Windisch Surface Developer _ what's correct

[Ian Grant]

Ron how do you explain "I found out that this was because Crawley thought the avoirdupois versions of the formulas were more accurate than the metric ones" which refers to Bill's previous line "Here's an interesting footnote on BJP formulas for Ian. There are some apparently super-precise renditions of some Kodak formulas such as D-61a that have amounts like 3.1 grams instead of the expected 3.0 grams."

In truth I've been aware of these differences in some Kodak formulae since the early 70's and knew that some of the formulae were later rounded to the nearest half gram for amateur use, where volumes being made up were smaller & scales far less accurate. It maybe the rounding down was done earlier in the US than the UK They wouldn't have bothered with roundin up or down the Avoirdupois weights as it wasn't relevant.

Ian

 

[Photo Engineer]

Ian;

It says nowhere that Crawley did the conversion, merely that he thought Kodak had done the conversion and rounding the other way (Avoirdupois -> Metric). You seem to have stated that Bill actually said Crawley had done it....

"Oh and Bill the primary weights given by Kodak are ALL metric, so Crawley didn't convert them from the Avoirdupois."

And, I did not read Bill's statement that way. Sorry.

As for rounding, the conversion back and forth from A->M and M->A is such that rounding should not yield the values we see. Rounding would change either the A or the M value but would not leave one of them (A) unchaged in all cases. In fact, using 5.0 or 5.1 and converting would cause a rounding of ~1.0 grains when in fact, no change was made. And, this value of 1.0 grains is well within the range of most Avoirdupois scales.

PE

 

[billtroop]

>Oh and Bill the primary weights given by Kodak are ALL metric, so Crawley didn't convert them from the Avoirdupois.

Did you ask him, Ian? I did, and what he told me I stated here. What you will find is that Kodak has, most confusingly, published this formula in a number of ways, which accounted for Crawley's confusion. It had been an issue that had interested him, and he was quite clear about it. H.D. Russell told me that the unrounded, metric version is the definitive, original one -- that is to say, the one printed in FDC. There was no way, HDR explained to me, that anyone at Kodak was going to formulate a developer for common use which specified 3.1g instead of 3.0 for the metol, or 2.1g of bisulfite instead of 2.0. It didn't make any sense, and he frankly expected me, he said, to have more sense than to raise the question when I should have known the answer.

 

[wogster]

>Oh and Bill the primary weights given by Kodak are ALL metric, so Crawley didn't convert them from the Avoirdupois.

Did you ask him, Ian? I did, and what he told me I stated here. What you will find is that Kodak has, most confusingly, published this formula in a number of ways, which accounted for Crawley's confusion. It had been an issue that had interested him, and he was quite clear about it. H.D. Russell told me that the unrounded, metric version is the definitive, original one -- that is to say, the one printed in FDC. There was no way, HDR explained to me, that anyone at Kodak was going to formulate a developer for common use which specified 3.1g instead of 3.0 for the metol, or 2.1g of bisulfite instead of 2.0. It didn't make any sense, and he frankly expected me, he said, to have more sense than to raise the question when I should have known the answer.

I have been following this, and wonder, how much of a difference would there be, in activity between 3.0 and 3.1g? There also raises the question or what was the original formula for, if someone at Kodak had originally developed a formula for 10L of solution, which used 31g, then converted that tor 1L, it would be 3.1g. If there isn't a huge difference, then it could end up later to have been changed to 3g

 

[Photo Engineer]

Paul;

OTOH, I have seen a big difference caused by 0.1 grams / liter of a developing agent. It depends. This is why I am very interested in this thread.

OTOH again, errors happen. Many formulas I know used milligram quantities of ingredients such as 0.05 g of KI or 0.235 g of NaBr in one liter. Sometimes this is critical. I cannot agree, apriori with either position so far as I have seen formulas both ways, rounded and exact for specific purposes.

In actuality, B&W formulas most often can be rounded, but not always. The 0.1 g above was from experience, but color must be more exact. Some formulas go to 3 decimals!

PE

 

[wogster]

Paul;

OTOH, I have seen a big difference caused by 0.1 grams / liter of a developing agent. It depends. This is why I am very interested in this thread.

OTOH again, errors happen. Many formulas I know used milligram quantities of ingredients such as 0.05 g of KI or 0.235 g of NaBr in one liter. Sometimes this is critical. I cannot agree, apriori with either position so far as I have seen formulas both ways, rounded and exact for specific purposes.

In actuality, B&W formulas most often can be rounded, but not always. The 0.1 g above was from experience, but color must be more exact. Some formulas go to 3 decimals!

