Divided D-23, D-76: Can they be made truly divided ?

[Ken Lee]

It has been pointed out (http://unblinkingeye.com/Articles/DD-23/dd-23.html) that Divided D-23 is not really a truly divided developer: initial development occurs in Solution A, and continues in Solution B.

The formula is given as follows:

Solution A: 75 g Metol, 100g Sodium Sulfite, water to make 1 Liter
Solution B: 2 g Borax, water to make 1 Liter

http://www.kenleegallery.com/images/forum/DividedDevelopers.jpg

Why can't we put the Sodium Sulfite in Solution B instead ? That's the same as D-23 itself. Is it because the film will develop in Metol alone ?

Is Divided D-76 not truly a divided developer either ?

[Jay DeFehr]

Both DD23 and DD76 are divided developers, in that they are divided into two separate stock solutions. They represent one of 2 types of 2-bath development. In this type, some development takes place in the first bath, and is rapidly accelerated and completed in the second bath. In the other type, no development takes place in the first bath, usually due to insufficient pH. Both types are effective, but the first type is more flexible, allowing expansion development.

You could make a D23 type 2-bath developer of the type in which no development takes place in the first bath by using enough sodium bisulfite in combination with the sulfite to lower the pH to a level at which the metol is inactive. Some sulfite is needed to preserve the metol, or the stock solution would quickly oxidize, and metol is only slightly soluble in water, so some sulfite aids in its solubility. You could try this:

A solution-

water 750ml
sulfite 100g
bisulfite 15g

water to 1 liter

B Solution-

water- 750ml
borax- 20g

water to 1 liter

The above is just a guess, as an example of one approach; bring down the pH of the A solution with bisulfite, and restore it in the B solution by increasing the borax. Alternatively, you might use 2g of carbonate instead of 20g of borax. A little testing will tell you what's required.

My question is: why do you want to convert DD23 to the other type of 2-bath developer?

[Ken Lee]

Thank you !

You make a wonderful point, and you ask a very insightful question indeed.

Ed Buffalo says the following (http://unblinkingeye.com/Articles/DD-23/dd-23.html), which has really grabbed my attention:

"So essentially, you develop in solution A until your high values are almost where you want them, then you place the film in solution B and develop until the shadow values are where you want them. If you are not experienced at development by inspection, you may have to do a little experimenting to get the times just right."

With Divided Pyrocat, where development only occurs in Solution B, I have had to return the film to solution A on a few occasions, to further develop the shadows - so it has often ended up more like a 2-Bath process than a fully divided one.

Using an Infra Red Viewing Device in the darkroom, it should be fairly easy to test Ed's 2-bath approach. I shoot sheet film and try to get each negative as close as possible to its desired appearance - before scanning, printing, or enlarging. The more adjustments we can do in the analog phases, the better.

[Jay DeFehr]

Ken,

The two types each have their advantages and disadvantages. The advantages of the development in A bath-type are the flexibility of expansion development, and the even development. I personally favor this type. The advantage of the other type of 2-bath is that the A bath generally is better preserved and longer lasting, and since development occurs so rapidly, they produce very sharp negatives. This type is also less prone to blown highlights. The other side of that coin is that they often compress values into a muddy tonality. I like some sparkle in my negatives. The other downside to this type is the potential for uneven development, especially with staining developers.

I'm currently working on a 2-bath developer of the develop in A bath type, for ultrafine grain and increased emulsion speed. It's very easy to make a 2-bath developer work, but much more difficult to make it work better than a standard developer. 5 or 6 years ago I experimented with ascorbate 2-bath developers, and was very happy with my results. My favorite was a metol/ascorbate/carbonate developer; no sulfite, no restrainer. I made the A bath up from dry chemicals before each session, and threw the remainder out after the developing session. I called it a Single Session approach, and I think it's very similar to the way Sandy works with his developers. I could develop a lot of film in a session without having to worry about replenishing, extending development times, or my developer quitting on me. I also didn't have to store a stock solution or wonder if it's still at full potency after the first session, because I just tossed it at the end of the session. It works great with my Jobo processor, because the solutions are pumped in and out of the containers for each batch of film developed, and everything is automated. I just fill the containers to begin, and then toss them after the session. The limitation on its practicality is that one should have a fair amount of film to process in a session- 10 rolls/8x10 sheets, or so is a good minimum. For smaller runs, one shot developers are more practical, for me. If you're interested, I still have a few of those formulas I was playing with. They each have their charms.

