of them off the top of my head but you can see there's a laundry list of why this wouldn't work. But I'm a stubborn product designer and I like to think of ways to make this work. So me and a few others started brainstorming. Actually I just started babbling about it at the pub and a few of my buddies offered their .02.

What if there was a specialized film pack the same dimensions as Impossible's, in to which a user would load 10 sheets of Instax film, load into the camera, and take 10 consecutive shots just like they normally would?

Addressing the loose tolerances of the SX70 rollers and my desire to keep the camera as original as possible, why not remove the SX-70 film door (it's really very easy to do... like really easy) and replace it with a newly designed door that contains it's own gear train, motor, battery supply, and delivers the print with the same torque and pressure specs to that of an Instax camera?

With a little help from one of my engineering buddies "bebeh" Kyle, we started developing this system. There's still a list of hurdles we need to get over but I feel like we're 80% to proof-of-concept. A few things we still need to work out:

- The original intent was to have this as a completely automated cycle. In other words, you press the shutter button and the camera does the rest... Just like Kodak used to say when people still used tooth powder and thought putting butter on a burn was the right thing to do. Problem here is the RPMs of the film door mechanism isn't the same as the RPMs on the camera. Speeding up the RPMs on the film door to match the SX70 drastically reduces the developer spread of the Instax film. And I broke my gear train on my SX70 due to the strain of trying to push the film sheet into ultra-tight rollers (luckily I know I guy that fixes these cameras. See? Having fun.).

To solve this, and for simplicity sake, it would be easy enough to move the rollers forward and let the camera cycle naturally without ejection. Then the user would push/pull a lever attached to the film door to mate the film sheet to the rollers. This lever movement would do two things:

1. moving the lever would mechanically mate the rollers with the film sheet
2. moving the lever would trip a switch that starts the film door motor initiating a film door cycle. This prevents the SX70's geartrain from clutching out and breaking.

So essentially this would become a two-step process. 1. press the shutter button for exposure allowing the camera to cycle 2. push/pull the developing lever to eject the print. I think we're a patient enough group to look past the fully-automatic process in favor of something a bit more manual. Or we could say "screw you" to both methods and slap a big ol crank on there so we can GRRRRRRIND our film from the betwixt the camera rollers.

So, that's where we're at. There's a few things that I'd be willing to live with like a reverse image, the need for an ND filter built in to the adapter pack, and my biggest user-unfriendly gripe which is the user would need to load the film sheets in a darkroom. I have some ideas for a day-lab like mechanism that would take the film from an Instax pack and transfer it to the adapter pack (pictured above) but just thinking of the engineering on this gives me brain-fever. That'll be saved for later. Our last steps are to get the roller tolerances correct, document a few test cycles, slap a visual and 3D presentation together in case anyone (permission to speak freely again) with a shitload of money wants to take control of this project, and get back to repairing and selling cameras and drawing pretty pictures. This has been a fun project and I hope more like this pop up later on.

All this writing that I hoarked up on your screen has built up to answering my original question... why am I doing this? When someone would ask me that I would look at them just as a nine-year-old would after their dad busted into the room screaming "WHAT THE HELL ARE YOU DOING?!" after finding that child lighting hairspray (Aquanet to be exact) soaked explosive caps on the carpet. Yes that happened. I give that kind of blank "f-if-i-know" stare. I guess I just really like instant film and my SX-70 and I wanted to see if I could do it. It's a simple answer. :)

Oh, and the filthy money. I want to make millions and millions of dollars and sleep on a pile of money every night with a woman under each arm. #yeahright