PE

I guess it would also depend on the percentage represented, .1g is a much larger part of 3g then it would be of 30g. I would assume that if Kodak was publishing a formula as 3.1g and later 3g that they had done sufficient testing to determine that, in that particular case the .1g was not a big enough difference to warrant the exact 3.1g.

 

[Ian Grant]

Data sources

Just to clarify: The Kodak formulae that Bill Troop mentions "which specified 3.1g instead of 3.0 for the metol, or 2.1g of bisulfite instead of 2.0" had been published in the BJP Almanac's - 1940's onwards, many years prior to Crawley becoming the editor in 1962.

They are also in the "Kodak Reference Handbook, 2nd edition" which I bought last year from a US APUG member, and I have back in the UK, so it's not a case of differences between Eastman Kodak (US) & Kodak Ltd (UK) publications. All the Kodak formulae I've seen have always been published in Metric form right back to the early 1900's

Bill makes a point of being accurate, all I can say is that the rounded off formulae aren't in any of the Kodak publications that I own, or have read, and I'm extremely surprised he's not seen this for himself.

Ron is right about the small differences between using 3.0 compared to say 3.1, may not matter. But when the Metol is reduced by 3.3% the Hydroquinone upped by 1.5% and the Bromide increased by 1.8%, as well as the Carbonate by 6.35% while the Bisulphite is dropped by 5% then there must be a far more significant difference. That's comparing the Official Kodak formula for D61a and the version Bill Troop published in the FDC

Ian

 

[billtroop]

HD Russell's greater point is this: look at Kodak formulas, and you will go very far to find one where these key amounts: (a) primary developing agent , (b) secondary developing agent, and (c) buffer, are all rounded off. You will go even further before you find a formula where they are all rounded off to just one-tenth either side -- i.e. the bare minimum.

Nothing can be more obvious than that the versions of D-61a with metol 3.1, HQ 5.9 and bisulfite 2.1 are misconversions.

Had the amount of metol been 3.5, HQ 5.6, bisulfite 2.4, it would not be so obvious.

Having straightened all of this out 20 years ago with HD Russell, it is surprising to have to be arguing it again and astonishing to find so small a point inspiring both passion and temper. Surely the calibre of what we talk about in photochem is a little higher than that?

Ian, what I have always found is that when a question arises about a probable error in a publication, you can seldom establish truth by reference to other publications, which all too easily either repeat the same error or introduce new ones. Instead, I always try to consult the relevant top expert, and commend this method to you.

The number of primary Kodak participants who were active in the 1920s to the 1940s or 1950s must now be very small. Personally, I don't know any. I talked to these people when they were still around.

So what would I be doing today? I would be looking at the work done from the 1950s to the 1990s and be going over it with people who were actually there and could speak with complete authority.

 

[Photo Engineer]

I would have to say that the differences in the 3 formulas that I have are rather disturbing. The ingredients change as Ian says, in the metric, but NOT ALL CHANGE in Avoirdupois. This is rather peculiar. Also, the change in carbonate is not clear as one is listed as monohydrate and the other as anhydrous and the values are not really an increase due to this little add in factoid.

As I stated earlier, I found a formula to vary mixed by two people, and I reformulated and found that the variations damped out by adding 0.1 grams / liter of a particular chemical to the mix. Therefore, either chemical batch variations or changes in mixing procedures caused a slight change in contrast with this developer in identical curcumstances when mixed in two different places and then tested by me.

At Kodak, formulas for all processes are initiated in KRL during normal R&D. The formula(s) are sent to manufacturing (Film or Paper depending on product) and to Photographic Technology. They are also sent to P&S (Powders and Solutions). Photo Tech works out any replenisher and verifies the formula responding to all other recipients with yes, no or changes. Everyone tests the changes and then P&S makes a batch for prototyping. Again it is yes, no or changes. Finally, everyone gets a final formula confirmation and it is cast in concrete. It takes a major act to change it, and if changed it moves from say Xa to Xb in naming so that there is a new name.

This rough description then ends the Park contribution. Formulas for publications are sent to technical writers who write those books. There is probably where some confusion takes place, as all formulas are only in metric! Transcription after transcription for the various volumes confuse the errors and we have what we do today.

These tiny changes are important. The three versions of D61A that I have would yield slightly different results. All would work reasonably well.

PE

 

[Ian Grant]

Nothing can be more obvious than that the versions of D-61a with metol 3.1, HQ 5.9 and bisulfite 2.1 are misconversions.

Had the amount of metol been 3.5, HQ 5.6, bisulfite 2.4, it would not be so obvious.

