That's it! I'm designing and building my own film processor.

[AlbertZeroK]

Bearing in mind that we're using the Jobo processors as the basis for this design, and the Jobos do not have any timing functionality, I'm looking at a followon project that would be a process timer, but as a separate device. There are not really any spare inputs/outputs on the processor we're using for this device either. That's to keep the cost down (although the CPU is not really a major part of the overall cost). I don't want feature creep to delay this version or drive the cost up...

We'll be making up a PC board. The current draft does have an optional header that would supply the SPI bus signals (the LCD display is an SPI device). You could use this to drive a different display if you wanted to. I don't think there's going to be enough room on the board to provide a prototyping area. Again, a cost consideration. We can get MiniBoards (3.8" x 2.5") at a very good price, so that is the size the PC board will be. The schematic will also be published (under GPL or FDL - not sure yet how that'll work), so you're free to use it as the basis for your own PC board design that adds extra functionality.

I think many people would agree that spending $5-10 more on a beefier processor that would allow for more features is a good thing. I would at least choose a processor who's foot print (PIN OUT) allows for using a more advanced processor should it be decided that you want to upgrade.

Anyways, their are a ton of ways around upgrading the processor including installing a daughter board

I just like expandability in anything I make, be it my dark room, software at work or electronics projects

 

[Steve Toner]

I think you are unclear on the concept. This is intended to be a low cost, open design replacement for the Jobo CP* processors. The vast majority of users/buyers will have no interest in modifying it. To them, it's just a tool to help them create images. Although I'm willing to help people figure out how to tweak it (e.g., non-reversing option as mentioned above), it is not intended to be a platform for hackers who want to add super whiz bang features or turn it into a dishwasher. It's to be made available as a kit to keep costs down, and because there simply isn't enough of a market to try to build a business out of it. We're not doing this to make money.

 

[AlbertZeroK]

I just think not adding a comprehensive timer to something like this is a huge mistake. I know the Jobo's don't have it, but when you are able to add that type of feature with a 50 cent buzzer, a 20 cent jack (for a foot switch) and a little programming, it seems like it doesn't complete the need.

This is where my being a Senior Software Engineer would look at the needs of the client, not the wants of the client, to solve their issue. You want a Jobo replacement, but a timer is an integrated part of the process so why not include it.

 

[bblhed]

Love the whole kit idea, and what I see so far. I would like to know how you plan to attach the tank to the motor? I have been watching this thread just to see how you do that because, well, I'm looking for ideas. I was thinking about doing something like this with a windshield wiper motor from a car, but I still can't figure out the attaching the tank easily part.

As for those that want you to add features to your design, the design is open source so they should be able to customize their own units, right?

 

[Sal Santamaura]

Exactly the mindset...

...This is where my being a Senior Software Engineer would look at the needs of the client, not the wants of the client...

That's why most software is so "feature" laden typical users find it unwieldy. K.I.S.S.

 

[AlbertZeroK]

That's why most software is so "feature" laden typical users find it unwieldy. K.I.S.S.

Complexity is not what turns people off feature laden software, bad UI development is, but I think that gets off point of the original topic.

 

[TSSPro]

I just donated my Jobo ATL 2 to the university where I'm doing my MFA...You should have gotten pissed off at your processor sooner

 

[Steve Toner]

I just think not adding a comprehensive timer to something like this is a huge mistake. I know the Jobo's don't have it, but when you are able to add that type of feature with a 50 cent buzzer, a 20 cent jack (for a foot switch) and a little programming, it seems like it doesn't complete the need.

This is where my being a Senior Software Engineer would look at the needs of the client, not the wants of the client, to solve their issue. You want a Jobo replacement, but a timer is an integrated part of the process so why not include it.

I am well aware that timing is a part of the process. That's why I included a basic timer. Many people will find this meets their needs, though perhaps not their wants. I use a stopwatch myself, and find this adequate. I also have a Jobo ProTime 100 and find that it has shortcomings that keep it from being very useful to me. For those who want a more sophisticated process timer, I have proposed that as a separate product. There are several good reasons for having the process timer be separate:

• You don't burden those who don't need/want it with the cost
• If the unit is portable, it can be used to time tray/hand/stand processing (I do my hand processing in a different room from where the rotary processor would live)
• Including it would delay the release of the rotary processor
• It's harder to build a good process timer than you think

If you want to wire a jack in parallel with the reset button so you can use a footswitch, go right ahead. That is a trivial tweak.

