New 2-bath developer part B discovery.

Alan Johnson · 2025-05-05T08:41:45-0400

I received an interesting message from Raghu K in India to the effect that he had found sodium ascorbate (not ascorbic acid) could be used as a B bath in Barry Thornton's 2 bath developer (BTTB) and I tried it out to see if the idea could be verified.
Some initial results suggested that 5g/L sodium ascorbate (about 2 level teaspoons) would be a good level for a test.

My test with Barry Thornton's two bath part A + various part B using FP4 type 517 cine film* EI=100, 22+/- 0.5C:
BTTB part A only ,5 inversions at start then 2 inversions on each minute, time 4m.
BTTB part A 4m, part B 5g/L sodium ascorbate 6m agitate 2 inversions at start and 2 inversions at 3 m, total time 6m.
BTTB part A 4m, part B 12g/L metaborate 4m agitate 2 inversions at start and 2 inversions at 2 m., total time 4m.
Sodium ascorbate 5g is closely equal to 2 level teaspoons.
*FP4 type 517 cine film is from a large quantity of outdated film sold cheaply by Analogue Cameras in UK, its EI is given as 100.

It can be seen from the attachment that there is significant development in part A alone, the density increases in both part B sodium metaborate and in part B sodium ascorbate.
Raghu's report appears to be true.
From the scans, it does appear that sodium ascorbate gives slightly higher contrast and possibly slightly finer grain. It's hard to see this on the <2mb scans attached here. Possibly most of the grain comes from the part A development.

What use might this be? One application seems to be that it is easier to buy sodium ascorbate than it is to buy the metaborate usually specified for BTTB as sodium ascorbate is a food supplement and 2 level teaspoons of it per liter should do a similar job to 12g/L metaborate. That is a preliminary result and comments welcome.

retina_restoration · 2025-05-05T08:55:49-0400

It might be that the slight increase in contrast in the ascorbate version could be solely due to the 6 minute time, versus 4 minutes in metaborate.
But that’s an interesting result. There doesn’t appear to be any significant advantage in terms of results, but if metaborate isn’t available, it’s great to have an alternative.

I’ve always found it puzzling (misleading) when people state “no development occurs in Part A” when clearly it does. Your experiment demonstrates this fact.

Milpool · 2025-05-05T09:35:32-0400

Barry Thornton bath A is only trivially different than D-23, so bath A is a normal developer, which is why the Thornton two-solution is basically the same as the old Adams two-solution, Stoeckler fine grain etc.

Ascorbate is superadditive with metol, so it makes sense that this would work given the carryover of metol and weakly alkaline pH of bath A (pH ~8).

The question is - is this giving different sensitometry / tone reproduction and/or image structure than adding the sodium ascorbate to bath A and simply making that a regular one-bath developer. This can be tested. If the sensitometry is the same a one-bath process is preferable (all other things being equal).

retina_restoration · 2025-05-05T09:44:47-0400

Milpool said:
Barry Thornton bath A is only trivially different than D-23, so bath A is a normal developer, which is why the Thornton two-solution is basically the same as the old Adams two-solution, Stoeckler fine grain etc.

Ascorbate is superadditive with metol, so it makes sense that this would work given the carryover of metol and weakly alkaline pH of bath A (pH ~8).

The question is - is this giving different sensitometry / tone reproduction and/or image structure than adding the sodium ascorbate to bath A and simply making that a regular one-bath developer. This can be tested. If the sensitometry is the same a one-bath process is preferable (all other things being equal).

I wonder if there's a developer that uses Metol and Ascorbate in a single solution?

Alan Johnson · 2025-05-05T09:50:01-0400

Milpool said:
Barry Thornton bath A is only trivially different than D-23, so bath A is a normal developer, which is why the Thornton two-solution is basically the same as the old Adams two-solution, Stoeckler fine grain etc.

Ascorbate is superadditive with metol, so it makes sense that this would work given the carryover of metol and weakly alkaline pH of bath A (pH ~8).

The question is - is this giving different sensitometry / tone reproduction and/or image structure than adding the sodium ascorbate to bath A and simply making that a regular one-bath developer. This can be tested. If the sensitometry is the same a one-bath process is preferable (all other things being equal).

