Deficiencies in DIY C-41 Chemistry

[Spektrum]

@Rudeofus, thank you for the suggestion, I guessed that what you wrote was directed at me, because you indicated the Polish company Keten, which I am familiar with. I searched the offer of this German company Greisinger but I did not find any probes for photographic applications in their offer.
However, I did find such probes in the offer of a company SENTEK from the UK:

P11/HA Glass High Alkaline Photographic pH Electrode - Sentek Limited

Glass bodied pH electrode designed for high alkaline samples and photographic solutions.

www.sentek.co.uk

Price 127.50 British pounds. BNC connector.
I wonder why they don't provide information on what pH meters these probes work with.
Can it be any pH meter with a BNC connector?

EDIT:
Here is a much cheaper offer (£81.00) from another company but it is the same electrode / probe, manufactured by SENTEK:

111-702 - Omni Instruments

P11/HA Glass Bodied High Alkaline pH Electrode for a pH range of 0-14. Combination electrode with a Silver/Silver Chloride reference wire and ceramic junction. Eo point 0 to +/-25mV. Shaft dimensions (approx) 12mm diameter and 120mm length. BNC connector with 1m cable. Operating temperature 0 to...

www.omniinstruments.co.uk

 

[Rudeofus]

I searched the offer of this German company Greisinger but I did not find any probes for photographic applications in their offer.
However, I did find such probes in the offer of a company SENTEK from the UK:

P11/HA Glass High Alkaline Photographic pH Electrode - Sentek Limited

Glass bodied pH electrode designed for high alkaline samples and photographic solutions.

www.sentek.co.uk

Greisinger used to offer one, but does no longer. They do sell very decent pH meters, off which I have one,

I wonder why they don't provide information on what pH meters these probes work with.
Can it be any pH meter with a BNC connector?

It appears, that all these pH probes follow more or less the same principle and therefore produce the same output voltages. Therefore they should work with any pH meter which fits their connector. Note, that pH meters need to have exceedingly high impedance, regular digital volt meters won't work.

There appear to be two types of pH in the Market: gel probes and liquid electrolyte probes. After seeing the color of my KCl electrolyte after a few months of rare use, I prefer the liquid electrolyte probes. They are expensive, but typically last for many years of infrequent use.

In general I have the impression, that pH probes are made neither by Greisinger or SENTEK themselves, but sourced somewhere, which may explain the "creative pricing" of these things. Despite this, it is sometimes beneficial to get "the same thing" from a reputable dealer anyway.[/USER]
[/QUOTE]

 

[originalwinslow]

Update:

I mixed a new batch of homebrew c-41 (same recipe/method) and compared it more qualitatively with test negatives from kit c-41 developer. This version is certainly better, but the deficiencies are still obvious in the same ways. I am wondering if perhaps my CD-4 is slightly oxidized and perhaps another .1g or so would push it closer to par with the kit chemicals. Both developers are testing at pH 10.2 with my meter.

I should note that the "control" kit developer was not fresh. It has had 18 rolls run through 1.5L and "replenished" with itself at a rate of 30ml/roll. This replenishment method is insufficient to bring the developer back to original strength, but I've already discussed this somewhat haphazard method on another thread so not much need to go into it further here. This developer should still give a decent control for the purpose of this test. I will re-test in fresh when new developer arrives and judge the differences, which I am also curious to see.

Scene was shot in warm evening light at f8 with Kodak Color Plus film metered off a gray card. We can see in the color chart that the reds and greens are still under-saturated, and the film base is more yellow. The HB negatives are also thicker, despite receiving 15 seconds less time in the soup. So the developer is more active and yet not fully penetrating all the dye layers...?

I can't rule out weighing errors entirely, but since RPC was comfortable rounding measurements of a few of the ingredients in his formula, I don't think the .03g inaccuracy my scale sometimes exhibits would create such a strong variance. Except maybe in the tiny amount of sodium sulfite? Could +.03g restrict the developer to this degree?

***FYI Conversions were done in NLP with no edits applied/auto white balance. Not a 100% accurate method because the algorithm compensates for irregularities, but perhaps koraks con help us out with a more precise conversion to really see what's going on.