As Ron points out the changes in D61a could have a significant effect, the change in Carbonate up 6.35% and the Bisulphite down 5% would have an effect on the pH and buffering, then add in the change in ratio of the developing agents and the increased bromide, it would be more active and give a higher contrass and possible increased grain.

Kodak's own Formulary from the Research Facility at Harrow shows the Formulae correctly with no rounding off and they always worked with Metric weights it's not likely to be a conversion error, particularly as this is a compounded formulae, sold in powder form.

If you look more closely the number of formulae where rounding off has taken place is more than first meets the eye, D61a, D72, D158 "Velox", D163, etc these were the main Kodak packaged developers. Perhaps not all the chemistry packing plants were Metric, so maybe this explains the greater accuaracy.

But as Ron notes the Avourdupois weights don't usually change, and if you were rounding up/down it would be done differently in the two systems anyway.

Either way Kodak published the formulae widely and in many publications for both anmateurs & professionals without rounding them off to the nearest half gram, and there's no arguing that fact and so to try and find excuses 60+ years later is crazy, you just have to accept that you're wrong Bill.

Ian

 

[Photo Engineer]

I would have to say that Bill may not be wrong so much as wrongly informed by people such as Crawley, and perhaps even the Kodak people who try to avoid giving out that sort of information. He is certainly entitled to defending the information as given him. What he says makes perfect sense to me, given what I know of the situation at that time.

Grant could be perfectly candid with me, but when editing his books I had to red line some items (IIRC) due to management guidelines, as the products referred to were still in R&D. Therefore, by analogy, Grant (and Dick Henn et. al. ) could not be entirely candid with Bill. And to admit to an error? Wow, unheard of!

As for published formulas - all of them were targeted at professionals. Amateurs were expected to buy prepackaged materials as were most professionals. And, the formulas were, by definition, developed by professionals at EK.

PE

 

[Ian Grant]

Crawley doesn't deviate from the Kodak facts.

Ian

 

[Photo Engineer]

Crawley doesn't deviate from the Kodak facts.

Ian

Given what Crawley has published and said to Bill, there are "errors".

Given that he did not work for EK, then people could not be entirely candid with him either!

So the "facts" he published were only "facts" insofar as Kodak wanted them known or insofar as they had been allowed to publish them.

I myself would have referred to a D61a and D61b for clarity. Crawley knew of more than 1 formula it seems but just ignored the differences, when I can clearly see some problems as you can yourself. I have to ask why he seems to ignore them.

PE

 

[Ian Grant]

Have you actually read Crawley's extended 1960/1 article Ron, or seen the formulae listings in the BJP Almanac's and Annuals he edited.

Unlike Bill Troop, Crawley didn't purport to have inside knowledge from any company in his original article, he merely examined some of the then commercially available developers. Kodak developers have been published over many years in the BJP, in fact the BJP pre-dates Kodak by roughly 30 years, and began publishing Kodak formulae when they were first release, I'd guess from around 1900 onwards. Crawley merely continued to publish the Kodak formulae that were already listed (which happen to be the ones in question), and they don't differ from the Koadak Formulary.

Crawley wasn't interested in what variations Kodak had in research files back then (1960/1) but rather in devising what he thought then were state of the art formulae. He had started the publication of formulae in the BJP all companies released formulae for publication. In another thread someone mentioned that Kodak didn't start selling chemistry until 1938, I've no idea how true that is but in their UK adverts in the BJP everything for Photo-finishers is listed except chemistry, and Australian/New Zealand branch has none either (separate adverts), in fact no pre-packed chemical at all.

In many ways Crawley is mainly useful for his writings in a more general sense and observing how he tackled developer design. His name was only introduced into this thread because Bill seems to attribute the 3.1/5.9 weights etc to him, while they were in fact Kodak's own weights, and very widely published.

Lets leave Crawley aside as he's totally irrelevant in the differences with the rounded off formulae and played no part in them, if or when Kodak published formulae they have to be accepted as being correct, after all they are primary sources of data.

The fact that a great many Kodak formulae have been published by others with the Metric weights rounded off means that they aren't strictly accurate and we have to accept that the original Kodak weights are in fact historically correct for the formulae.

Ian

 

[Photo Engineer]

Ian;

My resources were not second or third hand from people like Crawley however respected he may be. My resources were the actual people who did the work, and so even though I read the BJP from time to time, it was not a primary resource to me. My resource was the black book and the Technical Reports that I read every month as my "homework". Unfortunately, I cannot reproduce formulas in detail. I can also say that when I got data, it was not equivocal, it was exact.

What I do have now are 4 textbooks with the D61a and D61r formulas. Interestingly, there are at least 2 variants on D61a and 2 variants of D61r. Interesting also is the fact that the Avoirdupois changes do not match the magnitude nor direction of the metric changes in all cases.