From an article by Barry Thornton the two bath development cannot be replicated by one bath development:

In the second bath the developer soaked into the film emulsion is activated by the accelerator. In the highlight regions where the developed silver will be densest, the developer available in the emulsion is soon exhausted and development halts, thus automatically limiting the density of the negative at that point. The more the exposure, and the denser the highlight, the faster development ceases. In the shadows, though, there is little silver to reduce and there is enough developer to keep working there to push up the shadow detail density. The less light the negative received at this point the longer the development proceeds. Indeed there is a minor hump put into the characteristic curve of many films between the shadow and mid tones to give heightened shadow contrast. The effect is not the same as the well known technique of compensating development by diluting developers, which does work in holding back dense highlights, but can give muddy mid times and does not have the same automatic contrast equalisation as the two bath. Of course there is a limit to the contrast that can be equalised, but most negatives will print to good quality on 2 or 3 grades of paper with only the most extreme contrast range subjects requiring other contrast control methods for printing.

Milpool · 2025-05-05T10:11:28-0400

retina_restoration said:
I wonder if there's a developer that uses Metol and Ascorbate in a single solution?

There are some. In fact Ron Mowrey disclosed Kodak sometimes used this Elon-AA internally for test purposes:

Metol: 2.5g/l
Ascorbic acid: 10.0g/l
Sodium metaborate: 35.0g/l
KBr: 0.4g/l

Obviously the above is not intended to be a fine grain / solvent developer as it lacks sulfite. I'm just noting it for historical interest. The point is, you can use ascorbate as a superadditive agent with metol (or Phenidones). It's not quite the same reaction as with HQ but anyway.

Note using sodium ascorbate will be different than ascorbic acid from a pH perspective.

The issue with ascorbate in DIY developers as we know, is the potential Fenton problem in the absence of a strong iron/copper chelating agent such as DTPA.

Milpool · 2025-05-05T11:17:43-0400

Alan Johnson said:
From an article by Barry Thornton the two bath development cannot be replicated by one bath development:

In the second bath the developer soaked into the film emulsion is activated by the accelerator. In the highlight regions where the developed silver will be densest, the developer available in the emulsion is soon exhausted and development halts, thus automatically limiting the density of the negative at that point. The more the exposure, and the denser the highlight, the faster development ceases. In the shadows, though, there is little silver to reduce and there is enough developer to keep working there to push up the shadow detail density. The less light the negative received at this point the longer the development proceeds. Indeed there is a minor hump put into the characteristic curve of many films between the shadow and mid tones to give heightened shadow contrast. The effect is not the same as the well known technique of compensating development by diluting developers, which does work in holding back dense highlights, but can give muddy mid times and does not have the same automatic contrast equalisation as the two bath. Of course there is a limit to the contrast that can be equalised, but most negatives will print to good quality on 2 or 3 grades of paper with only the most extreme contrast range subjects requiring other contrast control methods for printing.

The difference is that two-bath / two-solution development generally tends to have a straightening effect on the characteristic curve, shortening the toe and/or shoulder. Some results from my own experiments below (may or may not be of value).

However the sensitometry and image structure typical of the normal two-bath process in which the second bath is an alkali, may or may not be the same as that produced if the second bath is instead composed of a (superadditive) developing agent (at roughly neutral pH in this case). This would need to be evaluated.

koraks · 2025-05-05T11:22:39-0400

Milpool said:
Some results from my own experiments below

Is it just me, or does your D23 curve show nothing that resembles the supposed compensating effect of a two-bath approach?

Milpool · 2025-05-05T12:34:37-0400

koraks said:
Is it just me, or does your D23 curve show nothing that resembles the supposed compensating effect of a two-bath approach?

The charts are somewhat out of context so perhaps not the best examples but basically the idea is you can retain a little more effective “speed” for lower overall gradients. The lower the aim gradient the more pronounced the differences are.

The red and blue curves in the first chart are representative of what I found to generally happen - higher emulsion speed and a little straighter overall. The most “extreme” results come from repeating the process (short times in bath A).

