Film processor design

[Sim2]

This may be a bit off-topic as far as a film processor goes but one bit of "auto-kit" that I would like to have would be an automated fill/dump facility for a print washer.

Some of the washing of fibre prints is, I believe, achieved by the diffusion process so rather than having the wash water running all the time I am thinking of a process where the print washer fills and stops, the print "soaks" for a period of time, the tank empties and refills and stops again, the print "soaks" etc for as many cycles are required to effectively wash the print. Could the intervals of "soaking" could be programmed? Could the temperature of the input water be controlled, similar to an intellifaucet?

Advantages could be reduced water usage, less time with the tap actually running (noisy). Disadvantages may be the different fixings needed for different print washers.

Off-topic for your processor I know, but if you don't ask.... Anyway, good luck with the project, should be fun!

 

[Mr Bill]

Hi, sounds like a fun and interesting project. Something I've thought would be fun for my personal use with Nikkor-style film tanks would be a little robotic arm to hold and agitate the tank for me. (This would include timing the inversion cycles, as well as rotating the tank, in increments during that process.) But I doubt there would be much public interest in such a thing, and in truth, not even worth it to me.

I think you have to start your design process based on the 1) the processing volume of a hypothetical user, and 2) consideration of whether they would use replenished chemicals or not. In my experience, the most economical way to work is with replenished systems, provided that one does enough volume. If you lean this way, the Jobo-style rotating tank option is out, as the exposure of developer to air will "kill" the developer in short order.

I've used, or worked with in some way, quite a lot of processing machines over the years. Just about everything except what they call "dip n dunk." Which is exactly one type that I think you should consider. In fact, the Lego machine (with the little car on a track) is exactly that style.

Such a machine could use fixed tanks that don't need to be filled or drained (routinely), and it could be left up to the user to handle the replenishment manually. Tanks are simply covered between processing runs. There would be no need for pumps or solenoid valves, saving costs as well as eliminating possible failure points and risk of leakage.

The guts of the machine would essentially just be an overhead guide rack with a movable carriage. The carriage would transport loaded film reels, etc., from tank to tank, lowering them into each solution, and agitating with an up/down motion, perhaps vibrating to dislodge air bubbles (do bench tests first, to confirm). Such a setup could be reconfigured with different length guide racks and programmed times, etc., to handle just about any process.

An elaboration to allow C-41 processing, would be a good temperature control system in the tanks. Commercially, water baths used to be common; they isolate heating elements from the chemicals, if nothing else. (If you heat individual tanks, you need to add a circulation system, too.)

Just thinking about it almost makes me want to try rigging something up. Best of luck with your project.

Ps, see the older book, "Handbook of Photographic Science and Engineering," the old industry bible, for info on construction materials as well as an overview of processing methods.

 

[MartinP]

Instead of a 'fire-and-forget' system, perhaps it would be simpler (equals cheaper, smaller and more saleable) to have a tempered system with two 'active' containers and the machine filling from each in turn via a device doing the same function as the Jobo's lifting arm. The user would then be responsible for swapping the last-used container for the next-but-one bath (perhaps also with a manual, or automated, rinse of the just-used container pipework), from the tempered area of the machine. Tanks used could be Jobo, as they are designed for rotation and seem to work well. It should be possible to have the machine running in daylight as that will be a big requirement for most users.

 

[alex2293]

This may be a bit off-topic as far as a film processor goes but one bit of "auto-kit" that I would like to have would be an automated fill/dump facility for a print washer.

Some of the washing of fibre prints is, I believe, achieved by the diffusion process so rather than having the wash water running all the time I am thinking of a process where the print washer fills and stops, the print "soaks" for a period of time, the tank empties and refills and stops again, the print "soaks" etc for as many cycles are required to effectively wash the print. Could the intervals of "soaking" could be programmed? Could the temperature of the input water be controlled, similar to an intellifaucet?

Advantages could be reduced water usage, less time with the tap actually running (noisy). Disadvantages may be the different fixings needed for different print washers.

Off-topic for your processor I know, but if you don't ask.... Anyway, good luck with the project, should be fun!

Down the road, if we design an automatic machine, with user programmable cycles, washing fibre prints should be fairly easy for the user to program. The way I see it, the user would have to set the process temperature, set the first step to filling the tank with water, choosing the agitation time, frequency and speed, the soaking time and the amount of cycle repetitions. Of course, the maximum print size is limited by the size of the machine.