So:

D61a D61r D61a D61a D61r
Elon 3.1 5.9 3.1 3.0 6.0
Sulfite 90 180 90 90 180
Bisulfite 2.1 3.5 2.1 2.0 4.0
HQ 5.9 11.9 5.9 6.0 12
Carbonate 11.5A 240A 11.5A 14.0M 280M
Br 1.6 3.1 1.7 2.0 3

A = anhydrous, M = Monohydrate
R = replenisher to make 6 L part a (no alkali) and 2 L part b (alkali) mixed 3:1 for use


Now, in spite of all of these changes the Avoirdupois values of Metol and HQ remain the same while the others change up or down accordingly. This is what I refer to above and what can make a big difference in the formula made up to any of the formulas using any measureing system.

And, the neat tables above will probably be trashed by APUG as usual as it already has upon import from Word. This is a big complaint I have about posting data like this!

PE

 

[billtroop]

Ian, nothing you say takes away from the simple fact that Crawley published his own conversions from avoirdupois. He did the maths himself. You jump to the conclusion that he was simply copying Kodak's data. But he wasn't. He was confused by all the data, and wanted to check it himself. You can verify this by giving him a call.

Also, nothing you say can take away that Russell informed me that the unrounded versions were correct. He was there. You weren't. Who is more likely to be right?

It is obvious to me, further, that when Kodak, in published formulas for general use gave amounts to the tenth of a gram, they generally assumed, as in the case of KBr, that 10% solutions would be used, since it was inconceivable that most of the people making up the formulas would be capable of weighing to within 10th of a gram accuracy.

You might just as well ask why D-76 has 2 grams of metol, rather than 1.9, or 100 grams of sulfite rather than 99.5.

By the way, haven't you seen the Kodak publications where D-61a actually is given with integers for the principal chemicals?

Ron, do you happen to have a copy of Haist handy? I wonder what he has?

 

[sanking]

All I can say is that it is a pretty sorry state of affairs that two centuries after France introduced a vastly simpler system of measurements, the metric system, people from the US and the UK are arguing about how conversions were made from one system to the other.

I would also add that if someone, anywhere, plans to publish a book of old formulas it would be nice to establish in advance some kind of standard protocol to show actual results in terms of curves, films speed, etc when these formulas are used with contemporary films. If this is not done, any comments on the potential usefulness of the formulas will be basically useless.

My personal opinion is that there is entirely too much intellectual masturbation about this subject and a lack of willingness to do good testing to provide empirical data.

Sandy King

 

[Ian Grant]

So lets agree that Troop is disagreeing with Kodak's own published figures

Ron, you and I don't dispute those figures, Crawley & the BJP are TOTALLY irrelevant because the Kodak original column figures are repeated correctly without mistakes (in the BJP Almanacs 1947, 50, 51, 53, 54, 57, 58, 59, 60, 61, 62 72 76 etc copies I have), as one would hope they would be..

I really CAN'T understand the problem you seem to have with that, and what on earth you're arguing about.

What Bill Troop is trying to say is that figures that both you and I agree on are wrong. Formulae do get changed slightly sometimes, the table's lost but the replenisher seems to differ slightly.

So lets agree that Troop is disagreeing with Kodak's own published figures regardless of where he saw them. You have them, I have them both in Kodak publications but Bill obviously doesn't !

Ian

 

[billtroop]

>Crawley doesn't deviate from the Kodak facts.

Oh Ian, that's just so silly. For heaven's sake, Crawley deviates from his own facts and has never been reluctant to admit it. On page 45 of FDC, you will find a note which reads: "A problem with the FX formulas is that they have often been printed with errors, even in the BJ Annuals, throughout the 1970s and 1980s. Photographers who have used some of these developers with poor results probably were using incorrectly published formulas. Effort has been made to ensure that the formulas published here are correct. Each one has been checked by Mr. Crawley."

That statement was read by Mr. Crawley, who was responsible for the errors. Even with all that, we didn't get FX-37 right in the first printing!

So please, Ian, don't try to tell us what facts GC deviates from in print. Of course he knows what the score is, but he has never had particularly good luck with printers, and that issue continues today with his wonderful contributions to Amateur Photography. See, for example,

http://www.amateurphotographer.co.uk/forums/showflat.php/Cat/0/Number/687729/an/0/page/0

for his gracious response to his latest spell of bad luck with the printers.

 

[Ian Grant]

Just accept you're wrong and stop flannelling.

We have at NO TIME been discussing Crawley's formulae or his own errors so stop the Bull shit.