There appear to be two types of pH in the Market: gel probes and liquid electrolyte probes. After seeing the color of my KCl electrolyte after a few months of rare use, I prefer the liquid electrolyte probes. They are expensive, but typically last for many years of infrequent use.

This is excellent information. When my electrode inevitably dies I will contact the manufacturer (Milwaukee Instruments) and see if they have any KCL electrolyte probes. Mine is of the gel variety. I am outside the EU so the specific brands mentioned here are less accessible.

 

[Spektrum]

I've been eagerly awaiting the results of your tests in the new bath. Honestly, it's much better than before. I have one suggestion: I see you're scanning with a Nikon D810 digital camera in the ADOBE RGB color space. I saw this information in the EXIF data after downloading the photos to my computer. From what I understand, NLP strongly recommends that the files be in the sRGB color space. This can also affect the accuracy of the negative-to-positive conversion.

Using Photoshop, I converted the color space of your JPG files to sRGB. Then I saved them as TIFFs and inverted them using the free Photoshop plugin Grain2Pixel. I didn't make any color corrections at all. See my conversion results attached. They look slightly different than yours in NLP.

 

[originalwinslow]

From what I understand, NLP strongly recommends that the files be in the sRGB color space. This can also affect the accuracy of the negative-to-positive conversion.

Wow, I'm amazed I have never seen this guideline. What an oversight. I will adjust my settings and see what difference it makes. Thanks for the tip.

They look slightly different than yours in NLP.

I'm seeing a strong yellow color anomaly in the sky of your conversion of the HB. The information looks like it's clipping, maybe an export quality issue? I don't see it in the negative.

Overall, I'm not surprised the different conversion methods produce varied results. Aside from that nasty yellow streak...

 

[Spektrum]

Yes, you're right, the "home brew" has a yellow tint in the sky (especially in the clouds). As I mentioned, I didn't make any color corrections. Only the Grain2Pixel plugin had to make adjustments. I worked on a compressed JPG file - this also introduces additional variables... But my subjective opinion leads me to think that I like my conversion of your photo developed in HB the most. Of course, this is just my humble and very subjective opinion.
Also, remember that web browsers don't handle Adobe RGB color reproduction very well. The standard online is sRGB if you publish your photos. Always convert to sRGB because people view photos on different monitors, and besides, as I mentioned, browsers sometimes go crazy with a wide gamut. Even with a decent, calibrated monitor, everyone sees colors differently (if they are in a color space other than sRGB). And this is important if you want to see other people's color assessments of your photos.

 

[originalwinslow]

The HB photo appears to have a light yellow tint across the entire image. Color Plus being a very "warm" emulsion shot in very warm light could add to this effect across both images. A factor to consider. I'm most concerned about the strong yellow blotch in the sky, and confused as to why it would appear in your conversion and not mine. Probably not an artifact of compression since it does not appear your kit chem conversion. The other differences in the conversions are mostly density and saturation, but that streak is more objective. It's either present or it's not...I am inspecting the negative closely and do not see any discoloration that would correlate with how it appears in your conversion. Strange.

Only way to discern this is with a manual conversion.

 

[koraks]

but perhaps koraks con help us out with a more precise conversion to really see what's going on.

This would start with photographic both negatives side by side in the same shot. Only then a meaningful comparison can be made. Something like this:

Assuming that the negatives were indeed photographed with the same exposure settings (ASSUMPTION!!!!!), a qualitative comparison would come out like this:

That's a very quick & dirty inversion, straight linear with no intention to make things look good. It's just the same linear curve adjustment applied to both images. Again, ASSUMING the negatives were photographed in the same way, it's clear that the one on the left has more density and thus ends up much lighter. It also seems to lean towards cyan more so than the one on the right. Again IF this is representative, we can take from it that the problem is definitely not a lack of activity in the homebrew chemistry - in fact, it's the exact opposite: the homebrew chemistry seems to have been more active.

Then again, if the photos of the negative featured an aspect of auto-exposure that would have spoiled the broth, all bets are off.

I'd also like to focus your attention on this issue here: I've tried to color balance the KIT shot to something similar that you got from NLP, and applied the same curve adjustment to the other negative:

I hope this drives home the point that you CANNOT use NLP conversions as a basis for comparison. Don't ever, ever try to do that, because you're going to get lost in the woods with no chance of recovery. Such a comparison is totally meaningless; as you can see, tools like NLP give a creative spin on an image that may be pleasing, but any consistency goes out of the window.