I think you have to start your design process based on the 1) the processing volume of a hypothetical user, and 2) consideration of whether they would use replenished chemicals or not. In my experience, the most economical way to work is with replenished systems, provided that one does enough volume. If you lean this way, the Jobo-style rotating tank option is out, as the exposure of developer to air will "kill" the developer in short order.

I was not aware that replenishment was not possible using a Jobo processor. Is the exposure of the developer to air is higher in a JOBO machine than in an inversion tank? Xtol replenishment is quite popular with inversion tank.

I've used, or worked with in some way, quite a lot of processing machines over the years. Just about everything except what they call "dip n dunk." Which is exactly one type that I think you should consider. In fact, the Lego machine (with the little car on a track) is exactly that style.

Such a machine could use fixed tanks that don't need to be filled or drained (routinely), and it could be left up to the user to handle the replenishment manually. Tanks are simply covered between processing runs. There would be no need for pumps or solenoid valves, saving costs as well as eliminating possible failure points and risk of leakage.

The guts of the machine would essentially just be an overhead guide rack with a movable carriage. The carriage would transport loaded film reels, etc., from tank to tank, lowering them into each solution, and agitating with an up/down motion, perhaps vibrating to dislodge air bubbles (do bench tests first, to confirm). Such a setup could be reconfigured with different length guide racks and programmed times, etc., to handle just about any process.

The Lego machine video is really interesting. I did not consider the dip and dunk process as I've only seen it using nitrogen-burst agitation. I've also never seen it done with the film on a plastic or stainless reel, I don't know how this would affect the developing time given from the developer spec sheet. Our goal is to make a user friendly machine that can be used with the chemical products spec sheet as a starting point.

Again, thanks for all your comments and input. We read all of them with great attention.
I'll post the link to our first survey here this week.

Alex.

 

[flavio81]

Hello everyone,

I've been reading here for almost two years and recently subscribed. This forum has been really helpful getting me started in analog photography.

I am presently studying mechanical engineering in college and five of my classmates and I decided to design an automatic film processor for our final design project.

I think the key to popular adoption would be "AFFORDABLE" and this would mean removing some features and leaving the most important ones.

In my opinion the essential features would be:

- Automatic Temperature control of the liquid inside the development tank to within 1°C or maybe less.
- Automatic agitation to a pre-programmed interval (for example different settings for stand development vs normal development)

I don't think that a system that automatically pumps-in or pumps-out chemicals is essential; i think that it would be OK for me to flush chemicals and pour the "next" chemical in the tank manually as long as your machine has:

- Automatic "alarm" that tells me when it's time to change to the next chemical. Furthermore if the device can connect to the internet via Wi-fi, then it could be able to do a "push notification" to my smartphone so i am notified via the smartphone that i need to run to the machine to change to the next chemical.

All above could be done using a controller like a Raspberry Pi board.

Now, if you want to add the DELUXE features, then yes, automatic pumping in-pumping out of chemicals would be great. Also, as mentioned above, some of us would like to use replenished developers so an auto-replenishment function would be GREAT...

Keep up the good work! Good luck!

 

[calebarchie]

I am currently designing a system similar to above. Been in development for several months and ongoing with reiterative planning, testing prototyping etc. This is just a side project as an ID student, keeping me occupied.

That said, as a final design project for mechanical engineering it may seem a bit underwelming. But if it is suitibly in line with your academic requirements then best of luck.

 

[Mr Bill]

Hi, regarding oxidation of developer: the typical inversion tank is somewhere (very roughly) about 90% liquid, with the remainder being air. In the Jobo-style system, this is reversed, being mostly full of air. Then as the rotary system turns, it brings a film of developer (attached to the drum surface) into exposure to air. So there should be quite a bit of developer oxidation going on.

It occurs to me that you may not be familiar with Kodak's Z manuals, the bibles for C-41, etc. Have a look for Kodak Z131, available online. Part 3 covers "sink-line" (like my suggested method) and "rotary-tube" (like Jobo). With rotary, Kodak recommends discarding developer after a single use.

My understanding is that commercial dip n dunk machines almost always use N2 burst agitation, but I don't have first-hand knowledge. But those machines are typically hanging strips of film, several feet long, in the solution; it would be very difficult to manually agitate that film. But if you were to have Nikkor (or other) reels semi-rigidly attached to a hanging arm, I really think (guessing) that you could could use a random sequence of up/down motions (solution flowing through the reels) to get good-enough agitation. You would have to do a few tests to confirm, but I don't see how it could be worse than the smooth rotation of a Jobo, which most photographers seem satisfied with.