[Ken Lee]

Thanks for the further explanation !

And thanks for your offer. For now, I plan to fiddle with Divided D-23, since I already have the ingredients on-hand. All 3 of them that is.

Of course, if you feel that any of your earlier formulas represents a compelling improvement, I'd love to try them too. I'd love to hear more about their "charms", and I'm sure that others will too.

These days I shoot FP4+ and HP5+ in 5x7. I'm less concerned with grain and acutance, than linear curves and good tonal control.

[Ken Lee]

Actually, the Divided D-23 formula I use is Barry Thornton's modified version, described here (http://www.awh-imaging.co.uk/barrythornton/2bath.htm).

[David Karp]

Ken,

Thornton changed his formula from 6.25g/L of metol to 6.5g/L after writing that article. The new formula is contained in his book "The Edge of Darkness." I corresponded with him after his death and he wrote to me that the formula in the book was his most current version and that it made a slight but noticeable difference. Whether .25g/L really makes a difference, I don't know, because I learned of his formula from the book and never tried the version from his website.

Just a heads up. Judging by what I have seen of your work, it seems as if the formula is working just fine for you!

[David Karp]

As I think about it, it probably does not matter for you, Ken, because you are watching the development. It might make a difference when developing by time in trays or tanks, in the dark.

[Keith Tapscott.]

Ken,

I corresponded with him after his death and he wrote to me that the formula in the book was his most current version and that it made a slight but noticeable difference.Was it through a séance or by Ghost-Post? Sorry David, I couldn't resist. :D

[Jay DeFehr]

Ken,

The ascorbate 2-bath developers I experimented with were more like DD-76 than DD23, being superadditive developers. I used ascorbate instead of hydroquinone to take advantage of the fact that the ascorbate/phenidone or metol pairs don't require the presence of sulfite for their superadditive effects. Since ascorbate is a surface developer, the opportunity to improve acutance without sacrificing grain intrigued me. I decided to try splitting up the superadditive pair between the two baths, with metol in the first bath and ascorbate in the second. This approach was something of a compromise, since it required I use some sulfite in the first bath to preserve the metol, and allow for development to begin there, so hydroquinone could also be used. I think this is essential for developing brilliant highlights, and glowing mid tones. These ends were furthered by using a carbonate B bath instead of borax, and at a higher concentration. The carbonate B bath also kept development times conveniently short; when I'm developing a lot of film, I don't want to develop each batch for 15 minutes, and I suspect rapid forced development has a lot to do with the speed increases these kinds of developers produce. This formula is significantly different from the ones in your table above, and may be of no interest to anyone, but here it is anyway:

M/C 2-bath

A

Sodium sulfite 25g
metol 2.5g
Water to 500ml

B

sodium carbonate 5g
ascorbic acid 5g
water to 500ml

Development time in the A bath determines contrast. Develop between 3 and 5 minutes (for scenes of normal-high contrast) at 70F with continuous agitation. Longer development times will produce more contrast. Development in the B bath is almost instantaneous, continuing only until the metol absorbed in the emulsion is used up in the reaction with the ascorbate and carbonate. 1-2 minutes seems to be sufficient, but 3 minutes won't hurt anything.

2-bath developers operate on the principle of excess, so they're very forgiving of sloppy formulation, penalizing mostly in the form of inefficiency, which does not reflect on image quality. This formula is not optimized, just a proof of concept, but the concept is sound because the results are very good. I've never seen another formula divided up this way, so I wasn't sure what to expect. I think the approach has potential. Like you, my primary concern is for consistent results, and excellent gradation, and any developer I test is compared to 510-Pyro, which sets a very high standard for consistency and gradation.

Here's another idea I've never seen published, for a staining 2-bath developer of the type in which development begins in the first bath.