See my conversion results attached. They look slightly different than yours in NLP.

Not 'slightly', they look totally different and your conversions also don't allow for a consistent comparison.

It's fine to experiment with different ways of negative inversion/color balancing for final output, but when trying to troubleshoot a color process, please be very careful.

@originalwinslow can you please photograph these two negatives in the same shot on a light table and post the result? Only then can we start with the comparison. The above is just an illustration of everything that can (and in fact, does) go wrong in the digital part.

 

[originalwinslow]

Assuming that the negatives were indeed photographed with the same exposure settings (ASSUMPTION!!!!!)

Yes, the reference images were shot on manual settings in the span of about 10 seconds to reduce the chance of any fluctuations in the natural light (cloud passing etc). They should all be almost EXACTLY the same exposure, as much as is possible outside of a studio lighting scenario.

the homebrew chemistry seems to have been more active.

Indeed, I mentioned this. The HB negative is more dense yet appears to not have the same color density. As I mentioned, the HB also received 15 seconds LESS in the bath.

I hope this drives home the point that you CANNOT use NLP conversions as a basis for comparison. Don't ever, ever try to do that, because you're going to get lost in the woods with no chance of recovery. Such a comparison is totally meaningless; as you can see, tools like NLP give a creative spin on an image that may be pleasing, but any consistency goes out of the window.

I knew I'd get a bit of chastising for this. NLP is used here as a basic reference for those who may also use the tool, despite the algorithmic magic it applies. Any automated program will create an unique interpretation of the information, hence the strong variations between results from two different conversion tools. NLP will (frustratingly) create slightly different conversion results even in identical scenes. This may be interesting to note for those who use differing automatic conversion methods. Thank you for your help with a qualitative manual conversion.

That being said, the difference in the red saturation/hue is similarly visible in your manual conversion vs the NLP interpretation.

@originalwinslow can you please photograph these two negatives in the same shot on a light table and post the result? Only then can we start with the comparison. The above is just an illustration of everything that can (and in fact, does) go wrong in the digital part.

I should have included this from the start, here is a side by side (shot/exported in SRGB).

 

[koraks]

Yes, the reference images were shot on manual settings in the span of about 10 seconds to reduce the chance of any fluctuations in the natural light (cloud passing etc). They should all be almost EXACTLY the same exposure, as much as is possible outside of a studio lighting scenario.

Ah, thanks; I meant the photos of the negatives though, but the 'scene as shot' conditions are definitely relevant, too. Thanks for the follow-up on that one as well!

NLP is used here as a basic reference for those who may also use the tool

It's not useful in that capacity. Sorry about the chastising; I know I've been very firm in my admonitions in this respect - deliberately so, although please don't take it personally.

That being said, the difference in the red saturation/hue is similarly visible in your manual conversion vs the NLP interpretation.

Yeah, but as you can see, in reality, NLP twists a couple of things into entirely random directions that really don't help in troubleshooting the chemistry.

In any case, something's really not in order with the DIY chemistry. I'm tempted to mix some test chemistry myself, but I've got too many other things going right now to try and replicate this problem. It's interesting though.

I should have included this from the start, here is a side by side (shot/exported in SRGB).

Thanks so much; it's really interesting - have a look at this:

This is a bit crude, but the principle here is to see what the qualitative differences are. The pattern we saw before still holds, but the density difference seems to be much less than in the earlier version posted, so it really does emphasize how important it is to photograph/scan the strips side by side. One caveat is that the photo was a little underexposed; the blue channel clips esp. on the bottom strip.
So overall the balance of the bottom strip is decidedly more cyan and more blue. The top strip is more red and more yellow. What's also interesting is that the top strip has a higher base density and this moreover seems to be uneven as well. There's a dark blueish grey band running along the center of the top strip; the edges look cleaner. This may be an artifact of the digital capture of the negatives (some flare), but it's something I'd have a closer look at if I were you to determine if there's perhaps indeed higher base density in the center of the top strip. If so, exhausted bleach and/or fix may play a role in this. Which would still not explain the color balance difference.

