stepper motor / PC control of emulsion making

[rmazzullo]

another delivery idea....

Just another idea for discussion. This is more than likely overkill, but it might generate other ideas.

You could have a long thin stainless steel rod (say 1/8th inch diameter) pointed at one end, traveling coaxially (telescoping) inside a larger stainless steel tube (1/4" inner diameter) submerged in the emulsion mix. The larger diameter tube is swaged, or somehow reduced in diameter on the submerged end, so the thin rod inside will not pass through the end of the larger tube. The pointed end acts as needle valve, and the ingredient is introduced at the side near the top of the larger tube, under pressure. The top end of the thin rod protrudes from the top of the larger diameter tube, and is acted on by a servo (I keep thinking model airplane types, but there must be an industrial strength commercial version that does the same thing) so you can vary the length of the needle valve that retracts and for how long very smoothly and consistently. Once you have the exact "delivery profile" figured out, it could be repeated from run to run. I have to figure out how to seal the delivery tube at the top (or servo) end so the added ingredient doesn't come out of the top, yet allows the rod to move up or down relatively freely. It might be easier to have two small diameter tubes welded side by side, with the ingredient in one tube, and the needle valve riding in the other, and joined at the bottom. There would be a hole on the side of the delivery tube, where the needle valve would slide past in varying amounts. What would have to be worked out is the shape of the ingredient discharge at the end of the delivery tube.

Additionally, you could have two (or more) of these delivery tubes running very close to, but on opposite sides of a small turbine mixer (I haven't figured that part out yet)submerged and carrying different ingredients and the motor driving the turbine could be a hobbyists Moto-tool. For short duration mixing the Mototool should be ok, as the Mototool's speed runs up to about 20,000 rpms. You could even vary the speed of the moto-tool right along with the delivery rate and amount of the ingredients, and track the results in realtime with your monitoring electrodes...all done via a PC. You could have the mixer, mototool, delivery tubes and servo mechanisms on a small lab stand, perched above the mixing vessel. You could have a fast mixer delivering and mixing ingredients in a local area of the vessel, and a second, slower mixer submerged directly opposite in the same container, to mix everything together.

I have no idea if this will destroy the emulsion, or make it super-homogenized and destroy it, or make it better. Something tells me that this latest fit of madness can and will do all three at any given time, depending on any number or combination of factors, and / or how those factors are implemented. If I knew some sort of mathematical CAD program, I could probably model different physical / rate of delivery arrangements to get the best result for a given container shape (square or round bottom), but I do not. I will have to perform this experiment and find out. For all I know, all this gadget could be good for is making mayonaise.

If anyone has any thoughts, please feel free to contribute.

Thanks,

Bob Mazzullo

 

[Photo Engineer]

I've seen a lot of good ideas here, but the problem is, a good dose of actual emulsion making will go a long way in understanding the process.

I'm trying to come up with a way of describing an addition method that will work, but the simplest is just a syringe and the next up is a burette, and then above that is a peristaltic pump. Each moves upwards in complexity and cost. Any of these can be accurate and repeatable.

PE

 

[rmazzullo]

Hello PE,

I agree with you completely, and know that I really have to do the work to understand what is going on. I will admit that I do get a bit carried away with the "thought exercises" though.

I am hoping the Mees and James, and the Haist books will help explain more of what is happening, and how the technology has progressed.

Thanks,

Bob Mazzullo

 

[epatsellis]

just as an aside...

As part of a part time business, I design, repair and retrofit CNC routers. Instead of the pic/AVR approach, why not use one of the many off the shelf 6 axis s/w packages out there? They are run on pc class machines, and use step/direction interfaces directly from the parallel port, if anyone is interested, email me for links. it's relatively trivial to program in G code, and the ability to have coordinated moves is quite likely the easiest apporach to getting consistency. FWIW, the pricing ranges from free to ~$150 or so.

erie

 

[PeterB]

Peter;

I could do that from memory, but I'm considering the ramifications of postng something like that due to the fact that I have too much information on confidential items. I have to edit my memory so that I can satisfy you and not disclose too much. We are treading pretty close here.