A

metol 5g
sodium sulfite 100g
water to 1 liter

B

sodium sulfite 0.25g (2.5ml of 10% solution, or 25ml of 1% solution)
pyrogallol 0.5g (5ml of 10% solution in glycol is very convenient, or 5ml of Hypercat A solution for a catechol variant, in which case the sulfite should be omitted)
sodium carbonate 2g
water to 1 liter

As you can see, the A bath is almost D23, and D23 could be used instead. Although D23 is a high sulfite developer, giving very fine grain and low contrast, not enough of the sulfite is carried over into the second bath to prevent staining. In this approach, the first bath is re-used for several films, but the second bath is used one-shot. The second bath contains the pyro, with enough sulfite to preserve it during development, and enough carbonate to force development to a rapid conclusion. That's the theory, anyway, I've not actually tested this developer. Catechol could be substituted for the pyro in the B bath, but the catechol would need to be increased, relative to the sulfite, since catechol is far more sensitive to sulfite than pyro is. The A solution of Hypercat would be a good option. If you want to test the above formula, you can be a pioneer! I won't be able to get into my darkroom for a few weeks, but when I do, I'll give it a try.

[Ken Lee]

- Wow -

[j.e.simmons]

Jay - good to see you back in the lab.
juan

[David Karp]

Was it through a séance or by Ghost-Post? Sorry David, I couldn't resist. :D

Oops. I meant before his death. :o

Just shows that I ought to proofread, always.

[Jay DeFehr]

Thanks, Juan. I've been enjoying myself when I've been able to find the time to get in there. Even as I type, my faithful companion and lab partner is running a test for me. It's nice to have some help!

[Jay DeFehr]

Julia has confirmed my theory is sound, and the 2-bath staining developers work, both the pyrogallol and catechol versions. When the film dries, she'll scan it so I can evaluate the developers' characteristics and contemplate adjustments. I'm very curious! I think this approach has potential.

[Jay DeFehr]

Here's an example from the catechol version of the 2-bath described in a previous post. The film is Fuji Acros. This is a quick scan without corrections or sharpening of any kind. I think there is potential here.

[Jay DeFehr]

On reflection, it seems to me this approach is not valid. It's not really a 2-bath developer so much as two consecutive developers, since both baths will develop film on their own, thereby negating the mechanisms responsible for the benefits attributed to 2-bath development. To secure those benefits the developer would need to be converted to a 3-bath developer! Sulfite + metol in the first bath, pyro or catechol in the second, and alkali in the third. Ludicrous. I'l give it a try and let you know what happens!

[Ken Lee]

"On reflection, it seems to me this approach is not valid"

Which approach ? The type where development occurs in both solutions ?

To me, the appeal is what Ed Buffalo referred to: being able to control high values and low values more or less independently, or at least sequentially.

What I have seen so far with DD23, is that nothing much occurs in Solution B. At least, I can't see much when I look with the Infra Red viewing device. I'm not convince that there is much actual difference between DD23, and D23.

But D-23 looks pretty nice on its own !

[Jay DeFehr]

Ken,

If there's any advantage to sequential development, as distinct from 2-bath development, it's not obvious to me. With DD23, development in the first bath proceeds as usual, beginning in the highlights and working towards the lower values. When the highlights have attained desired density, the film is removed from the first bath, and placed in the second, where the film is starved in the highlights, but continues developing in the shadows (in theory). This asymmetrical development is the basis of compensation and the chief benefit of 2-bath development. In sequential development, the film is simply developed in two different developers, in sequence. There is no asymmetrical development, and prolonging development in either bath increases contrast. The film has to be starved of developer in the highlights, at some point for there to be compensation. Towards this end, a staining developer of type C, in which development begins in the first bath, rests in the second, and is completed in the third is being tested, even as I type. It's ridiculous, I know, but sometimes I just like to test wild premises for the fun of it. In this developer, the first bath is the metol + sulfite, after which the film is rinsed to stop development and wash away the metol and sulfite. The second bath is a solution of pyro , in which no development takes place, but the pyro is absorbed into the emulsion. The third bath is a sodium carbonate solution, which should force the development of the absorbed pyro and form a stain. So, no superadditive effects, but there should be some compensation and some stain, and a lot of nonsense.

Julia just informed me the film is developed, and the images look correctly developed, with a green-brown stain. 123 Pyro! Although wildly impractical, this does constitute a fairly unique developer; it uses metol and pyro, but not as a SA pair, begins as a high sulfite, fine grain, non-staining developer and ends as a compensating, acutance, staining developer! It could be made into three stock solutions with good keeping properties, and all three solutions could be re-used, but that's about the end of its practicality. But 2-bath developers are so tired. If two is good, surely three must be better, right? Hahahhahah!