I don't have any other or better explanations yet than what we have discussed before, although your recent example does seem to suggest that the problem was not a fluke. There's a consistent difference between your commercial chemistry and the DIY approach. The question is still which one's the right one. Given what you said about the commercial stuff being reused + replenished, there's a very good chance that your DIY chemistry is actually performing closer to spec at this point.

 

[originalwinslow]

It's not useful in that capacity. Sorry about the chastising; I know I've been very firm in my admonitions in this respect - deliberately so, although please don't take it personally.

If someone is strong in their convictions it's oftentimes for a reason, no offense taken.

View attachment 402709
There's a dark blueish grey band running along the center of the top strip; the edges look cleaner. This may be an artifact of the digital capture of the negatives (some flare), but it's something I'd have a closer look at if I were you to determine if there's perhaps indeed higher base density in the center of the top strip. If so, exhausted bleach and/or fix may play a role in this. Which would still not explain the color balance difference.

Both are possible. The fixer was close to exhausted in that dev but still within acceptable range. My scanning setup is also not *perfect* with flaring. Something else to dial in.

***EDIT I just realized that I have negatively affected the comparison with a bad scan. I don't know why this didn't occur to me before. I did not scan the side by side in an entirely dark environment, and the strong ambient light in the room certainly affected the output. The other scans were done properly, but it is likely hard to critically judge color with the ambient light affecting color balance etc. Both negative strips have the same ambient light falling on them so judging differences between the two is still valid, but judging overall color balance is likely out the window. Regardless, the final conclusion probably stays the same.

I don't have any other or better explanations yet than what we have discussed before, although your recent example does seem to suggest that the problem was not a fluke. There's a consistent difference between your commercial chemistry and the DIY approach. The question is still which one's the right one. Given what you said about the commercial stuff being reused + replenished, there's a very good chance that your DIY chemistry is actually performing closer to spec at this point.

I figured it would be difficult to diagnose the issue causing the color difference. In my opinion, the lack of saturation/color shift leads me to believe it is still farther off spec than the used kit chemicals. I will re-test when the fresh batch arrives. It will be interesting to see the effects of degradation in the developer.

Frustrating to be so close and yet so far.

 

[koraks]

I did not scan the side by side in an entirely dark environment

That explains a lot; good catch! And yes, it influences the outcome, but doesn't change it fundamentally.

An A/B test with fresh, factory-made C41 developer should give some further insight.

Frustrating to be so close and yet so far.

Yeah, very much so!

 

[Rudeofus]

Update:
This is excellent information. When my electrode inevitably dies I will contact the manufacturer (Milwaukee Instruments) and see if they have any KCL electrolyte probes. Mine is of the gel variety. I am outside the EU so the specific brands mentioned here are less accessible.

If you can get reasonable deliveries from Milwaukee Instruments, then why not go with them. They call their probes with liquid KCl electrolyte "refillable". Note, that you can easily make that electrolyte yourself for very little money, but you have to make sure, that the KCl concentration of your electrolyte matches the concentration given in the probe description. Deviations from KCl concentration have direct impact on meter reading. You can cover up only so much through calibration.

PS: I keep reading about the general surprise, that your home brew is so much more active than your kit chems. You did write, that your kit chems have already gone through many rolls, so little surprise here from my side. We had a member here "Stefan Lange" aka stefan4u, who would use all his home brews only single shot just to stay within spec.

PPS: Kodak published cheat sheets for C-41 and other processes. If you can somehow translate your observations into "deviation from C-41 reference test strip", then you may be able to fine tune your home brew.

 

[originalwinslow]

They call their probes with liquid KCl electrolyte "refillable".

Great, thanks.

PPS: Kodak published cheat sheets for C-41 and other processes. If you can somehow translate your observations into "deviation from C-41 reference test strip", then you may be able to fine tune your home brew.

Most of the information here seems to be based on densitometer readings which I don't currently have access to, but good to have on hand for precise troubleshooting.

I am considering the culprit may too much sodium sulfite. It is possible my scale is imprecise at measurements lower than .5g, and realistically the deviation of this particular ingredient could be relevant. Bonavolta's site mentions this for the RA-4 formula, I would think the same principle applies here.

