Boil Off / Dilution discrepancy

[Populuxe]

I am on Windows XP, BS 2.0.57

My calculated pre-boil gravity/volume is 1.037/10.37
Post-boil is 1.044/8.84

Now if I use dilution tool to go back from 8.84G to 10.37G the final gravity will be 1.038.

It could be either bug or a way to make sugar out of propane or ... I don't know.

I don't care too much about the dilution tool but want to make sure the boil of calculation is correct.

Thank you.

[BeerSmith]

Hi,
  I'll take a look at this but two things might be coming into play.  One is rounding error (1.037 vs 1.038 is not a huge difference).  The other is cooling loss (shrinkage).

Brad

[gizzygizmo]

I think the actual issue may be related to something we're seeing with preboil volume/OG.  It appears preboil OG is calculated based on the thermally expanded volume. 

If you take the preboil SG, multiply by preboil volume (displayd at 150degF), then divide by OG you will get your cooled post boil volume.  However; you measure preboil SG "at temp" which means you cool it down... so the volume is really 3% smaller.   Your SG is 3% "off" here due to the shrinkage calculation expanding the volume of preboil wort.

Set shrinkage to 0% and watch your preboil SG go up by a few points (5-6 points on a 1.070ish beer).   Now multiply preboil SG times preboil volume (this will be cooled volume, notice it's smaller now), divide by OG and you see the cooled volume matches.

[Oginme]

You are correct that the pre boil gravity is based upon the thermally expanded volume and that this creates a known error in BeerSmith prediction.  For most brewers, the measurement of volume and gravity are well within the error of the calculation, but it does throw off the sugar point balance when making dilutions or additions on the cold side.  I had brought it to Brad's attention some time ago, but it does not seem to be high on the list of corrections.

Note:If it were to be corrected it would occur in the next updates of BS3, since development is stopped on the BS2 version of the software.

[gizzygizmo]

Also, i've noticed the rounding... sometimes that will get you here.  1.03744 would round down vs 1.033751 roudning up.   So if it's very close and just barely rounding up to get your 1.038, then when you work backwards you may be just on the edge.

Like said above, even the shrinkage issue is only a 4% error... so most people don't even notice this unless they're converting recipes or working with odd volumes/losses/deadspaces, etc.