[Ken Lee]

"It's ridiculous, I know"

It's not ridiculous. A similar idea had occurred to me, but I lack a grasp of photo chemistry. When using Divided Pyrocat HD, I repeated the process a few times. This gave the negative greater overall density (less extreme compensation you might say), and reduced the muddy look.

It ended up being 6-bath. :rolleyes:

How about a divided developer, followed by regular developer: develop the shadows fully, while restraining the high values - and then boost the high values to taste.

[Bob McCarthy]

"On reflection, it seems to me this approach is not valid"



But D-23 looks pretty nice on its own !


Actually I came to this same conclusion. I started testing Sandy king's two bath and my subject lighting wasn't so extreme as to take advantage of the process. I did ,however, develop some sheets to completion in D23 1:1 and got marvelous negatives. D23 isn't a listed developer for TMax (400 in this case) on the massive or elsewhere I could find. "bout 9 minutes will get you in the ballpark.

bob

[Ken Lee]

To that point, I'm beginning to wonder about the importance of film developer for Large Format in general. I can see how grain, sharpness, acutance, and edge effects could be important for roll film or 35mm, but for scanning big sheets, it seems that the only real considerations are film speed, linear tonality, and contrast control. Am I missing something ?

Here's a test of some 5x7 HP5+ with the Barry Thornton DD-23 - which seems not much different than D-23 when you watch it develop under Infra Red. It seems quite linear to me.


http://www.kenleegallery.com/images/forum/HP5+ DD23 4 min 4 min ISO 400.jpg
Test shot: HP5+ in DD-23, 4 min/4min ISO 400

I may go back to simple D-76, or even simpler: D-23, with its 2 ingredients. By the way, what's the advantage of D-76 over D-23 ?


http://www.kenleegallery.com/images/forum/DividedDevelopers.jpg

[Jay DeFehr]

6 bath? Hahahhaha. The power of perseverance! Still, I remain dubious regarding the utility of a 3-bath developer. What advantages could it possibly confer? Since staining is at odds with compensation, the extra bath required to produce the stain seems superfluous, if compensation is the goal.

I'm still a little fuzzy on how scanners see stained negatives. I don't think the stain has the same effect on contrast in scanning as it does in printing with either graded or VC papers, but I suspect the response is closer to graded paper than to VC paper. The dye image color information is converted to neutral density when the image is converted from color to grayscale, in a similar way it is when printed on graded paper, but I'm not sure how much density. In other words, I think the effect is similar, but I'm not sure it's equal in degree.

If we agree that scanners respond to stained negatives in essentially the same way graded papers do, staining and compensation are interfering effects and cancel out; stain increases density and contrast exactly inversely to the way compensating development decreases density and contrast. This interference produces very straight characteristic curves, with full film speed and luminous shadows, glowing mid tones and delicate highlights; in short, all the good things we want in our negatives. If one wants maximum compensation, a non-staining developer should be used. Hypercat with a little sulfite does the trick nicely, which should come as no surprise since the Windisch, and Maxim Muir compensating formulas are so similar to Hypercat, except that they use enough sulfite to kill the stain, and sodium hydroxide as their alkali.

I'm looking at Julia's scans of her 3-bath development test, and I have to say, they look very good. Unfortunately, she mistakenly used a very old roll of Acros that has been exposed to heat and humidity for several years, and so there are some interesting effects unrelated to development. Still, looking past these defects, the images look very good. Sharpness is good, gradation is very nice, grain is nil, and there is quite a lot of compensation, it seems. The image I'm looking at now shows the interior of my kitchen, and the through the window I can see the tool shed in perfect detail, even though the difference in illuminance must be many stops. All in all, a very satisfying result.

This developer, as used, is very simple in formulation, it contains: metol, sulfite, pyrogallol, and sodium carbonate. If I were to optimize this formula, it might prove beneficial to add something more, at the very least a preservative for the pyro solution, which would likely be sodium bisulfite.