In the RA-4 developer the Sodium Sulfite acts to inhibit excessive dye formation and the Sodium Chloride is the restrainer. Since small quantities are involved, you may have trouble attaining consistency from one batch of developer to another if you use an Acculab scale to directly measure these, since its accuracy is only +/- 0.1g and that can (in this case) introduce a variance of 10% or more. If you don't have access to a torsion balance to weigh the sulfite and chloride and you are experiencing batch-to-batch changes in the developer's performance, you may instead wish to prepare percentage solutions as follows:

He goes on to mention that the Zone V developer, which has nearly 10x as much sodium sulfite as the PE recipe, restrains overexposure in the highlights but does not mention anything about restraining color dye formation or how that would have an effect on hue or saturation.

Sulfite in CD-3 developers is used to prevent excessive dye formation, and I've seen the same phenomena occur with CD-4, so it may be possible to use this developer to push color neg film and at the same time slighly increase the Sulfite content to prevent the highlights from getting blocked up (which this developer will quite easily do). In this regard, adding additional Sulfite could be likened to the effect of using an extreme compensating Catechol developer on B&W film - you get fully developed shadows without blocked-up highlights. This is still theoretical, and sometime I hope to have the time to sweat the details out with a 100-foot roll of film, a Macbeth Colorbrator, and my transmission densitometer.

I could try displacement measurement of the sodium sulfite to be more precise. Probably prudent to just buy a more expensive scale and eliminate any measurement errors as a possibility.

I'm not sure what else to tinker with besides this and the amount of CD-4. There is a possibility it has degraded slightly in the humid climate, even though it's only 5 months old/air tight. Should I store it in the freezer?

 

[Rudeofus]

Most of the information here seems to be based on densitometer readings which I don't currently have access to, but good to have on hand for precise troubleshooting.

Any film scanner, even the really awful cheap ones, is a color densitometer. I use vuescan and with the right key presses it shows me density of every image patch my mouse cursor just hovers over. I strongly assume, that vuescan is by far not the only software which can do this.

Probably prudent to just buy a more expensive scale and eliminate any measurement errors as a possibility.

Scales, which resolve 1mg, are not "more expensive". You will need such a scale anyway, so you can accurately measure the iodide. I use two scales, one, which goes up to 20g in 1mg increments, and one which goes up to 500g in 10mg increments. Both were really really cheap (read: less than US$20) and last for many years.

This is one of the issues with well reputed but old resources: they make a big fuss about things which are non-issues today. "You need an expensive scale which can measure fractions of a gram" LOL. "You need an expensive densitometer, ideally one for color" LOL. "Here are five recipes, which will cover your dark room needs" LOL. I am a proud analog photographer, but this doesn't prevent me from using a scanner plus software, cheap electronic scales, spread sheets to calculate and verify ingredient lists, and the internet.

I'm not sure what else to tinker with besides this and the amount of CD-4. There is a possibility it has degraded slightly in the humid climate, even though it's only 5 months old/air tight. Should I store it in the freezer?

Think about this cheat sheet again: why would Kodak publish such a resource, when labs bought their chems anyway?

Storage alters chemistry. Environmental conditions alter chemistry. A thermostat controlled bath at 100°F will have a different temperature, if your ambient temperature changes from 60°F to 100°F. Thermometers are quite inaccurate, even the good sensors claim +/- 1°F. At the same time your photographic process amplifies an image by 108 to 109, yet is supposed to get density targets down to 1% of full scale. Since nobody can reach such accurate and reproducible conditions, lab work started with a test strip and then some adjustments to the process, as shown in the cheat sheet.

In our case this means: we process roll after roll after roll, and if we detect a reproducible image defect over multiple dev runs, only then do we make attempts to fix them. We will be patient and account for process variations. We know, that our chemical ingredients are crap compared to the properly analyzed and sometimes custom made chems Kodak used. Process variations are no biggie, since we mostly scan, and post processing can handle massive color imbalances or crossover. We also don't worry about these endless claims of "blocked up highlights", which are trivially easy to correct these days, if they even exist with modern C-41 materials.

 

[koraks]

I am considering the culprit may too much sodium sulfite.