Oh, and I am not aware of any "50 cent buzzer" that can interface directly to the output of the microcontroller. Could you provide a reference?

 

[Steve Toner]

Love the whole kit idea, and what I see so far. I would like to know how you plan to attach the tank to the motor? I have been watching this thread just to see how you do that because, well, I'm looking for ideas. I was thinking about doing something like this with a windshield wiper motor from a car, but I still can't figure out the attaching the tank easily part.

As for those that want you to add features to your design, the design is open source so they should be able to customize their own units, right?

Tony is working on the mechanicals. I'm just doing the electronics and firmware. I think he's planning to check in here soon with an update of his own.

As to your second question: yes, of course. Even the rude and pretentious ones.

 

[AlbertZeroK]

Even the rude and pretentious ones.

Hey, I resemble that remark.

The truth is, my comments are ment to open a can of worms and perhaps get the design changed in a way that benefits expandability. I think I've done that or atleast tried. So I'll get off my soap box.

 

[olleorama]

I don't know if it's any good to you, but I use a quite simple heating and controlling set-up that I think a lot of home-developers might be interested in. My setup consists of:

• a pid controller, I have one from CAL, which is pretty nice, and it was free for me, but there are good chinese units available from ebay (like this)
• a solid state relay (from the same seller if you like)
• a pt100 sensor
• a heating element (salvaged from a kettle, not running at full power)

As I only had to buy a pt100 sensor and managed to salvage the heating element from a kettle (not going on full power...) I got away really cheap. I use it in a sleeping mat insulated plastic box from IKEA. Lets me develop film and prints (in tubes) real cheap and very consistent.

I'm gonna buy or make a motor base any of these days. I like Usagis version, will probably copy it some how.

 

[michaelbsc]

• a pid controller, ... there are good chinese units available from ebay (like this)
• a solid state relay (from the same seller if you like)

These work well. I have one. I'm planning to build a controlled temperature cabinet for curing things.

The controller can also be purchased to use a thermocouple instead of an RTD, and in some of the listings there's a Type K thermocouple included. See here http://cgi.ebay.com/PID-Temperature...065?pt=LH_DefaultDomain_0&hash=item3367877699

MB

 

[olleorama]

Most controllers can use both I think, at least the two I have tried IRL (how's that for statistics? ). You just dial in type of sensor.

I have no idea which kind is most accurate, but would be glad to know!

 

[michaelbsc]

Most controllers can use both I think, at least the two I have tried IRL (how's that for statistics? ). You just dial in type of sensor.

I have no idea which kind is most accurate, but would be glad to know!

Sensor accuracy depends on the manufacture of the unit. Platinum has a well known characteristic curve, as do all thermocouple alloys. But a poorly manufactured style A is less likely to be accurate than a well made style B.

Having said that, this stuff is so far down the technology curve that it is nothing but a commodity. Both RTDs and thermocouples have been around longer than anyone on APUG. Thermocouples are well over a hundred years old. And commercially viable RTDs go back many, many decades. If anyone makes junky commodities, they go under pretty quick. The likelihood is that your standard measuring device is less accurate than your commercially purchased thermocouple or RTD.

If you really, really need a good measurement, and you're pretty close to sea level, then mix an ice bath made with ice from distilled water, and boil distilled water. That's zero and one hundred Celcius, and I promise you cannot afford a device that will accurately measure the minute changes from fluctuations in atmospheric pressure.

More important than accuracy is repeatability. If you set it at 100.5, and it always comes back to 100.0, you'll be OK once you dial in your process. but if you set it to 100.5, and it comes to somewhere between 100.0 and 101.0 randomly, then you'll never dial in your process.

MB

More reading

http://www.omega.com/temperature/z/thertd.html

http://www.omega.com/rtd.html

http://www.omega.com/thermocouples.html

 

[Steve Toner]

Most controllers can use both I think, at least the two I have tried IRL (how's that for statistics? ). You just dial in type of sensor.