I may diagram one single delivery system and leave out details to obscure some of the fine points.

PE

Hi PE,

After typing this reply, I have realised that you began another thread on automated emulsion making (there was a url link here which no longer exists). I'll let you decide which thread you want to put any reply to.

Obviously you have thought about how much info you are comfortable revealing without upsetting Kodak, and what I have written below seeks to address your comment: "I have too much information on confidential items. I have to edit my memory so that I can satisfy you and not disclose too much. We are treading pretty close here."

Make the descriptions generic enough so that you don't reveal any of Kodak's IP. Alternatively, you can always be specific with info that is already in the public domain, and processes for which patents have expired. You have highlighted an interesting dilemma here, do Kodak's secrets go with them to the grave ? Given that Kodak no longer manufacture B&W papers, perhaps they would be open to the idea of publishing their trade secrets which don't relate to their colour papers, or their film emulsions - I mean what do they have to lose ?

Clearly your manual process does not disclose IP, otherwise you wouldn't be running workshops on it. So document that process and we'll automate it for you, and hopefully we may come up with a different way to do it that doesn't infringe on a Kodak patent. (if our 'automated' way happened to do things in a similar way to a non-patented but secret Kodak proess, then that's tough bickies for Kodak !)

regards
Peter




One could argue that

 

[rmazzullo]

35mm film leader

Hello all,

This question is not intended to divert this discussion from the topic too far but...

Out of curiosity, does anyone know if 35mm film leader usually used for motion picture work, is supplied with or without a subbing layer? I do not know if the sprocket spacing is the same as on 35mm still camera film, but I did a google search and found sources in clear acetate and polyester, in 1000 (one thousand) foot rolls.

You know where I am going with this. I was just curious about the potential. That's all.

Thanks,

Bob Mazzullo

 

[Photo Engineer]

Fully automated delivery system

Here is my description of a fully automated precise delivery system....

I will describe only one system.

First, imagine that you wish to deliver 0.2 molar Silver Nitrate for 1 minute and then hold and then deliver 2.0 molar Silver Nitrate for 10 minutes.

This design will include 2 jars, one of Silver Nitrate at 2.0 molar and another jar of distilled water. It will include 2 pumps and associated pipes and automatic valves.

Ok, first, the Ag and water jars will have to have recycle lines and valving so that the pumps will be up to speed when delivery starts, so here goes.

1. Start Ag at 1/9th the rate needed and water at 9x silver and set for recycle.

2. At time zero divert both to the reactor thereby providing 0.2 molar Ag solution and at 1 minute shut off the Ag and allow the DW flow at 1/9th higher speed to push the exact amount of Ag into the reactor at exactly the same rate as previously for the entire combined flow, then shut off DW entirely.

3. Hold as desired while Ag is in recycle and the recycle ramps up or down to the proper rate for the next addition. DW is OFF.

4. At time zero, start Ag at full flow and DW in recycle at same flow. Pump Ag for the time required and then shut it off while DW is turned on to run to the reactor pushing the correct amount of Ag into the reactor.

This is done to prevent line volume from distorting delivery volume. The total exact volume must be delivered if automated. Any error will result in formula fluctuations or waste of Ag solution.

This system will also allow for ramped flow rates and ramped concentrations which gives the system described very great flexibility. It also leads to great complexity and an interesting problem for the programmer.

So, you guys considering automating delivery should be aware that when a formula calls for 20 ml of Ag solution in 1 minute, they mean exactly that, no more and no less. Variations will cause fluctuations in the resultant formula. Here is an example of a system that will carry out this accurate delivery.

Now, multiply this by the need for running salt (in control) or two salts (one in control and one not), dopants, antifoggants, gelatin, auxiliary materials such as stabilzers, acids and bases, and you get an idea of the complexity of the systems involved.