This would inhibit dye formation so I'd expect lower gamma. You got the opposite.

As @Rudeofus indicated, a precision scale is dirt cheap. I assume you already have one. If not...well, get one.
They're typically marketed for weighing gold and jewelry. I always have a distinct feeling they're mostly aimed at the market of green, crumbly, sticky 'gold' with a decidedly pungent smell.

I'm not sure what else to tinker with besides this and the amount of CD-4. There is a possibility it has degraded slightly in the humid climate, even though it's only 5 months old/air tight. Should I store it in the freezer?

As longs as it's stored in a dry, sealed container, it will last just fine for years.

 

[Spektrum]

Scales, which resolve 1mg, are not "more expensive". You will need such a scale anyway, so you can accurately measure the iodide. I use two scales, one, which goes up to 20g in 1mg increments, and one which goes up to 500g in 10mg increments. Both were really really cheap (read: less than US$20) and last for many years.

As @Rudeofus wrote, cheap jewelry scales work well, but look for products priced around $20, not $5.
The cheapest ones are the worst junk, and buying them is a waste of money.
Take a look at this scale, for example:

It's sold literally everywhere. Price: $5-6 USD. Sellers claim the scale's accuracy is 0.01g. I have two of these scales: one purchased in 2015 and the other recently. This scale, purchased in 2015, is slightly more accurate, but its actual accuracy is still 0.5g. Don't buy it.

Similar situation with this garbage:

It looks a bit more professional, but it's suitable for people measuring food weight while dieting.

These scales may be branded with different "manufacturer" logos, but they're still the same crappy scale produced in the same factory.

But the scale I bought for about $20 has started to meet my expectations.
The seller states the scale's accuracy is 0.001 g.

The scale can be calibrated. A calibration weight is included. Actual accuracy is 0.01 g, and the results are repeatable. 0.01 g is perfectly sufficient.

If you want to use potassium iodide, I wouldn't try to measure 1.2 mg (0.0012 g) with a scale. It's better to dissolve 1g of potassium iodide in 1000ml of distilled water and then use an insulin syringe.
This aqueous solution is quite stable if stored in a dark glass bottle (ideally wrapped in aluminum foil) in the refrigerator. It lasts about a year.
Here's a discussion on the shelf life of this solution: https://www.photrio.com/forum/threads/potassium-iodide.11655/

 

[Spektrum]

I'll just add my suggestion for measuring the pH of C-41 developer solution and using a pH meter.

Ready-made Milwaukee M10010 pH 10.01 calibration solutions are available for purchase.

This isn't a powder to dissolve in distilled water, but a ready-to-use 20ml solution with a pH of 10.01 at 25 degrees Celsius (77 degrees Fahrenheit). See the photo for how the pH of this solution changes at different temperatures.

By opening the solution sachet and inserting the probe, you'll have a reliable pH close to the one you need for C-41 developer. This way, you can check your meter for accuracy. The cost of one sachet is low – under $2 in Poland.


EDIT:
I have a question for @Rudeofus, @koraks, and other experienced users who use pH meters to determine the pH of their C-41 developer solution.

Available developer recipes accurately specify the pH of these solutions, but I haven't seen any information about the temperature at which measurements should be taken.
Since I dissolve the reagents at around 50 degrees Celsius, measuring the pH immediately after dissolving them won't be accurate. (?)

I'm wondering about this because, since the pH of this calibration solution changes so significantly with temperature changes, it seems to be an important issue.

 

[originalwinslow]

My scale is in the $25 range resolving down to .01g but my concern was whether it is able to accurately detect very low-weight measurements. It seems from the reaction here that this piece of equipment should be ok. I will still try displacement measurement to see if there is a difference. It's easy enough to mix 50g of SS in 500ml of distilled water and add ~38ml of this solution.

This would inhibit dye formation so I'd expect lower gamma. You got the opposite.

If the sulfite only affects the gamma of the image rather than the development progression of individual dye layers then maybe I am falsely attributing its function and therefore the potential result of an erroneously high sulfite mix. I noticed you posted this in Large Format Photography forum some years ago.

"Adding more sulfite is not a good idea as it inhibits dye formation and hence influences color balance. What's more, this will likely happen differently in higher and lower layers of the emulsion, so color balance will become a veritable mess."