By the way, I overlooked the issue of aerating the C-41 bleach. You could run a few rolls okay, but after that it needs to be exposed to air to get its bleaching power back (a Jobo handles automatically by the mechanism of agitation). So unless you build in an aeration system, you need to give the operator an easy way to get the bleach out so they can shake it up in a jar, or whatever they do.

By the way, regarding inversion tanks, my pet theory of how the agitation works is that a large air bubble travels through the reel on inversion, and this is what semi-randomly disrupts the developer/film interface. I think that in order to get best agitation, the bubble has to go through a different part of the reel each time, so the tank ought to be rotated a third of a turn, or so, each set of inversion cycles. I've never actually studied this, but it's how I imagine it to work. (If your school has a chemistry lab, ask someone to fill a volumetric flask with water, and go through the motions of "mixing" it (via inversion); you'll see how the bubble behaves.)

Ps, I would be real hesitant to use replenishment long-term with an inversion tank system. It MIGHT work fine; someone's actual experience would be more valuable than my gut feeling here. The rate of replenishment would have to outweigh the rate of oxidation of both the developing agent and the preservative(s). As a note, even if it works with b&w, color might be a problem - color developers have low concentrations of preservatives, such as sulfite. If you added a lot of sulfite to a color developer, it will drastically inhibit formation of the color dyes.

Pps, Richard Henry, in "Controls in Black and White Photography," has a short section on "Uniformity of development," which might be of some use to you - he tried a variety of common agitation methods.

 

[alex2293]

I think you've got a great point Flavio81, designing a simple, reliable, affordable machine that has only the essential features. Deluxe features should be in option to keep the base price low.

It is a bit underwhelming Calebarchie, but we accept the fact that we have to fill the consumer's needs in order to make the commercialization possible. If most consumers want a basic affordable machine and we design a fully automatic one that has all the features one could think of, we probably won't sell many because clients are not ready to pay that much for feature they do not need. In the end, I think there will still be some challenge to design the basic one and have affordable price tag. Take a look at a new Jobo, it does not have automatic liquid transfer, but the price is still about 3700 USD.

Thanks again Mr Bill for all the info, I'll take a look at the Kodak Z131 manual and "Controls in Black and White Photography".

 

[Diapositivo]

I wouldn't go beyond the complication of a Jobo.
Kitchen digital timers cost a dime, can be bought in different colours for the different phases, and have precision of a second.

I use a photographic "triple" timer but nowadays I would just buy 3 or 4 kitchen timers of different colours for E-6. Besides, the timing of the last baths is not really crucial. One pays real attention only to the development bath, that's something to be done exactly. No phone calls ;-)

I personally would not include wi-fi stuff. Even taking aside the cost, those things can brake, mulfunction, conflict with other wi-fi devices in the house (or the house nearby) and the producer would be bothered with help calls by clients for functions that are extraneous to the core of the product.

The user can use a smartphone timer (or multiple timer) and turn it on manually after pouring the liquid (as we do now with the Jobo, don't we) and then he can go away for a few minutes knowing that the smartphone will emit a sound in case they forget or that phone call lasts longer than expected.

I would stress compatibility with common plastic bottles so that one doesn't have to chase particular flasks as is the case with Jobo. In order to avoid contamination flasks should be used only for the same chemistry when using one-shot solutions. People developing B&W, colour negative and slides would need at least 12 different flasks. Buying new Jobo flasks in this number begins being quite expensive!
For the short duration of a one-shot preparation common mineral water plastic bottles should fit the task very well (not for conservation, only for use of diluted solutions) so the machine should be compatible with common flasks.

I would also avoid the digital thermometer of the Jobo. These things are now sold for cheap and I would not incorporate a thermometer or anything having a display. For a manual device, what matters is that the temperature can be maintained on a stable temperature. Pump, heater, thermostat. And a decent rotation mechanism, like the one Jobo has, or even identical to the one Jobo has, for compatibility.

If it wasn't for rotation, the entire Jobo system could be easily replicated with an acquarium pump, an acquarium heater, a submersible thermometer, and some system to avoid the bottles to float.

