Getting more rolls out of C-41 chemistry / aka. Have you tried the Adox/Tetenal/Rollei/Bellini C-41 kit?

I've been thinking about doing an experiment for a long time - after every four 120/135 films, add 1 g of CD4 and make a pH adjustment if necessary. Since I do not have a color densitometer, my experiment will be based only on my visual and subjective impressions of the result.

lamerko said:

after every four 120/135 films, add 1 g of CD4 and make a pH adjustment if necessary.

Click to expand...

That will correct for 2 out of 3 things that happen when developer is used. The third is accumulation of halides. To handle this, only a proper replenishment system will work where some of the used developer is discarded and replaced by a fresh mix that's free of halides.
The amount of CD4 to add will be a gamble and 1g sounds like way too much for 4 films.

I chose this filling scheme precisely because of the halide accumulation. Of course, if I see that the density increases, I will adjust the scheme.

How does your scheme compensate for the effects of halide accumulation?

The main problem with accumulationcan is that they start to act as limiters. The increased amount of CD4 should compensate for this.
However, the big machines work with replenishing chemistry - apparently, halide accumulationcan be compensated even by Kodak and Fuji standards.

It's not that simple. Halide accumulation will affect the shape of the curve. Adding CD4 will not compensate for this!
The way the big machines work is like I said above; as part of replenishment, part of the used developer is drained and replaced with fresh replenisher. This replenisher has a low (or even no) amount of halides in it, so that the final tank mix that results after replenishment is again on spec.
Again, adding CD4 to a used developer cannot restore it to fresh performance. It will offset some of the effects of use and oxidation, but not all. The result may be good enough depending on what you're looking for; I'm not saying it won't work at all, but it won't work particularly well.

The volume of the replenisher relative to the volume of the working solution is too small to flush out the accumulated halides. Rather, a constant total volume is maintained.
Of course, there will be deviations - at least this replenishing will not be accurate. But I am not looking for full compensation - I know I cannot achieve it. But this method should give much better results than extending the time for 16+ films per liter

lamerko said:

The volume of the replenisher relative to the volume of the working solution is too small to flush out the accumulated halides.

Click to expand...

I'm sorry, I don't follow that statement. I think the manufacturer of the chemistry is in a better position to know this. Replenishment rates are dependent on the surface area of the film processed and that seems to me a perfectly viable way to actually account for halide buildup - which quite literally is really the reason why the system works this way. I'm not sure why you'd try and argue against this; it makes no sense.

lamerko said:

But this method should give much better results than extending the time for 16+ films per liter

Click to expand...

Potentially, if you hit the addition rate of the CD4 just about right. If you're off too far and/or err in the pH adjustment, the results will actually be worse than just reusing the developer. The risk of both failure modes is quite high.

lamerko said:

The volume of the replenisher relative to the volume of the working solution is too small to flush out the accumulated halides. Rather, a constant total volume is maintained.

Click to expand...

If you're talking about commercial C-41 replenishers, they actually DO supply enough volume to counteract the accumulated halides, etc. I've actually done (and seen) enough chemical analysis to know that this is true.

The way this works is: the developer tank solution is formulated with a significant amount of development byproducts. As I recall (might be wrong) it's something like 1.3 g/l as NaBr. As film is developed more bromide ion is released, obviously increasing the concentration. In order to get back down to the original spec concentration (~1.3 g/l NaBr) we add enough volume of fresh replenisher to DILUTE to this concentration.

As a note, although discarding some of the replenisher DOES remove some of the NaBr from the system, it DOES NOT lower the concentration. We have to actually dilute the developer tank solution to lower the NaBr concentration.

So the question is, how much dilution (via replenisher volume) is needed? As koraks has suggested, different replenishers might have different concentrations of halides in them. It actually depends on what the design spec for the replenishment rate is. Kodak has long had a C-41 developer replenisher known as LORR, andit is spec'd with an aim replenishment rate of something like 25 or 30 ml per nominal roll of film . If one is willing to accept that LORR replenisher contains essentially no halides (ie, NaBr) the presumption is that this is the lowest possible replenishment rate for a C-41 developer.

Stated a little differently, if one wants to lower the byproduct concentrations of a used C-41 developer tank solution back to the aim spec they would want to add a volume of about 25 to 30 ml per "roll" of (normally exposed) film. Note that this is sort of an average; it really depends on how much actual silver is developed. In the real world a pro lab would periodically process special test strips (aka process control strips) to determine if they need to make adjustments to their replenishment rates.

Just a follow up reply of the kit capacity of Ilford/Jobo vs Kodak/PSI. I processed a roll of Portra 800 using the latter, and results are just day and night compared to Ilford. The same process even came out a bit overdeveloped with slightly darker edge markings, but Portra 800 is known to handle it without problem. Same grading process with exposure lifted up a little. No colour correction done, only matching characteristic curves according to Portra 800 data sheet.

