Sunday, 2 May 2010

Whitbread colour cell

I knew I'd seen this. And that if I kept on going through the Whitbread Gravity Book I'd come across it. The size of the cells used by Whitbread for their colour measurements.

Take a look:

The SB Pale Ale has 1" cell (that's one inch) written above the colour value of 32. Brown Stout has 1/16" (one-sixteenth of an inch) above it's value of 1 + 17.

That explains why the numbers for Stouts always look way too low: because the measurements were made using a much smaller cell than for paler beers.

Glad that's been explained.


Zythophile said...

The dirty bastards!

Ron Pattinson said...

Zytho, I know, I know.

Talking of dirty bastards, I'll be posting about London filth in a couple of days.

Barm said...

You mean those cards you find in telephone boxes? Never seen them in any other city.

Graham Wheeler said...

Glad that's been explained.

Naw! Too simplistic. The numbers do not make sense which ever way you look at it. I've looked at it time and time again and I've never been able to make sense of Whitbread stuff once they start using those damn red glasses, particularly the number 40.

In the example you have given, using a 1/16" path length, you are in the absurd situation where the stout is a lighter colour than the brown ale. The brown should be typically 100 units and the stout 250 ish.

If you ignore the red glass, and miltiply the brown glass (17) by 16, we get 272 for the stout. Fair enough. But multiply 21 by 16 and we get 336 units for the brown ale - unlikely - and we have not taken the red glasses into account.

A 1-unit red glass will be a barely perceptible tinge of red and can be ignored or just added to the brown, and will give results close enough for government work. But a 40-unit red glass should be opaque, so that does not make much sense for a visual comparator if you can't shine a light through it.

Simply compare the same beer pre 1955 and post 1955. The post 1955 will be EBC. The Truman beers table posted on 2nd April is one example. Trubrown 1954 is given as 17+40. Trubrown 1955 is given as 105 EBC. Sit down with a calculator to try to make 17+40 match somewhere around 105 and you'll still be at it a week later, no matter path length you choose.

Nope. Something odd goes on once they start using red glasses. There is some other scale factor involved; certainly for the No. 40 red (Particularly as the Lovibond red glasses only went up to 20).

You will need the Whitbread procedures manual to get to the bottom of it.

Ron Pattinson said...

Graham, the Brown Ale wasn't measured using a 1/16th inch cell. Just the Stouts.

Graham Wheeler said...

Graham, the Brown Ale wasn't measured using a 1/16th inch cell. Just the Stouts.

That really poses more questions than answers. One being that with just one unit of red, which should make a very small difference, they could have normalised the stout readings to 1", as per the standard procedure. Why didn't they?

With the intermediate ranges where they (usually) hold the brown at 40 and vary the red (the opposite to what I said earlier) it is probable with so much more red that normalising to 1" will not be representative of the real world. The closest is to assume a half-inch cell, un-normalised, but it still does not really match comparable values where the data exists.

Basically they have three different measurement methods that are not directly comparable. It seems very unscientific.