Actually I would try to engineer only the rotation/inversion part and then give a list of suitable products compatible with the device (heater, thermometer etc.) keeping all that complexity outside of my firm (magazine investment, client support).

A glorified kitchen sink will work wonderfully. Just give us rotation and maybe a "lift" like in the Jobo! The rest is easy and I would keep it outside the product.

 

[Steve Goldstein]

As far as I can see nobody has mentioned a key parameter - size. This would be a make-or-break choice for me as I work in a micro-darkroom and will do for quite some time more.

 

[alex2293]

Thanks again for your input Diapositivo and Steve, we'll consider it.

As I've already said in this thread, we created a survey. If you could just take a minute or two to take your time to fill it out, it would really help us. Just click on the following link.

http://surveynuts.com/surveys/take?id=59946&c=759635729JLPT

If you have any comments, feel free to send me a PM or write it in this thread.

Once again, thank you for your help.

 

[Diapositivo]

As far as I can see nobody has mentioned a key parameter - size. This would be a make-or-break choice for me as I work in a micro-darkroom and will do for quite some time more.

This leads me to another consideration.
As we know, Jobo did some "big" machines and some "small" machines.

I suggest separating the two functions.

An "agitating machine" is only devoted to agitation and pouring. That corresponds to the left part of the upper part of the Jobo. It should end up being fairly small even for large tanks. If no "lift" is provided, it might end up being fairly small.

Then there is the "bath machine". That has the circulating pump, the heater, and the thermometer, sold as a kit. The user will put them inside a large plastic basin or a small plastic basin according to its need. The "bath machine" can also reside separated from the "agitated machine" which will allow a better exploitation of darkroom space (or bathroom space).

The agitating machine would require, to do things well, a basin of its own with a thermostat of its own (no circulating pump needed in that case, I suppose). The two basins should be brought to the same temperature with the help of the digital thermometer which is sold with the kit (or maybe two digital thermometers will be in the kit, that stuff is not overly expensive, one remains with the agitating machine, and the other in the basin with the flasks).

The same basic product might serve the small darkroom and the large darkroom and maybe even a makeshift darkroom.

 

[Diapositivo]

Alex, I completed the survey.

I suggest not falling into the devilish temptation of trying to serve the bigger market (such as 135 or 120 roll film) at the expense of the niche market (supposing sheet film is less common).

One of the reasons people use to justify the purchase is also the future polivalence. I might do no more than 2 rolls at a time today, but I would be interested in being able to develop 4 rolls at a time tomorrow. I might only use 135 today, but I would not spend money for something that will not adapt when I want to explore different pastures.

The new product, even though unsophisticated as far as functions are concerned, should be as polivalent as possible. Rolls, sheet film, paper, B&W, C-41, E-6. It would have a wider consumer base and also a wider appeal for every client IMO.

If need be sacrifice automation, raise price, but do not tamper with flexibility of use ;-)

 

[Mr Bill]

I might do no more than 2 rolls at a time today, but I would be interested in being able to develop 4 rolls at a time tomorrow.

Would it really bother you if the processor only took 2 rolls at a time, so you have to do multiple runs for 4 or more rolls? If the machine is doing the work, do you really care that you have to restart it a second (or more) time(s)? Or is the issue the maintenance between runs for a Jobo? (A replenished dip n dunk would have no maintenance - only a set of fresh reels plus the addition of replenisher at whatever interval one uses.)

If need be sacrifice automation, raise price, but do not tamper with flexibility of use ;-)

I agree that flexibility is important, but mainly in the ability to reconfigure the times and number of solutions. If you want the single machine to do every process you want, this seems to bring the risk of error - perhaps you loaded a wrong chemical somewhere.

To me, one of the main priorities for a machine is high consistency followed by a fail-safe design. And if you have the possibility of, for example, processing a color film in a b&w setup, then I don't see this as good. So I'd sacrifice the flexibility of multiple processes, at will, for the safety factor.

An open question to all who desire a rotary tube design, have you ever used replenished systems? If not, why not? (The photofinishing industry does this almost exclusively in order for 1) lowest chemical costs, 2) minimal chemical handling, and 3) minimal effluent loading.)

On a different note, I don't put much faith in surveys like this, other than to establish how much processing is being done (and thus, how much potential market there is). In my experience with a large studio chain, surveys NEVER identified any breakthrough ideas. But after the fact, when some game-changing idea has been put into use, people would say they never had any idea it would be so useful. Now they can't do without it. I think a well-designed, easy to use, foolproof, and "cheap" processor might be such a thing.