Again, it seems that kit capacity of Ilford/Jobo is very questionble. YMMV.

Good to hear you're seeing an improvement. Let's keep an eye on other user experiences with the Jobo/Ilford chemistry; it's good to be aware of a potential issue with it.

Owlch said:

only matching characteristic curves according to Portra 800 data sheet.

Click to expand...

Offtopic, but is this something you'd be willing to do a write-up on at a later date? I'd be interested to see how you go from the product datasheet to a correction curve.

koraks said:

Good to hear you're seeing an improvement. Let's keep an eye on other user experiences with the Jobo/Ilford chemistry; it's good to be aware of a potential issue with it.


Offtopic, but is this something you'd be willing to do a write-up on at a later date? I'd be interested to see how you go from the product datasheet to a correction curve.

Click to expand...

I use DaVinci Resolve to do grading, with NamiColor DCTL as the first IDT node. It automatically converts the linear scan into log space, where RGB alignment should be done within. After manually aligning RGB in both Dmin and Dmax (aka black and white levels) using shift, black point for Dmin and gain for Dmax, you get an image that is somewhat mapped to Cineon film log, but still does not actually match the toes and shouders of the film characteristics – as we all know they (almost always) don't have parallel RGB curves, especially in highlight/shouders:



And the image at this stage looks like this:


Next to do is basically to adjust black point and gain according to the toes and highlights of the curves. For instance, Portra 800 after pushing would have significantly earlier roll-off on G channel, so G gain should be decreased. Also, B channel at highlight is steeper, so increase B gain a bit to match. I admit it's a little subjective here, but every move is based on film characteristics, not by taste. Toes don't need adjustment here, becaue we can see Portra 800 has excellent parallel RGB toes. At this point the colour cast due to unmatched curves should be removed, and the cool overcast sky is rendered correctly. With a second node of Color Space Transform to make it right for display, it would look like this:


It's pretty much a graded result, except a little underexposed. Lift the global gain as wish, then done. I call it "density-referred" workflow because it respects the actual density distribution of the negatives.

Right, I see; thanks for illustrating that so well. I've never worked with Da Vinci, but the overall process I can wrap my head around. I was wondering in particular about how you'd get from a curve in the datasheet to a digital adjustment curve, but that part is now clear to me. I understand you use the datasheet more or less as inspiration to do a visual adjustment manually.

I think in fact the most important aspect of your process is the NamiColor conversion, which is conceptually interesting at the least very promising. I wonder if there's a similar solution that doesn't rely on DaVinci.

koraks said:

I think in fact the most important aspect of your process is the NamiColor conversion, which is conceptually interesting at the least very promising. I wonder if there's a similar solution that doesn't rely on DaVinci.

Click to expand...

It mainly does two things. One is log conversion and this should also be doable as long as you have linear scans. I know PhotoShop can do, but I forgot the how-to, sorry.

The other is to fit to Cineon base. That is it lifts the base from 0 to around 95 out of 1023 without clipping highlight. IMO this is just a legacy stuff of Cineon, and definitely not a must for modern film digitisation. I do it just for the sake of ease in Resolve because HDR output based on Cineon is much easier.

You can Google it and you should be able to land on NamiColor’s GitHub page. It has some useful info there. Also the concept behind requires a light source that somewhat resembles Status M to minimise layer crosstalk during scanning. There’s a project called Scanlight and the page is also very useful!

Thanks for the follow-up.
I think I see the Cineon 'hack' in your resulting histograms above and they made me worried a bit about highlight compression, which I noticed in the images you posted initially. This is in part what prompted my question, actually.

Owlch said:

You can Google it and you should be able to land on NamiColor’s GitHub page.

Click to expand...

Yeah I did before I posted me response above. Thanks! And I'm aware of Scanlight.

Probably the best advice is what Fuji Hunt says in the instruction leaflet of their 5L Film X-Press kir. It is that the developer can be reused. but for perfect results, it is one shot only. That said. acceptable negative is not the same as perfect, as it is always at least color and exposure corrected, even in optical printing. But please, don't try making records of reusing chemistry, when film is so much more expensive than C-41 chemistry. For manual tank developing. the real capacity is around 10 rolls per liter - maybe even 12 with 200 ISO or slower film, but no more than that.

When film was the only medium, and minilabs were in every corner, many of them were poorly managed. This was especially true of the labs in big grocery stores and in similar places, aimed to the general crowd. Still, those negatives, and even badly made prints were still acceptable for most common people. What I'm trying to say, is that unless it is the almost perfect, professional quality for exhibitions you are seeking, C-41 and RA-4 are quite forgiving processes.