I'm a little surprised it works as good as it seems to, but much remains to be seen. How well will it work with other films? How will it deal with under/over exposure? How much non-image stain does it produce? How much film speed does it produce? There are a million questions like these to answer if this developer is to be taken seriously, and I'm not sure it should be.

[Jay DeFehr]

Ken,
the advantages of D-76 over D23 are: increased capacity (develop more film/liter), better film speed, and increased energy (shorter development times, better expansion development, push processing). D23 1+1 makes for long development times.

[Bob McCarthy]

Ken,
the advantages of D-76 over D23 are: increased capacity (develop more film/liter), better film speed, and increased energy (shorter development times, better expansion development, push processing). D23 1+1 makes for long development times.

Highlight development seems like a good answer. I agree regarding film speed, but that's of little import to me ( 1/2 stop, maybe a little less). Does D23 roll off the shoulder a bit compared to D76? Thats what it looks like to me. With TMax that may be an advantage.

I find capacity to be acceptable, 12 sheets of 4x5 or 4 sheets of 8x10 is my norm. (4x5 is by combi - 8x10 is by tray).

Some of the appeal, is I mix enough for a session, and dump. It never goes stale.

bob

[Jay DeFehr]

Bob,

I like soft working, single agent developers, and D23 is a very good one, but it can be a little too soft working for my taste. GSD-10 works better for me, and the kinds of photos I make, but if I had to use D23 instead, I wouldn't suffer much.

[sanking]

Let's remember that there are really only two major reasons to use a two-bath developer with large format sheet film.

#One, we expose many sheets of film in conditions that make record keeping for each of them difficult or impossible.

#Two, there is a scene of very high contrast and we need to compress highlight detail.

There is really no substitute for #One. Only a truly divided developer achieves the objective.

For #Two we can use a two bath developer, or standard N- development, or we can use stand or semi-stand development. Each method can be equally effective in compressing highlight detail.

BTW, if we have to keep returning the film to Bath A to bump up the shadow detail, the film would have been underexposed. To avoid that, simply expose the film for the deepest shadows and the film should develop fine without passing it again through Bath A.

Sandy

[patrickjames]

Divided developers can be useful for roll film, but for large format I think staining developers have advantages over them. I occasionally use a two part of my own formulation for 35mm just to switch it up a little. The biggest problem I think two part developers have is inconsistent development, which is a big deal and a good reason not to use them.

[Jay DeFehr]

Hi Patrick,

There is quite a bit of discussion above about the different types of divided developers, and the advantages/disadvantages of each. I think the differences between type A (development begins in first bath) and type B (development begins in second bath) divided developers are more significant than the differences between staining versions of each.Your post seems to ignore the fact that there are divided staining developers, and now there are divided staining developers of both types, though the type A staining developer requires a third bath to work. Which type is your developer, and what do you mean by, "inconsistent development"?

[sanking]

I have personally not seen any inconsistencies with two bath development. Indeed, for the article that I published in View Camera a couple of years ago I ran a number of sensitometry tests and found that the results in terms of density and tonal scale were consistent and predictable. In that article I tested both two bath D23 and Diafine. In contrast to divided D23 Diafine is a full bown two bath type that does not develop at all in Solution A.

One problem to look for with true two bath developers like Diafine and divided Pyrocat-HD (and other similar developers) is uneven development. This occurs because when the film is transferred from Bath A to Bath B development is virtually instantaneous and any unevenly wetted surfaces of the film are subject to uneven development. The way to avoid this is, 1) place a small amount of a wetting agent in Solution A, and 2) work with individual sheets of film rather than several sheets at a time.

BTW, two bath development works as well with rotary development as it does with tray. However, what I found was that in order to get approximately the same contrast with the two methods of development it was necessary to dilute the regular strength of both solutions 1 + 1 with water.

Sandy

[Ken Lee]

"This occurs because when the film is transferred from Bath A to Bath B development is virtually instantaneous and any unevenly wetted surfaces of the film are subject to uneven development. The way to avoid this is, 1) place a small amount of a wetting agent in Solution A, and 2) work with individual sheets of film rather than several sheets at a time."

Aha ! That explains it.

(Uneven development was one of the reasons I've been looking for something else).

Thanks !

[Ken Lee]

"the advantages of D-76 over D23 are: increased capacity (develop more film/liter), better film speed, and increased energy (shorter development times, better expansion development, push processing). D23 1+1 makes for long development times."