However there is nothing spectacular going on here. I also found uniformity to be unfavourable relative to normal processing, and the negatives are grainier as expected.

It’s possible you get some enhanced edge effects if that’s important to anyone but I made no attempt to evaluate that so I really don’t know.

Raghu Kuvempunagar · 2025-05-05T13:47:11-0400

Alan Johnson said:
That is a preliminary result and comments welcome.

Hi Alan,

First of all, a big thanks to you for this.

In my tests I used Ascorbate as a single use Part B. As this point might not be obvious from a reading of OP, I thought of mentioning it here.

-- Raghu

snusmumriken · 2025-05-05T15:55:49-0400

koraks said:
Is it just me, or does your D23 curve show nothing that resembles the supposed compensating effect of a two-bath approach?

I have wondered the same. Puzzles me, even though I use BT2B and am happy with the results in a wide variety of situations.

Milpool · 2025-05-05T16:50:21-0400

snusmumriken said:
I have wondered the same. Puzzles me, even though I use BT2B and am happy with the results in a wide variety of situations.

“Compensation” is usually intended to mean a disproportionate reduction of highlight contrast / density in relation to shadows and midtones. That is typically what people assume is happening when they use dilute developers with reduced agitation. Two solution development appears to give a different result. I don’t think my trial results are unique / anomalous in that respect - I remember an article by Sandy King years ago in View Camera magazine in which he came to the same conclusions.

Photographic processing is often like that - we assume some particular thing is happening (for a variety of reasons) and even if it isn’t, we remain none the wiser because of the latitude inherent in end-to-end process.

lamerko · 2025-05-05T17:20:16-0400

I think the ascorbate in the second bath will not work as an accelerator like the metaborate, but more like a second developer. Is it possible to partially "reverse" the oxidation of the metol and create a super superadditive reaction with it?

Alan Johnson · 2025-05-05T17:41:45-0400

lamerko said:
I think the ascorbate in the second bath will not work as an accelerator like the metaborate, but more like a second developer. Is it possible to partially "reverse" the oxidation of the metol and create a super superadditive reaction with it?

It is possible, as in Ryuji Suzuki's DS-12 and DS-2:

Film Developer Recommendations - Silvergrain Labs

web.archive.org

However, these take a few minutes to work at pH 9.75- 9.8.

I don't know the pH of Thornton's part A . Sodium ascorbate is about pH 7.5 according to Gainer. Not clear to me if the reaction may be just a continuation of the sulfonation of metol oxide that occurs in part A but at the lower pH of sodium ascorbate.

Raghu Kuvempunagar · 2025-05-05T23:30:40-0400

koraks said:
Is it just me, or does your D23 curve show nothing that resembles the supposed compensating effect of a two-bath approach?

What would you say about the divided Pyrocat curve which can be found here? I guess compensating effect in this context is intended to mean lower contrast development but not necessarily with a pronounced shouldering effect.

koraks · 2025-05-06T01:44:33-0400

Raghu Kuvempunagar said:
I guess compensating effect in this context is intended to mean lower contrast development but not necessarily with a pronounced shouldering effect.

That's probably where my misunderstanding stems from. Although I'd wager to say that I find 'compensation' a poor choice of words if what's intended is 'linearity with low gamma'.
The chart you linked to is nice, but it uses four reasonably to very linear films to begin with, to illustrate that a linear result is obtained. A more illustrative chart would be one that plots the HD curve for divided and normal development for the same film. It's also somewhat unfortunate that the chart peaks out at fairly low absolute densities.

relistan · 2025-05-06T05:57:17-0400

Very interesting, thanks for posting @Alan Johnson and @Raghu Kuvempunagar . Effective two bath developers all seem to do significant development in bath A and that's reinforced here.

Given the effect demonstrated, since the pH is much lower here than for a metaborate bath B, the effect must be assumed to be either:

Because the dev time was extended
Because of superadditive effects of ascorbate and metol

It would be quite interesting to see what it looks like with 6-8 mins in part A only. That would help to eliminate how much is superadditive vs extended time. My _guess_ is that the superadditivity is about half of it.