No, I'm not trying to discourage you and no, you don't need the above for a home system. What you do need is a system that can fill a jar and deliver it fully to the reactor in a specified time. And, you need this capability for every item you wish to deliver in a precise manner. I'm merely showing you one way I know it can be done.

Now, you run with it.

PE

 

[rmazzullo]

PE,

If you are able to (within the limitations you mentioned before) can you please shed more details on the purpose of recycling?

I am thinking this is to mix the required proportions to get the desired solution strength ahead of the timed injection into the reactor, but something tells me this may not be the case.

Thanks,

Bob Mazzullo

 

[Photo Engineer]

No, recycling is to get the pumps moving at the right speed before delivery starts and so as not to waste solution. If the pumps are not running at the right speed, then the delivery rate will ramp up with pump speed and delivery will not be at the right flow rate / unit time unless the flow is started prior to real delivery but is recycled back into the originating vessel. There is no mixing of DW and Ag before flow commences. Plumbing is designed to insure that.

PE

 

[wirehead]

So, you guys considering automating delivery should be aware that when a formula calls for 20 ml of Ag solution in 1 minute, they mean exactly that, no more and no less. Variations will cause fluctuations in the resultant formula. Here is an example of a system that will carry out this accurate delivery.

Now, multiply this by the need for running salt (in control) or two salts (one in control and one not), dopants, antifoggants, gelatin, auxiliary materials such as stabilzers, acids and bases, and you get an idea of the complexity of the systems involved.

No, I'm not trying to discourage you and no, you don't need the above for a home system. What you do need is a system that can fill a jar and deliver it fully to the reactor in a specified time. And, you need this capability for every item you wish to deliver in a precise manner. I'm merely showing you one way I know it can be done.

Now, you run with it.

PE

So basicly our semi-automatic-amateur-setup would replace a fairly small number of containers of chemicals only at the maximum concentration with a larger number of auto-syringes? In our case, you'd have two syringes, one with 0.2 molar silver nitrate and one with 2 molar silver nitrate?

 

[Photo Engineer]

So basicly our semi-automatic-amateur-setup would replace a fairly small number of containers of chemicals only at the maximum concentration with a larger number of auto-syringes? In our case, you'd have two syringes, one with 0.2 molar silver nitrate and one with 2 molar silver nitrate?

Yes, that is correct. The only 'automation' would be a plunger actuator for each and a timer for each with a push button to start the plunger. The syringes would have to have a head of air in it to force the last of the Ag solution out with air.

If you wanted to run anything else with either syringe, then you would have a 2 or 3 syringe system delivering the several solutions at the same rate. This method would not allow for control of vAg though but it would be close enough.

PE

 

[rmazzullo]

testing automated equipment...

Thanks PE,for the post describing a basic automated delivery system. This is blockbuster information.

Is it possible to build an emulsion mix, without the silver nitrate or other sensitizing agents, in order to fine tune the emulsion mixing, coating and drying 'sections' of an automated setup? It would seem obvious to test equipment that is built without wasting silver nitrate and other chemicals that are expensive, and light sensitiive.

Are the physical properties (viscosity, 'smoothness', etc) of an emulsion significantly affected by the addition of the light sensitive ingredients? If so, can these effects be compensated for when testing equipment builds?

Thanks,

Bob Mazzullo

 

[PeterB]

Here is my description of a fully automated precise delivery system....
....PE

Thanks for that PE. I've decided that the extent of my interest for now will be to try to understand as best I can the process behind one delivery system. I have so little spare time on my hands right now that I type this reply during an act of procastination and semi-awakeness !!!