I have no idea which kind is most accurate, but would be glad to know!

This document: Dead Link Removed has a good discussion of the tradeoffs between different sensor types (accuracy, stability, cost, linearity, response time, ...)

 

[olleorama]

Anyway, I hope this sort of cheap solution helps somebody to take the plunge. I was looking like crazy for a cheap solution, tried the aquarium heater route, tried asking industrial automation firms etc.

This is without doubt the best setup I can imagine for controlling temperature, for the price.

 

[O HanraHanrahan]

any progress?

 

[Steve Toner]

I haven't heard from Tony in quite some time. I sent him a breadboarded temperature control unit for evaluation (before I added the motor control) and never got a response.

I've had the PCB layout ready to go for quite some time, but have held off ordering because I haven't heard anything...

 

[Barry S]

Tony stopped responding to any queries on this project--so it's dead as far as I know. Steve, thanks for your work--it's a shame this stopped when the need is still there.

 

[holmburgers]

Too bad indeed, but hopefully he'll regain his interest some day.

Odd that there's no response, Dead Link Removedappears to be quite frequently updated.

It's easy to get frustrated when you've begun a project and people are expectant of results but life gets in the way. I wonder if this type of undertaking doesn't often get a lot of its initial momentum from wanting to please an "audience", when in reality, if conceived of in a vacuum the motivation might not be strong enough.

Slow & steady, all in due time...

 

[Richard Jepsen]

I'm impressed with your passion. As in life, simplicity is often the path to happiness. Slow down and hand process monocrome film in a holding bath of tempered water.

 

[rawhead]

I wonder if this project should've been a Kickstarter one, where somebody could've gathered enough initial funding from people like us to insure development costs and some money on the side, to keep motivation up.

Being constantly under the fear that my CPP2 could die any day, it's a shame.

 

[hoffy]

I wonder if this project should've been a Kickstarter one, where somebody could've gathered enough initial funding from people like us to insure development costs and some money on the side, to keep motivation up.

Being constantly under the fear that my CPP2 could die any day, it's a shame.

Afraid of what? If it dies, you fix it! If you can't get genuine parts, you make do with something else!

OK, I had a dead CPE2 - two issues

1. DC PSU had died
2. Motor burnt out

To replace the PSU, I disconnected it and wired in my variable Lab Power supply. So, I have 240V AC running the heating circuit as per the original design and 20V DC running the motor circuit.
To replace the motor, I bought a constant duty 24V DC motor and reduction gearbox. This did require a bit of messing about to get it fitted. I had to:

• Get a coupling made. The original motor has a long 6mm shaft, the replacement motor has a stubby ¼ inch shaft – lucky I have a good friend who has an engineering workshop
• Cut a hole in the back of the unit. The motor is about 50mm longer then the original, so it has to stick out the back.
• Remove a lot of the internal webbing where the old motor used to sit
• Enlarge the hole that the shaft comes through, to fit the coupling
• drill new bolt holes to mount the motor

Apart from getting the coupling made, all of the above took me a Sunday afternoon to complete (OK, granted, the Jobo sat on my bench in pieces for a good 10 months before hand!). The major hassle I had was getting a few UNF Screws for the motor (I got the motor through a mate – should have checked these details before I bought it! Seriously, who uses UNF these days? I could have sworn most of the world has been metric for a very long time, from an engineering stand point). It’s probably not the nicest job, but it should work.

So, the moral of the story is use what you have and do your best to get it working again!

 

[Steve Toner]

I wonder if this project should've been a Kickstarter one, where somebody could've gathered enough initial funding from people like us to insure development costs and some money on the side, to keep motivation up.

Being constantly under the fear that my CPP2 could die any day, it's a shame.

THe plan was to go to kickstarter when it was real enough (i.e., we had built working prototypes) as a way to make it available to the masses. As it stands, I'm out a couple of hundred bucks in parts cost. No biggie.

There was never any intent to make this a profitable venture. I suspect, as someone else mentioned, that he just got busy with other things and had to drop something.

 

[MaximusM3]

Judging from his Flickr page, it seems that he's dumped the Jobo for a Fuxi X100 and having a good time with it.