LPD Powder Developer Mixing

[brass majestic]

I see that B&H and a couple others are selling LPD powder in a huge tub container - enough to make 25 gal.
Now would that be 25gal of stock solution or working solution? My other question is; can one successfully divide the contents into 25 equal portions of powder (by weight) and be assured that each portion contains exactly what one would get, chemistry wise, in one of Ethol's 1 gal size can of powder?

 

[MattKing]

can one successfully divide the contents into 25 equal portions of powder (by weight) and be assured that each portion contains exactly what one would get, chemistry wise, in one of Ethol's 1 gal size can of powder?

No.
There is no way that you can assure uniformity.

 

[Kino]

I used to mix LPD in 25 gallon batches for the OSU Department of Photography Still Labs in a Pako mixer and pump in up into overhead holding tanks...

Loved the smell of that developer!

 

[Leigh B]

can one successfully divide the contents into 25 equal portions of powder (by weight) and be assured that each portion contains exactly what one would get, chemistry wise, in one of Ethol's 1 gal size can of powder?

No... NEVER... not even on a good day with a strong tail wind.
You can't even divide powder volumes (of multi-component formulas) in half, or in any other portions.

There's no way to control the proportions of the individual chemicals.

- Leigh

 

[tezzasmall]

You can't even divide powder volumes (of multi-component formulas) in half, or in any other portions.

There's no way to control the proportions of the individual chemicals.

- Leigh

Do you mean we can't take a formula and even with really accurate scales, make a smaller proportion of a formula e.g. half amounts? OR as I've just thought, do you mean not to try and divide up a PREPACKED developer like D6 / ID11 etc? That would make more sense to me.

I have done it in the past with good results.

Terry S

 

[Pentode]

Once powdered chemicals are mixed together, pre-packaged or not, you can't divide the resultant total quantity with any guarantee that the proportions will be correct.

If you're mixing your own ingredients then, of course, you can reduce quantities while keeping the proportions the same, but once powders are mixed together they can't be divided.

 

[Leigh B]

I have done it in the past with good results.

Terry,

That means nothing.

The fact is that you cannot proportionately divide pre-mixed chemicals with any degree of accuracy.

A developer contains many ingredients, some of which are included for stability or longevity.
Dividing such a mixture may change those constituents without affecting its immediate usability.

- Leigh

 

[Gerald C Koch]

Having performed many quantitative chemical analyses I can testify just how hard it is to get statistically consistent results from what are thought to be representative samples. When you subject an evenly mixed sample consisting of different particle sizes, the larger particles work toward the top of the mix. With premixed products such as developers this happens very easily during transport. Companies like Kodak spend big bucks on powder mixing machines to insure that their product is uniform when it leaves their care. Once this occurs all bets are off. You have to mix the entire contents to insure a predictable developer. Despite how many times you have successfully subdivided a bag is no guarantee that the next time will be correct. This is the classical statistical fallacy. Believing that past results somehow determine future results. Flip a coin 20 times and it is always heads. What are the chances that the next result will also be heads. The odds are the same as before 1 in 2.

 

[Maris]

Having performed many quantitative chemical analyses I can testify just how hard it is to get statistically consistent results from what are thought to be representative samples......

I'll confirm that. A regular laboratory challenge was fire assay of gold bearing gravel. Too many big chunks and the assay was low. Too much fines and the assay was high. Millions of dollars of speculator's money could ride on the result. I even had devices called riffle splitters to help randomise sub-samples. Statistically, the smaller the sub-sample the more it was likely to be non-average. You could (theoretically) mix up half of a well randomised powder developer and expect a workable result. Mix up a tenth at a time and no aliquot would be quite the same. The mathematics of all this is rather interesting but it doesn't get photographs made.

 

[grahamp]

This was recently addressed in (if I have the link right): (there was a url link here which no longer exists)
Quite a few of us, it appears, have worked or are working, in industries where representative sub-sampling is an issue.