The WORST part of film developing is getting the temperature right. Here is my solution.

The worst part of developing film is getting the temperatures correct. The key to quality and consistency is temperature so it’s something that you need to get right each and every time. It takes me longer to get everything to 20˚C than it does to develop, stop, fix, and wash the film, hang it up to dry and pack everything away. I hate it.

So I made a spreadsheet to calculate the amount of cold and warm water needed to get my film processing temperature right and my life has been changed. You can get a copy for yourself.

The rest of this post is the story of the process of making the spreadsheet, which is less about making the spreadsheet and more about working out what the problem was and what I wanted a solution to look like.

Defining the problem

The hardest part of solving any problem is defining what the problem is. The second hardest part is finding a solution that people want to use. I design and make software, and after identifying the problem I ask the end-users how or what a solution might look like (that is, what the the user interface or work flow might look like or be) and then work out what needs to be done behind the scenes to make that happen.

Here is my problem:

Getting the temperature right is the WORST part of developing, it takes me hours and just thinking about it makes me want to do back to digital.

Ok, after we’ve got that out of the way we do some detailed analysis and discover:

1. My developing tank needs 1.3 litres of chemicals: I have a 2 litre bottle of Stop Bath, a 2 litre bottle of Fixer, and three 2 litre bottles of water for pre-wash and developer.
2. The 10 litres of processing chemicals are room temperature, ~24˚C. I need to get it to the processing temperature of 20˚C within about 0.1˚C.
3. I have six 1.25 litre bottles of water in my refrigerator that are chilled to around 7˚C to use for the tempering water bath.
4. I have an unlimited supply of water from my tap which at the moment is 21˚C to use for the tempering water. The temperature changes a couple of degrees each way with the season.
5. I use a 26 litre insulated drink cooler. It takes 6 litres to make the tempering water bath for the five 2 litre bottles.

Defining the solution

My ideal solution scenario:

I want to put the bottles into the cooler, pour in some chilled water and some tap water, and just leave it to sort itself out. When I come back later I want the five bottles in the cooler to be 20±0.1˚C and ready for processing.

From this description, I need to make <something> that takes as input:

• the volume of the bottled stop, fixer and water
• the temperature of the bottled stop, fixer and water
• the volume of the water bath in the cooler
• the temperature of the chilled water
• the temperature of the tap water
• the desired final temperature

The <something> needs to tell me:

• the volume of chilled water to pour into the cooler
• the volume of tap water to pour into the cooler

Thermodynamics and Heat Transfer

The math required to work all this out is not terribly hard, but with a couple of simplifying assumptions it can be made reasonably easy:

1. The stop and fixer have the same density and thermal properties as plain water.
2. The chemical containers transfer heat well and don’t themselves contain a lot of heat to cool down (or need a lot of heat to warm up). For this last assumption, plastic containers are preferable to glass or metal.
3. The cooler needs to be insulated to stop heat from transferring in or out.

The basic thermodynamic concept is that we are getting three different volumes (more correctly, three masses) of fluid, each with a different amount of heat, and allowing them to average-out. The heat can move between the water bath and the solutions in the bottle, so the actual liquids won’t average out, just the temperatures. The cooler needs to be insulated to stop heat from transferring in or out from the room.

On Thermometers and Measuring Temperature

For most practical purposes measurement reliability and repeatability are more important than accuracy. Not that being a couple of degrees off is OK, but ±0.3 will be fine as long as the thermometer measures the same every time.

During testing for this project I used three cheap (like $5) digital thermometers. All three of them are consistent (that is, if I measure something more than once I get the same reading) but I discovered that two give identical readings, while the third is about 0.75 degree lower. I compared all three to an old Kodak mercury thermometer and discovered the odd-one was the most accurate and the two that read identical were reading low. Would this matter?

Not for practical purposes, as long as I just use one thermometer. If I used the low-reading thermometers for my film developing, my solution temperatures would actually be a bit high (about 20.6˚C instead of 20.0) so I’d be slightly over-developing the film. If I was happy with the results I was getting then everything would be fine. If I thought my negs were a bit contrasty I’d maybe decrease development by 30 seconds or so. Either way, developing time needs to be adjusted to suit the equipment. A problem only occurs if I change thermometers…

Simultaneous Equations, Absolute Zero, and the Metric System

Three days later: I just spent a couple of hours last night solving simultaneous equations: I probably should get out more. I’m also now pretty good at converting from Celsius to Kelvin because thermodynamic calculations work from absolute zero.

Yes, I made a spreadsheet that takes as input the volume and temperature of the chemical solutions and calculates the volume of cold and hot water to mix together to make a water bath to reach the target processing temperature. It uses the metric system, get with the plan already (I’m looking at you, Liberia and Myanmar). It makes a few simplifying assumptions (eg, that all the liquids have the density and specific heat of water) but my tests so far indicate that it works to within 0.1˚C. Woo hoo.

Not a perfect solution

The biggest issue with getting accurate results are measuring the temperature of the solutions. I put six 1.25 litre bottles of water in the fridge to chill, and each one comes out with a different temperature by a couple of tents of a degree. I’m now using one 5 litre bottle for the cold water and will post how that goes.

Update 2022-04-12: it takes 1 to 2 hours for the solutions to stabilise temperature, and the assumption that the cooler’s insulation prevents ambient heat from transferring does not seem to be valid. I put 6 litres of cold water into the cooler and measured the temperature change after about 90 minutes, and discovered that it had risen by 1 degree. I have updated the spreadsheet to include this 1 degree factor and the results (after about 2 hours) are within 0.1 degree of the target 20˚C.

If you find this helpful please leave a comment. Thanks.