View Full Version : Digital control of agitation in BTZS tubes and drums.

Dave Jeffery
16-Mar-2010, 13:31
I'm building a processor that will rotate both drums and BTZS tubes at the same time for 4x5 film development.

A programmable digital motor controller can be programmed to rotate a drum, reverse rotation and also sweep the motor back and forth. There is also some speed adjustment possible in both directions.

Ultimately I would like to program both drums and tubes to spin with the same cycle so lots of sheets of film can be processed at the same time. The agitation cycle must also allow individual tubes to be processed at different times. Both the drums and the tubes can be removed quickly but it seems like the initial agitation should happen for both the drums and tubes at the same time to get the development started. The drums can be pulled early or idividual tubes can be pulled early, and a duty cycle can continue for the tubes or drums left in the processor.

By adjusting the size of the pulleys I can make the drum turn at the same rotation rate as the tubes, faster or slower.

An array of light tight drain and fill flanges will allow a number of tubes and drums to be quickly drained and filled with stop bath so the time and effort needed to handle large drums and 6 tubes will be simplified greatly.

This will be for developing TMY-2 in divided D-23 and possibly 2-bath Pyrocat.

The most important idea is to avoid too much agitation which is why a number of people don't use rotarty processors. Now that the exact timing, speed and rotation control of the drums and tubes can be programmed with a digital motor controller I'm hoping that over-agitation can be totally avoided.

A few questions to gather some ideas.

1. Should the drums rotate more than one revolution in relation to the tubes say at 1 1/4 turns for the drum to 1 turn of the tube?

2. With agitation cycles timed to occur at regular intervals at 1/4, 1/2 and 3/4 points of development more cycles of this same interval can be programmed allowing development time to be staggered for various drums and tubes. One drums cycle could be started and just prior to the next agitation cycle starting, another drum or a number of tubes could be agitated separately and when the processor stops turning they are inserted into the processor for the rest of the cycle. Can anyone imagine why this wouldn't work?

3. The motor can be programmed to turn very slowly. Instead of just having the drums and tubes sitting at a complete stop would a very slow rotation be advantagious at any time during the cycle such as during the initial stage of development?

4. As a new light tight drum drain and fill design is in the works I can either modify and existing Jobo Expert drum or make a drum that holds individual PVC tubes. Is there anyone that thinks that the Jobo drums are a whole lot better than a large drum with individual PVC tubes?

5. Are there any design ideas that you think would be important for a drum that holds PVC tubes? Do the tubes need to be held to the inside of the drum or will having them roll over each other be OK and possibly help the agitation?

6. Can anyone imagine why film cannot be processed in drums and tubes at staggered intervals but at basically the same time as mentioned above?

I didn't have time to proof read this so please excuse any hacks. Gotta run.

Any feedback is appreciated.

Dave Jeffery
16-Mar-2010, 14:06
Can you imagine if this work well?

"Can someone email me the code for TMX in Mocrophen?"

17-Mar-2010, 03:05
any picture to demo your idea? I am interested. compared with you considerations, my roller seems like toy.

Robert Hughes
17-Mar-2010, 08:44
"Can someone email me the code for TMX in Mocrophen?"

itoa(DSGetHWLibErrorStatus(), tmpstr, 10);
lstrcat(tmpstr, " is error code; DSEngineOpen failed");
MessageBox(GetActiveWindow(), tmpstr, "Dscope Test Prog", MB_OK);
goto cleanup;


Any questions? :p

17-Mar-2010, 08:58
I think the hardest part of this project is definetly the mechanics. I don't believe Rotation rates are really that important. The Jobo used to have two RPM settings but they dropped that in favor of one setting (60 or 75 rpm I think?). Jobo did a lot of research on their products so I think they probably standardized on one setting for very good reasons. This makes the code very easy to write- in fact all you really need is a geared motor that will turn at 60-75 RPM and an H-Bridge or ESC circuit to change it's direction.

When it comes to making a new light-tight drain and fill drum out of PVC.. well good luck with that. I have not been impressed with the quality and ease of use of the PVC BTZS tubes that you can make it home. Not worth messing with to me if Jobo drums are still available.

Dave Jeffery
18-Mar-2010, 00:19
"itoa(DSGetHWLibErrorStatus(), tmpstr, 10);
lstrcat(tmpstr, " is error code; DSEngineOpen failed");
MessageBox(GetActiveWindow(), tmpstr, "Dscope Test Prog", MB_OK);
goto cleanup;

Any questions? "

Good one Robert!!!!