Either Kodak are right or you are and the world will accept Kodak's published formulae, if you want to change them or publish them inaccurately that's your business.

Ian

Ian, nothing you say takes away from the simple fact that Crawley published his own conversions from avoirdupois. He did the maths himself. You jump to the conclusion that he was simply copying Kodak's data. But he wasn't. He was confused by all the data, and wanted to check it himself. You can verify this by giving him a call.

Also, nothing you say can take away that Russell informed me that the unrounded versions were correct. He was there. You weren't. Who is more likely to be right?

It is obvious to me, further, that when Kodak, in published formulas for general use gave amounts to the tenth of a gram, they generally assumed, as in the case of KBr, that 10% solutions would be used, since it was inconceivable that most of the people making up the formulas would be capable of weighing to within 10th of a gram accuracy.

You might just as well ask why D-76 has 2 grams of metol, rather than 1.9, or 100 grams of sulfite rather than 99.5.

By the way, haven't you seen the Kodak publications where D-61a actually is given with integers for the principal chemicals?

Ron, do you happen to have a copy of Haist handy? I wonder what he has?

 

[billtroop]

Ian, you have yet to explain why Kodak also publishes the correct formula for D-61a - - as I do.

All this effort spent to defend a misprint in a formula of academic interest that will probably never be used again by any sane person! Can't you find a better use of your time? And for heaven's sake please learn to make statements without resorting to CAPS and bold, which make even the coolest person look cranky. I just cannot stand seeing bold in blogs!

 

[billtroop]

P.S. Very generally on the issue of errata in 'definitive' sources. When I told Dick Henn I was using Crabtree and Matthews's Photographic Chemicals and Solutions as a primary source for some of the older Kodak formulas, he said, 'I wouldn't do that. Internally, we had fourteen pages of errata for that book. It is not a reliable source for formulas. Unfortunately, I just wouldn't know how to get a copy of the errata for you. I don't have one myself.'

Other general errata: Haist told me that they had discovered in the 1950s that most pH measurements made in Kodak research prior to that time were erroneous.

Nelson told me that the densitometer used for Loyd Jones's work was later discovered to be unreliable. That did not affect the general truth of Jones's colossal work, but it did lead to some baffling if minor erroneous observations, such as the famous 'hump' in paper curves. Needless to say, there was never any public acknowledgment of any of these errors.

Work through some of the equations in TH James and you will find that some of them just don't add up.

Errors are an inseparable component of the photographic literature. It's left to people in the field to get things right! As Sandy King would probably say!

 

[Photo Engineer]

Ian;

I am pointing out that the different formulas do not appear to match the changes in the Avoirdupois measurements, and I wish to point out that this difference is either larger or smaller but not definitively due to rounding. This points to other errors creeping in besides rounding. I don't disagree with you or Bill, but I do interpret Bill's comments a bit diffferently than you do. I have already addressed that in two other posts here.

Bill;

In the 50s and before, it was common to measure pH with a Wheatstone Bridge and antiquated electrodes. Then we had to do conversions from salt effects, concentration offsets and temperature. So it could be said that in the "early days" pH was an approximation unless one did all of the corrections. In the 60s, Kodak converted to Beckmann pH meters with the top of the line electrodes and temperature compensation hardware. Not so much an error, but a refinement by several orders of magnitude. This usually resulted in a correction of 0.1 - 0.5 in pH which is huge but not unmanageable.

Sandy;

Kodak has indeed published more than one version of a formula and they did it for professionals. The formulas I see for D61a have apparent errors which are big, small and in-between. What does it mean? IDK. D61a is hardly used at all today and never was a big runner AFAIK. Why all the changes between Na2CO3 and Na2CO3.H2O? Why the odd replenisher? There are dozens of questions that dont even deserver an answer let alone experimentation.

We should move on to more modern or more useful developers.

PE

 

[sanking]

Sandy;

Kodak has indeed published more than one version of a formula and they did it for professionals. The formulas I see for D61a are big, small and in-between. What does it mean? IDK. D61a is hardly used at all today and never was a big runner AFAIK. Why all the changes between Na2CO3 and Na2CO3.H2O? Why the odd replenisher? There are dozens of questions that dont even deserver an answer let alone experimentation.

We should move on to more modern or more useful developers.

Ron,

I totally agree.The point is we should learn from the old formulas and then try to improve on them. I know that is what you do and I applaud your positive approach to these discussoins.


Sandy

 

[billtroop]

Finally, Ian, why not pick on Kodak instead of me? In all the publications I have seen from the 70s and 80s, Kodak always printed D-61a correctly. I do not of course take credit for this, but am glad to note they got it right in the end.