Anyone who has ever processed real analog film in a darkroom probably remembers two things: the awkward fumbling in absolute darkness while trying to get the film loaded into the developing reel, and the tedium of getting the timing for each solution just right. This automatic film-developing machine can’t help much with the former, but it more than makes up for that by taking care of the latter.
For those who haven’t experienced the pleasures of the darkroom — and we mean that sincerely; watching images appear before your eyes is straight magic — film processing is divided into two phases: developing the exposed film from the camera, and making prints from the film. [kauzerei]’s machine automates development and centers around a modified developing tank and a set of vessels for the various solutions needed for different film processes. Pumps and solenoid valves control the flow of solutions in and out of the developing tank, while a servo mounted on the tank’s cover gently rotates the reel to keep the film exposed to fresh solutions; proper agitation is the secret sauce of film developing.
The developing machine has a lot of other nice features that really should help with getting consistent results. The developing tank sits on a strain gauge, to ensure the proper amount of each solution is added. To avoid splotches that can come from using plain tap water, rinse water is filtered using a household drinking water pitcher. The entire rig can be submerged in a heated water bath for a consistent temperature during processing. And, with four solution reservoirs, the machine is adaptable to multiple processes. [kauzerei] lists black and white and C41 color negative processes, but we’d imagine it would be easy to support a color slide process like E6 too.
This looks like a great build, and while it’s not the first darkroom bot we’ve seen — we even featured one made from Lego Technics once upon a time — this one has us itching to get back into the darkroom again.
Design looks solid in principle, but I’m not sure about how light safe a reel agitation mechanism would be. For color you still have to throw it all in a computer controlled tempting bath. So if course I think it’d be cool if this did temperature control too.
Reel agitation mechanism is exactly as light safe as developing tank, that is used to build it. The .stl files of servo mount are designed to work with AP tanks (that are still in production), and the OpenSCAD parametric model included in the git repo can be customized to use any other tank with stick agitation.
The temperature control is done with LTLT-cooker aka Sous Vide, which is submerged into the same water bath as the machine, as shown in the video. The lower part of OpenAutoLab is water-tight, and the vessels have magnetic mounts, so they don’t float in water when empty.
If you mean that the temperature should be managed by controller itself, I did that in previous version. But if I want amateur enthusiasts to build OpenAutoLab, I shouldn’t bring unnecessary complexity, if cheap off-the-shelf external device solves the problem, and I definitely should not spice the PCB up by bringing 230v into it :D
I forgot Jobo tanks have that light-baffled spoke thing down the center don’t they? So there’s a built in way to move the reels that’s already light tight?
There is an alternative to proper agitation: “stand” developing, which works with dilute developer, longer processing times, and maybe one or two agitations after the initial one to release air bubbles. It also involves deliberately letting developer exhaust itself in the brightest parts of the image, automatically “dodging” them down to restore some detail. As a result it has a slightly different look than traditional developing, which some photographers very much like. I was playing with it, using Caffeinol as my developer, for my Graflex large-format SLR; need to get back to that project.
Agitation is fully controllable from the settings menu, one can perform stand and semi-stand development with this machine without changing any code. Semi-stand development in caffenol was one of the motivations to build the first version of what became OpenAutoLab, because waiting for an hour and not forgetting to agitate every 10 minutes was too hard for me, let the robots deal with it.
Many times I thought about adding a servo or DC motor to the top of an AP tank to help with agitation, but like you mention I now I semi-stand develop my black and white 120 film in 510 Pyro with only 1 agitation and find I get much better results. For colour film though, I use a Jobo machine with water bath. I’ve been processing ECN2 movie film in C41 chemicals recently; I might take a look at this for our lab.
There is a nice chemistry for C-41 which is more tolerant on temperature:
https://cinestillfilm.com/products/cs41-simplified-color-processing-at-home-quart-kit-c-41-chemistry
You need just a thermometer to adjust the processing time, but not a perfectly controlled environment.
(doing this at home with a simple sous-vide heater in a water pot)
It’s not some special magic chemistry, any c41 dev kit can be used to develop in room temperature by compensating slower reactions with longer developing times. There are tests online with direct comparison of developing in standard temperature vs room temperature. The thing is, you can’t really compensate all of the reactions together with diffusion speed into all of the layers of the film, so you always end up with color shifts. Yes, it’s not that critical and can be easily fixed digitally, but buying a 40€-sous-vide and setting it to 38 degrees is even easier.
🙄 As a professional master printer, former fine art lab owner with about 50 years of experience, manual processing is by far easier and allows for far greater control of the development than any automated system. And like the countless of other automated film processing systems that have come….and gone…over the decades, this one will be just some new toy that some people will try just because they don’t have the experience knowing what hand processing can actually achieve.
As someone who has only 15 years of manual processing, of course I may miss some fine details. When developing black-and-white film with this machine (it is all about negatives and has nothing to do with prints), I can control exactly the same parameters as I do with manual processing: I choose the developer, its dilution, its temperature, development duration, agitation strength and times when to agitate. If there’s more to it, I just don’t know about it: and will happily listen if you can share some insights.
And you are absolutely right about a toy: the whole premise was to build a cheaper and repairable contemporary alternative to machines like Jobo AutoLab, which, by the way, are not gone, although not produced anymore. Noone should take a device made out of aquarium pumps, valves and tubes too seriously :)
Indeed. I think the key is in your intro. “As a master printer with 50 years experience “. Having put in the hours and gained the knowledge, and no doubt made a few mistakes on the way, you have no need for a machine. Put on a decent CD, find the right station, and away you go. Seriously, isn’t this just a variation on a Jobo?
As someone who works on precision industrial film developing machines (Agfa), I really appreciate the effort put into making this process automated, but I also see how far this has to go before it becomes a polished product.
Good luck with your venture 👍
Oh, it’s not supposed to be a “product”. There are products, which are somewhat similar, on the market: there are used Jobo machines, newer Filmomat, Dev.a and Osiris F1, all of them are more polished than anything I could develop in my free time for the rest of my life :). But that’s not the goal of the project.
OpenAutoLab is inspired by the first RepRap 3d-printers. It’s a proof, that film processing machine suitable for non-professional enthusiasts can be simple to build and repair, can be well documented and contain no secrets like magic button combinations, or special parts (which who knows where to get if the manufacturer stops producing them), and can be further developed by the community if there is interest in it.
I don’t do color but I have an old Jobo processor. Long ago I decided to use only Diaphine except for special projects, so I have virtually no dependence on time and temperature, within reason. In fact, that looks like a Jobo tank and reels? Maybe I missed something. The fully automated nature of this project is pretty attractive