Here is the Arduino open-source board


The code is available for Windows, Mac And Linux and some control code is here


Motor Control

Eagle files for an L293-based Arduino motor driver shield.

Code the parallax HB-25 motor controller item code #29144.

DC motor control using an H bridge.

Adafruit servo/stepper/motor shield - controlling 2 servos, & up to 2 stepper motors or 4 DC motors. (Schematics, layout, and library).

Datasheet for the L293 motor driver IC. Blog diagram of how to connect the L293 to an Arduino. (Peltier Elements are working with this IC, too)

TankShield - a tank tread shield for the Arduino that shows how to turn an arduino into a robot

DC Motor Control - Speed and Direction

DC motor control using L293D (with and without PWM).

directional motor control with a L293D

Library for Pololu Qik2s9v1 controller

Schematic for driving DC solenoids, using a transistor connected to an Arduino pin.
A Solenoid Tutorial.

Mega Servo
Control up to 12 servos with the MegaServo library.

The code is not as out of reach as it may seem. I just found out that Photo Therm programs control chip for their processors so the idea was sound.and there seems to be a very active open-source community for robotics. It's $100 for a Phototherm chip.

The intial idea was to build a unit to turn an number of tubes. Then it seemed logical to turn drums and tubes together and a system that does this is somewhat easy to make. The idea then is to develop a lot of film at once which means getting the chemicals in and out of the tubes and drums efficiently. A motor could simply reverse back and forth but using a cheap digital controller would make it easy to set the amount of agitation fairly precise and in a repeatable fashion and hopefully avoid overagitation. Now it turns out that in order to make the process more automatic the chemicals need to be pumped in and out of the drums as well.

One would probably be better off just buying a Jobo ATL or PhotoTherm SSK processor as recommended by Ian Mazursky in this thread.


Thanks for that post by the way numnutz.

I avoided the automatic processors as they were expensive and I was concerned about a part failing and the unit being useless or very expensive to fix. I was not familiar with the PhotoTherm SSK at all. Now that I look at those designs it makes a lot of sense to rotate the film in a drum and pump the chemicals in and out.

I'm in a quandry right now and have to do some more research. I am still trying to find the Hurst motor that reverses with and interupt and consider the drum draining in a light trap again. The idea of doing a lot of sheets at once seems good but if much time is needed to swap chemicals during the process a person could get a lot more done doing less sheets in an automatic unit.

I really enjoy building things though so it's not like that is a huge drawback.

>Not worth messing with to me if Jobo drums are still available.
Thanks for that. I've done a lot of work making underwater housings etc and have worked with PVC and epoxy a lot. No sense making something if the Jobo drums are OK though. I read a respected member here mentioning that he doesn't use Jobo drums just pvc tubes in a large drum which I thought would be a lot cheaper and easier to make. The reason I didn't buy a Jobo Processor was due to the weakness of the lift and motor and I thought the ATL was too expensive.

>I think the hardest part of this project is definetly the mechanics.

Sealed bearings are 10 for $20 and the seals $5 each. Off the shelf 5/16" SS shafts 3" in length are $3 each. I would use bearings spaced apart on the side with the drive pulleys and simple bushings on the non drive side. The frame would be aluminum angle with about 2 1/4' strips of cutting board (great material) screwed to it that the bearings and seals would be set into. Anywhere there needs to be a lot of strength material can be removed and replaced with epoxy fiberglass. I still have a lot of thick woven fiberglass matt around which makes casting parts very easy. I need to look at the belts and pulleys more but a 1/2" belt seems fine. The belt length is not critical as the motor mount is easy to make, and it's base can be ajustable.

A couple more days of research is needed for me to decide whether just to make a simple roller system, build a more automatic system or just buy a good used ATL or SSK. The availability and cost of parts will determine a lot. The advantage of homebuilding is the ability to keep the system running for as long as I need it. Finding cheap small pumps to move chemicals may aslo be a big factor.

A simple roller system should be easy to make and I can do some sketches for you if you like. The Jobo drums could have fittings for the roller system fiberglassed to them or the drive shafts could be make with a compatable end for the Jobo.

I have to study things a lot more to see how much time and expense is involved for a more robust system.

Thanks for your input Domaz!