Thank you for the explanation. This forum (the people of course) is a marvel.

Diluted DD-23 - Aha ! I used it straight.

I found it necessary to cut down development time to 3.5 minutes. (When developing 15 sheets at a time, that's not enough for development by inspection: it's more like development by cardiac arrest) :rolleyes:

[Jay DeFehr]

Testing divided developers is complicated, and differs with each type of divided developer. Tests should be run for time in each bath, concentration of the first bath (type B developers), and pH of the second bath. There's more than just pH to the second bath, too. It's possible to make solutions of sodium hydroxide or sodium carbonate at the same pH, but the concentrations of the solutions will be very different. A dilute solution will neutralize locally, producing increased edge effects and compensation. That's why so many compensating formulas use hydroxide with catechol; catechol won't fog, and the pair makes for an excellent compensating developer. I digress, please excuse me. I guess my point is that there's more than meets the eye to a 2-bath developer, and while it's easy to make one work, they're devilishly complicated to test. That's part of the fun, I suppose!

[Bob McCarthy]

" D23 1+1 makes for long development times."

Thank you for the explanation. This forum (the people of course) is a marvel.

Diluted DD-23 - Aha ! I used it straight.

I found it necessary to cut down development time to 3.5 minutes. (When developing 15 sheets at a time, that's not enough for development by inspection: it's more like development by cardiac arrest) :rolleyes:

Ken, I jumped on 1:1 because of the shorter development times compared with straight D23. With development by inspection I would think that would also apply, perhaps even more so.

Texas summers don't help either


Bob

[patrickjames]

Hi Patrick,

There is quite a bit of discussion above about the different types of divided developers, and the advantages/disadvantages of each. I think the differences between type A (development begins in first bath) and type B (development begins in second bath) divided developers are more significant than the differences between staining versions of each.Your post seems to ignore the fact that there are divided staining developers, and now there are divided staining developers of both types, though the type A staining developer requires a third bath to work. Which type is your developer, and what do you mean by, "inconsistent development"?

I was not ignoring what is stated in the above posts Jay, I was simply giving my experience. It was very late when I posted so forgive the lack of clarity, but by "inconsistent development" I meant inconsistent across the negative, as in streaks and/or blotchiness, and not inconsistent from one batch to the next. With a two bath there will always be developer resting on the emulsion instead of being absorbed in it, and I tend to believe that is what causes the inconsistencies. The two part that I have been messing with is a metol/glycin formulation for A. From what I have seen so far, it isn't as prone to problems probably due to the glycin, but I am not too interested in two part developers or reinventing the wheel so to speak. I normally use Rodinal for roll film and Pyrocat-P for sheet film. They are of course both superb developers and I think it is hard to beat them, at least for the kind of negatives I want, which is of course a personal choice. In the last few years I have tried too many developers to mention and I stick these days to ones I know to work consistently for me.

[Jay DeFehr]

Patrick,

I hope I didn't seem harsh in my post; I didn't mean to. "Ignore" was probably a poor choice of words on my part that implied willfulness. I should have said "overlook", or something similar. Thank you for clarifying what you meant by inconsistencies; the two alternatives you describe are exactly what I wondered about, and I agree with your assessment of the problem. I think Sandy's recommendation of a wetting agent is a good one, and I intend to incorporate it into my divided development. I too have a few developers that I know very well and use for anything I think is important, but I also enjoy tinkering with developers and development methods. It's a lot of fun for me, and the developers I do rely on all resulted from my tinkering. Glycin is a very good choice for a 2-bath developer, by the way, and it doesn't need any help from metol, either. Thanks again for your post, and I hope I didn't seem rude.

[patrickjames]

No problem Jay. I was overly simplistic, which is a problem I have sometimes. I like messing with different developers too, but these days I don't necessarily have the luxury. The metol that I add is secondary to the glycin in the two bath that I use. I will eventually do more with it, especially messing with the pH of the B bath. I also have some other ideas for the A bath, but I am not too interested in redoing everything so it will take quite a bit of time to mess with it. I tend to tackle problems when I need to instead of just for the fun of it.

[Jay DeFehr]

Good luck with your developer, Patrick, and have fun!