 

[Mr Bill]

In retrospect, answering my own question about replenished systems, I think it may be beyond what most users are willing to deal with. I've dealt with it for many years, on a really large scale, so it's old hat to me, and I tend to forget that the regular photographer has almost no experience in this sort of thing.

To handle replenished C-41, one should really run process control strips (from the manufacturer), check them with a densitometer, check the pH or specific gravity of bleach, etc. There are roundabout ways to handle these things if one understands the drill, but I think that people who are primarily shooters won't want to deal with it. You might be able to process for 30 cents per roll, or so, but the cost savings, in total dollars, aren't that great.

A really clever machine designer could build in all sorts of features, even a crude densitometer, but it would take either a lot of tech support for the user, or troubleshooting software, either of which are way beyond an ME student. I think it would take a major movement with a user group providing its own support.

 

[Diapositivo]

Would it really bother you if the processor only took 2 rolls at a time, so you have to do multiple runs for 4 or more rolls? If the machine is doing the work, do you really care that you have to restart it a second (or more) time(s)?
[...]
I agree that flexibility is important, but mainly in the ability to reconfigure the times and number of solutions. If you want the single machine to do every process you want, this seems to bring the risk of error - perhaps you loaded a wrong chemical somewhere.
[...]
To me, one of the main priorities for a machine is high consistency followed by a fail-safe design. And if you have the possibility of, for example, processing a color film in a b&w setup, then I don't see this as good. So I'd sacrifice the flexibility of multiple processes, at will, for the safety factor.
[...]
An open question to all who desire a rotary tube design, have you ever used replenished systems? If not, why not?

I use one-shot dilutions (see below). That actually is two-shot solutions because I do reuse once the solutions but only if the second run follows immediately the first. The entire developing phase is a bit time-consuming, as I have to wash carefully all flasks, I have to dilute exactly the solutions, I have then to wait in the meanwhile for the Jobo to go in temperature (all the plastics, both trays, the flasks, the water, and it takes more than half an hour, certainly I can write nonsense on APUG while I wait ) and the entire 6-bath process, which is actually a 10-bath process - a rinse before developing, a rinse before fixer if I reuse, deep and long rinse after fixer, and a last bath on stabilizer outside of the tank, takes one-hour from beginning to end if we include the final rinse of the machine.
If I do two shots I end the entire work after around 90 minutes, actually more because there is the final rinsing of every single piece.
If I had to process rolls from a holiday or some "event", or if I begun developing film for somebody else, it would be practical to have the possibility to do 8 rolls in two shots. But usually it can be one shot in, maybe, three weeks. So it's very variable.

Besides, my second-hand Jobo CPP-2 originally was the backup machine of a laboratory. This is not an uncommon use, I believe. A small lab can have an automatic developer, and a manual one as a backup.

The reason I use one-shot solutions is:
- fresh solution every time, no worry about health of chemistry;
- very repeatable results;
- easier conservation of the chemistry when kept concentrated;
- less stink in the bathroom;
- the one-shot method is ideally suited to rotary processing.

I never considered using replenished system. I see it as something fit for laboratories or professionals who develop themselves and develop almost every day.

Regarding the risk or errors, I said what I said having in mind a replacement for my Jobo CPP-2, a manual machine. A manual machine should be very flexible in size of tanks (including large paper sheets and sheet film) and number of phases. I understand an automatic machine could be dedicated to only one kind of process, but then again, IMO should have the flexibility to treat paper and sheet film. You can't produce a model for paper, a model for rollfilm, a model for sheet film... even if the final user uses it only for one purpose.

 

[Wallendo]

I think pricing should be your primary concern. It may well be technically possible to build a better processor than an advanced Jobo, but there is a very small market for such a device. People who already have an advanced system are unlikely to change, especially to a system with no reliability history, customer support, etc.

I suspect there is some market, however, for a device that allows for temperature control of solutions and easy emptying and filling at low cost. I would buy such a device, so at least you have a market of one.

 

[Diapositivo]

Considering the absurd price at which Jobo processors are sold new, there is a huge margin of profit for a second entrant even if keeping the price pretty steep ;-)

I hope the device is convincingly cheap though, home-made development could be the last ditch for analogue photography, and at the same time its future. Ferrania said they would pay great attention to the world of home developers, I think home developers are key to the survival (and future progress) of analogue photography.