Is this not a potential cause you previously mentioned? An incomplete formation of the lower (magenta/cyan) dye layers?

Any film scanner, even the really awful cheap ones, is a color densitometer. I use vuescan and with the right key presses it shows me density of every image patch my mouse cursor just hovers over. I strongly assume, that vuescan is by far not the only software which can do this.

I sold my old V500 flatbed to a friend very recently and noticed that, while I showed him how it worked, there was a densitometer feature within the Epson Scan software that I had previously ignored. I wasn't sure about the accuracy of this tool. I might have to buy it back from him to have access to this.

In our case this means: we process roll after roll after roll, and if we detect a reproducible image defect over multiple dev runs, only then do we make attempts to fix them.

Two batches may be insufficient to call it reproducible, but the same deficiencies presented themselves in both.

I'm wondering about this because, since the pH of this calibration solution changes so significantly with temperature changes, it seems to be an important issue.

Milwaukee advertises that their meters compensate for pH changes due to temperature. I do not know how accurate this compensation may be. Perhaps there is an ideal temperature for taking measurements specifically of photo chemicals. I have seen varying figures. I usually test @80F (room temperature here) as this is the temperature at which I calibrate.

 

[Mr Bill]

I have a question for @Rudeofus, @koraks, and other experienced users who use pH meters to determine the pH of their C-41 developer solution.

There was a time, long time back, when I was taking pH readings on around 40 to 50 color process samples per day. (This is very roughly upwards of 10,000 per year.) Mostly "seasoned" chemicals out of processing machines, including color developers, bleach, bleach-fix, and fixers, but maybe 5 or 6 fresh color developer mixes per day.

The most difficult were color developers, especially the seasoned ones. (Bleach, bleach-fix, and fixers were easy, by comparison.) We were using research-grade pH meters with 3-decimal place readings and the Kodak specified pH electrodes of the day (back then it was the Corning "triple-purpose pH electrode," and when these went out of production something else was recommended.) It was something of an art to reliably (and continuously) get good readings on the color developers. We would keep a pair of pH electrodes in use, swapping them out daily (one in use, the other "resting.") It would typically take between one and two minutes to get a stable reading on a seasoned color developer; if the reading was still drifting after two minutes then it' was time to do some troubleshooting of the meter/electrode system.

We used separate pH and reference electrodes, which simplified troubleshooting. (Fwiw I believe the Kodak document I referenced earlier gave a pretty complete rundown on the electrode troubleshooting, as well as cleaning procedures.) For someone using a combination electrode (pH and reference electrodes combined into one unit) the troubleshooting is more difficult. For example, how do you troubleshoot a clogged junction on the reference electrode? With separate electrodes you can swap out the reference electrode; you don't have this option with a combination electrode.

FWIW we used to get roughly a one to two year lifespan out of our pH electrodes. Eventually the response gets so sluggish that it's not worth dealing with any longer. To be clear I'm speaking of "glass bulb" pH electrodes, not the more modern isfet (or whatever it is) style.

 

[Mr Bill]

Milwaukee advertises that their meters compensate for pH changes due to temperature. I do not know how accurate this compensation may be. Perhaps there is an ideal temperature for taking measurements specifically of photo chemicals. I have seen varying figures. I usually test @80F (room temperature here) as this is the temperature at which I calibrate.

You should not trust the automatic temperature compensation. Best to get your developer samples to the spec temperature, then use the specific pH value given for the pH buffers at that temperature.

As I recall the Kodak color developer pH aim specs were given at something like 20 deg C (it might have been 25 deg C). So you would ideally have all of your gear at this temperature, and use the 10-buffer values given for 20 deg C when you calibrate your meter.

Given your nominal 80 deg F room temperature I would suggest to set your own developer aim value for that temperature like so. Start out with the original aim spec at 20 deg C (or whatever it is). Do what it takes to get your pH buffers, the developer sample and the electrodes to that temperature, more or less. Then read the developer pH at that aim spec. Next step is to determine your own developer aim spec for the temperature you want to use. So let everything stabilize at the new temperature (80 deg F?). Now recalibrate the meter, this time using the 10-buffer value specified for 80 deg F. Finally, read your developer sample at 80 deg F. Assuming that your developer had the correct pH value at the prior temperature spec you can now take your new 80 deg F reading as your new personal aim spec for 80 deg F developers.