It's good that Raghu points out that the bath B has to be one-shot when using ascorbate. Otherwise, the preservative and superadditive nature of ascorbate would make bath B quite a good developer after only 1-2 rolls. I found this to be the case in all the ascorbic-acid-based two baths I worked on where even the carryover from A rapdily spoiled B because the preservative nature of ascorbic acid prevents the normal exhaustion/breakdown of the developing agents in B.

snusmumriken · 2025-05-06T06:02:20-0400

koraks said:
That's probably where my misunderstanding stems from. Although I'd wager to say that I find 'compensation' a poor choice of words if what's intended is 'linearity with low gamma'.
The chart you linked to is nice, but it uses four reasonably to very linear films to begin with, to illustrate that a linear result is obtained. A more illustrative chart would be one that plots the HD curve for divided and normal development for the same film. It's also somewhat unfortunate that the chart peaks out at fairly low absolute densities.

You can see such a chart here. If I understand the implications correctly, the metol produces a low contrast image with good shadow development in Bath A. In Bath B, development is completed until locally exhausted, which will increase contrast overall but tend to favour development of the shadows again. So the bulk of the effect is in straightening out the toe without rendering the highlights unprintable, rather than in exaggerating the shoulder.

If that is indeed correct, I feel vindicated in my practice of giving continuous agitation in Bath A, limited agitation in Bath B.

relistan · 2025-05-06T06:09:28-0400

snusmumriken said:
You can see such a chart here. If I understand the implications correctly, the metol produces a low contrast image with good shadow development in Bath A. In Bath B, development is completed until locally exhausted, which will increase contrast overall but tend to favour development of the shadows again. So the bulk of the effect is in straightening out the toe without rendering the highlights unprintable, rather than in exaggerating the shoulder.

If that is indeed correct, I feel vindicated in my practice of giving continuous agitation in Bath A, limited agitation in Bath B.

That is how it's supposed to work. As you re-use bath B, however, it becomes a stronger and stronger developer. Re-use more than about 3 times will start to have a noticeable effect on the development because B starts to become a decent developer. @gorbas tested this out in another thread a few years ago.

Limited agitation in B makes sense (this is what I do), but for consistent results you need to have agitation when you first pour it in, or the surface carryover can given uneven development.

koraks · 2025-05-06T06:14:31-0400

snusmumriken said:
You can see such a chart here.

I don't see it; all I see is 2 charts that only show divided development. Am I misinterpreting the labels? I'd also wager to see that the plots are variable and support either the compensation or the linearity hypothesis, depending on what you prefer to look at.

Anyway, I don't want to take this into the direction of the purported use of 2-bath development. The thread is about the use of ascorbate for bath B, let's get back to that.

snusmumriken · 2025-05-06T06:24:00-0400

relistan said:
That is how it's supposed to work. As you re-use bath B, however, it becomes a stronger and stronger developer. Re-use more than about 3 times will start to have a noticeable effect on the development because B starts to become a decent developer. @gorbas tested this out in another thread a few years ago.

Limited agitation in B makes sense (this is what I do), but for consistent results you need to have agitation when you first pour it in, or the surface carryover can given uneven development.

I fully accept both points. It’s just that many of us would have expected starvation of developer in the highlights to create a pronounced shoulder. Bath B is cheap enough and easy enough to re-make frequently.

relistan · 2025-05-06T06:51:36-0400

snusmumriken said:
I fully accept both points. It’s just that many of us would have expected starvation of developer in the highlights to create a pronounced shoulder. Bath B is cheap enough and easy enough to re-make frequently.

Yep! Was not arguing with you.

New 2-bath developer part B discovery.

Alan Johnson

Attachments

retina_restoration

Milpool

retina_restoration

Alan Johnson

Milpool

Milpool

koraks

Moderator

Milpool

Raghu Kuvempunagar

snusmumriken

Milpool

lamerko

Alan Johnson

Film Developer Recommendations - Silvergrain Labs

Raghu Kuvempunagar

koraks

Moderator

relistan

snusmumriken

relistan

koraks

Moderator

snusmumriken

relistan

New 2-bath developer part B discovery.

Attachments

Moderator

Moderator

Moderator

Privacy & Transparency

Privacy & Transparency