 

[calebarchie]

I will echo some users here and say a survey and other quantitative based research is generally poor (especially as a first step) in finding a true need, problem or opportunity for design intervention.

Qualitative data is far more valuable in this regard, like this thread.
As for being underwhelming I'm referring to academically, unless you actually plan to really bring it to market, not so much the product itself.

 

[Mr Bill]

The entire developing phase is a bit time-consuming, as I have to wash carefully all flasks, I have to dilute exactly the solutions, I have then to wait in the meanwhile for the Jobo to go in temperature (all the plastics, both trays, the flasks, the water, and it takes more than half an hour,

I get it. In contrast, using the system I had proposed, the only cleaning is the film holder - whether it be a couple of reels or some sort of basket or rack. Everything else is the same, no waiting to reach temperature. The only "chemical mixing" is to add a bit of replenisher from a storage tank.

The only problem with replenished systems is that you may have to change the rate periodically, and that you need to be able to know when to do so, and how much. And if you don't do it right you can get into trouble - this is when you need the know-how.

I've spent a large part of my work life heavily involved with this sort of thing, and I could never understand why a lab of any decent size would use a rotary tube processor; I guess that with a certain processing volume there is no other game intown.

 

[alex2293]

I read all the posts that followed mine where I posted the link to our survey.

I understand that this thread gives us a great amount of qualitative information and it is really valuable to us. It gives us a large amount of needs, problems and opportunities. However, completing the small survey will take you 2 minutes at most. The point of the survey is not to ask you directly what do you need/want, I've done this here and got great feedback, way more than expected. The point of this survey is to put down some numbers on quantitative issues like the usage frequency, format, process, price, etc. I think that both types of data, qualitative and quantitative, are complementary and helpful together.

Thank you again.

Alex

 

[CropDusterMan]

Alex,

Do me a favor, if you and your friends come up with a good working design, please use this as your music
in your promo/demo video. Gets really good at 30 seconds...us older folk will recognize it.

 

[Antonio Marques]

As a consumer, I'd love a 'set and forget' approach. I don't mean everything had to be automated, if there are steps that are much easier if done by the user then fine, but if the repetitive tasks can be automated away, it'd be great. Because if I have to commit the same time to it than if I were doing it all manually, what's the point? Uniformity of result isn't too important to me, if it were I'd be just shooting cellphone.

 

[mshchem]

Keep it small. Precise temperature control, pumps for agitation. Shouldn't need more than 1 liter baths. Don't try to accommodate 6 bath e6. Just four bathes max. Use Jobo 1500 series reels. 1 liter round tanks would accommodate 4 roll of 120 (2 per reel ) Avoid hard tooling by using existing reels and columns . You can't afford injection molding tools.

Consider a radical idea make it for use in a darkroom lights off

4 1 liter PVC pipe tanks temperature controlled, a small fountain pump to agitate the liquids.
Load up your reels plunk them in the first tank for 3 min and 15 seconds, beep, manually transfer reels to bleach , next fix, wash with a simple syphon type washer. Final rinse done.

With the Kodak C41 RA bleach and fixer you would only need to be in the dark for 5 minutes..

I always did it manually, but picked up a Jobo CPP2, for 175 bucks tinkered it works great.

All the mountains of used stuff out there is your competition.
I would focus on something that would make it easy for a young enthusiast to be able to process roll film in a dorm room closet with excellent temperature control.

My two cents, best of luck make something cool! Mike

 

[Europan]

Let me be devil’s advocate. I shouldn’t purchase your processor because I can do it all by hand. More than that, manual processing is flexible, cheaper, better, and less prone to failures. Except when I faint

I know several machine processors, Colenta, Photomec, ARRI, Debrie, Calder, JOBO. The big ones need to run as much of the time as possible in order to be economical, the smaller ones have too little bath volumes for longer runs. So they’re too expensive for the purpose. It’s a catch 22. End of DA speech

One point that needs close attention is cleanability. Do yourself the favour and design everything so that the processor can easily be cleaned. No narrow cavities where you can’t put your fingers in. There will be tenacious residues in tanks, hoses, pumps, on everything in contact with a bath. There will be silver-compound muds and dyes all over. If you’re not familiar with photo chemistry, invite someone into the team who is. Good luck!