Anyway if I could sufficiently understand the process to a level that I would be able to put it into an (Engineering) requirements specification, then I would be happy.
So my questions
i) Does this process bear any association with (there was a url link here which no longer exists) diagram ? I like to work with diagrams as well as words. As an Electrical Engineer, I design and communicate in circuit diagrams and program in LabVIEW (a graphical dataflow programming language)

ii)You mention two jars. One with 2 molar silver nitrate , the other with DW. Yet in replying to wirehead you agree with his interpretation that the two jars are really 0.2 molar silver nitrate and 2 molar silver nitrate. Which one is it?

iii)If you intend your process description to have a relationship to (there was a url link here which no longer exists) diagram then why don't you mention anything about mixing in the Br- ions ? Do you allude to the Br- when you write "Now, multiply this by the need for running salt (in control) or two salts (one in control and one not),"

iv) What do you mean when you say one salt is not (being) controlled ?

regards
Peter

 

[Photo Engineer]

Thanks PE,for the post describing a basic automated delivery system. This is blockbuster information.

Is it possible to build an emulsion mix, without the silver nitrate or other sensitizing agents, in order to fine tune the emulsion mixing, coating and drying 'sections' of an automated setup? It would seem obvious to test equipment that is built without wasting silver nitrate and other chemicals that are expensive, and light sensitiive.

Are the physical properties (viscosity, 'smoothness', etc) of an emulsion significantly affected by the addition of the light sensitive ingredients? If so, can these effects be compensated for when testing equipment builds?

Thanks,

Bob Mazzullo

Bob;

A part of my budget every year included water runs of 'emulsions' that were intended to test the timing of events when scaled up or down and testing new formulas for proper timing, so yes the equipment must be tested.

The emulsion itself has a different viscosity than each of the ingredients alone. This is primarily due to the binding of silver halide to the gelatin, a process called peptization. My students practice coating plain gelatin in the light, and when they being coating the same level of gelatin / emulsion, they are astounded at the drastic change in viscosity and coatability.

PE

 

[Photo Engineer]

i) Does this process bear any association with (there was a url link here which no longer exists) diagram ? I like to work with diagrams as well as words. As an Electrical Engineer, I design and communicate in circuit diagrams and program in LabVIEW (a graphical dataflow programming language)

ii)You mention two jars. One with 2 molar silver nitrate , the other with DW. Yet in replying to wirehead you agree with his interpretation that the two jars are really 0.2 molar silver nitrate and 2 molar silver nitrate. Which one is it?

iii)If you intend your process description to have a relationship to (there was a url link here which no longer exists) diagram then why don't you mention anything about mixing in the Br- ions ? Do you allude to the Br- when you write "Now, multiply this by the need for running salt (in control) or two salts (one in control and one not),"

iv) What do you mean when you say one salt is not (being) controlled ?

Peter;

Answers as follows:

1. This diagram is a simplfied schematic of what I describe in more detail with the Ag/DW example.

2. Wirehead was asking how to do it with syringes injecting solution into the reactor. In his case, dilution in-line cannot be done, so you need 2 syringes, one for each concentration of silver. The two jars, Ag and DW make up one delivery system. Another might be X and DW. Again, two jars but a second full delivery system. None of this is shown in the simplified schematic.

3. That would involve 2 delivery systems, and I did not want to go beyond one in the word description. Actually a real system has many many jars coupled in pairs with many many delivery systems under this type of overall reaction method.

4. Silver runs at 0.2 M or 2.0 M. Salt would be at 0.2 M or 2.0 M to match the silver, but the salt flow would be at a rate to achieve a constant vAg via feedback control. You can see that in the diagram referenced here. However, if the salt contains both KI and NaBr, the chatter induced by the feedback control would cause chatter in KI deposition and therefore it would cause banding in the emulsion of KI rich sites which is not good. For this reason, KI is often run at a contstant rate from a third delivery system made up of a pair of KI and DW jars which feed into the reactor so as to avoid any chatter in the NaBr line flow.

Now do you see how complex it can be???

So, to design an automated formula that is both modern and yet not automated to the extent used at EK will take a bit of my design skills before I can say - "this will work".

Posting things here and getting feedback is a way of my understanding what can be done without spending a fortune on equipment.

PE

 

[rmazzullo]

Bob;

The emulsion itself has a different viscosity than each of the ingredients alone. This is primarily due to the binding of silver halide to the gelatin, a process called peptization. My students practice coating plain gelatin in the light, and when they being coating the same level of gelatin / emulsion, they are astounded at the drastic change in viscosity and coatability.