Ps, fwiw pH 10 buffers are known to shift pH after a short exposure to air. So you would ideally use a fresh buffer sample when you calibrate your meter. (I don't recall seeing this mentioned here.)

 

[originalwinslow]

Ps, fwiw pH 10 buffers are known to shift pH after a short exposure to air. So you would ideally use a fresh buffer sample when you calibrate your meter. (I don't recall seeing this mentioned here.)

I have read this. I bought 300ml bottles of buffer solutions. They are air-tight but if any amount of air exposure causes pH variation then this method of selling/storing the buffer would be useless as they would begin to shift from the moment they are opened. The product description/reviews said this would not be an issue, but who knows. This is why Milwaukee sells the single use packets. Another variable to consider for pH accuracy.

 

[Rudeofus]

Two batches may be insufficient to call it reproducible, but the same deficiencies presented themselves in both.

If you have two sort of matching samples, then you do have something to work with. Make sure, that the necessary corrections are larger than the deviation between these samples. If your samples have more variation than their average deviates from the target, then try to make the process more consistent rather than changing the composition. It's a lengthy process, but the good thing is, that you can still fix all the way points in post processing, so you don't really lose images.

Milwaukee advertises that their meters compensate for pH changes due to temperature. I do not know how accurate this compensation may be. Perhaps there is an ideal temperature for taking measurements specifically of photo chemicals. I have seen varying figures. I usually test @80F (room temperature here) as this is the temperature at which I calibrate.

There are pH probes, which have a built in thermo couple to measure temperature, otherwise you will have to get a separate one. The separate one may be better, since it lasts forever, whereas you will have to change your pH probe every couple of years. My pH probes have typically lasted 5-10 years, but obviously with many orders of magnitude less use than Mr. Bill's.

There will be possible deterioration of pH probe itself, but you also have to look at its connectors. As I have mentioned before, pH probes have insanely high impedance, which means that even tiny amounts of dirt or moisture trapped in the connector will render it unusable until completely cleaned and dried. Connectors can also corrode, which can also screw up measurements.

 

[originalwinslow]

If you have two sort of matching samples, then you do have something to work with. Make sure, that the necessary corrections are larger than the deviation between these samples. If your samples have more variation than their average deviates from the target, then try to make the process more consistent rather than changing the composition.

Makes sense. If the variation between two batches made from the same recipe is significant then it would be impossible to attribute changes to an adjustment of the formula. This would be the case between my v1 and v2, as the differences were significant even though they exhibited similar deficiencies. If my v3 batch is consistent with v2 I will have a baseline to adjust from.

 

[koraks]

Makes sense. If the variation between two batches made from the same recipe is significant then it would be impossible to attribute changes to an adjustment of the formula. This would be the case between my v1 and v2, as the differences were significant even though they exhibited similar deficiencies. If my v3 batch is consistent with v2 I will have a baseline to adjust from.

Just don't forget to do the test with fresh, factory-made C41 developer.

Is this not a potential cause you previously mentioned? An incomplete formation of the lower (magenta/cyan) dye layers?

You're overlooking one very obvious issue: the image in a C41 negative is entirely made up of dyes. If there would have been too much sulfite, you'd see insufficient image density in one way or another. That doesn't happen. Overall, contrast on your DIY negatives is too high. So too much sulfite cannot be the cause of that. There's still the possibility that you have two problems going on at the same time of course. I can see how this might be the case if e.g. your scale is just very inaccurate. This can happen if they get very dirty inside, but in that case the calibration would likely fail, too, and you generally see odd fluctuations if you just leave the scale untouched; instead of a stable value, it will generally show erratic and changing values. Have you noticed anything along these lines? That's about the best I can come up with in terms of a systematic problem, other than pH measurement issues.

Also, can you post a photo of the dry chemicals? It's kind of hard/impossible to exclude any identification/labeling issues this way, but if there's a gross problem somewhere, it might be visible. Photos of bottles/jars/labels and especially of the powders/granules/etc itself, please.