PE

PE,

Is it possible to use another, (hopefully cheap) compound to bind to the gelatin, to get close to the viscosity / coatabilty of the finished emulsion? It might take some searching, but it would bring tuning the equipment closer to the ideal. Could it be as simple as adding a non light sensitive salt to the gelatin in place of the silver nitrate, for testing?

Thanks,

Bob

 

[Photo Engineer]

Yes, provided you had the means to test viscosity. I have found that this is really not needed and was not done at Kodak. The pumped or running materials were all almost like water. It was only the gelatin solutions that differed.

PE

 

[rmazzullo]

electrode information

Hello all,

This is a link to an article that gives some basics on electrodes much like what might be used here:

http://www.currentseparations.com/issues/14-2/cs14-2d.pdf

Thanks,

Bob M.

 

[rmazzullo]

one more...

some more background information on electrodes as used in electrochemistry; interesting circuit diagram.

http://www.collectionscanada.ca/obj/s4/f2/dsk3/ftp05/MQ64941.pdf

Thanks,

Bob M.

 

[rmazzullo]

sorry all,

One last document describing the basics of pH.

http://www.emersonprocess.com/raiho...ory_And_Practice_Of_pH_Measurement_199909.pdf

Bob M.

 

[rmazzullo]

possible low tech delivery system ?

Hello all,

In keeping with the 'low cost' aspect of using external controls for emulsion making, I have a suggestion for a possible mixing system / feedback loop based on one of the earlier posts regarding using lead screw actuated syringes.

Say you have 'x' syringes containing separate ingredients for an emulsion mix. Each syringe is actuated by a small variable speed motor controlled lead screw. Alongside each leadscrew, is another shaft that has "limit" switches mounted on it but can be moved along the extra shaft and secured as needed.

The action of the lead screw on one syringe moves an actuator (metal projection, etc) that passes and trips the limit switch (or switches), starting or stopping other syringes / steps. Each of those motor driven syringes have their own limit switches set at respective points to start, stop or delay other syringes, valves, etc.

If you have electrodes in the emulsion, the feedback signal could speed up or slow down one of the motor driven syringes to keep the vAg at a desired level.

The wind up is that you may not need a PC to control emulsion mixing on a small scale. An arrangement of switches, small relays, and motors, with a feedback signal supplied by the electrode(s) used to alter the rate of delivery of important ingredients.

There are shortcomings to this approach (need to hold voltage to a motor, and resultant speed, constant, hardware, etc ,etc) but, this is only an idea.

Thanks,

Bob M.

 

[Photo Engineer]

Bob;

It depends on what you want to make. I can make an Azo/Lupex like emulsion and a Brovira/Kodabromide like emusion with no control whatsoever. Early emulsions were single run and needed no control or were double run with no control.

So, the 'look' of the 20s - 50s could be done, as I have repeatedly said here, with no control. I don't want to lose sight of that fact.

PE

 

[rmazzullo]

Hello PE,

For the sake of discussion, assume you have a working, precise and repeatable delivery and feedback 'system'. Is it possible to tailor the emulsion using the vAg value as an 'indicator' (one vAg value indicates an emulsion with 'x' properties, while another vAg value indicates an emulsion with different properties)? This is also assuming you are working within an acceptable "range" of vAg values.

This may have been covered before. If so, can someone point me in the direction of the right thread?

Thanks,

Bob

 

[Photo Engineer]

Yes, I thought I mentioned that.

The range spans from cubic to octahedral to T-grain to K-grain depending on vAg. I believe the actual values are posted here in this forum.

But, you can also do this without control by picking the salt level. I gave some salt / pX/ pAg values elsewhere as well, maybe the same thread. At worst, without control, you get a dominant crystal and then a mixture of others which depends on what pAg value you drift towards and by how much.

PE

 

[rmazzullo]

Hello PE,

Yes, you did.

It was all explained previously. I lost track of where it was.

Thank you, regardless.

Bob