Kettle boil off rate

The Homebrew Forum

Help Support The Homebrew Forum:

hoppyscotty

Landlord.
Joined
Oct 13, 2020
Messages
507
Reaction score
234
I'm just in the process of setting up and establishing the equipment profile for a 3 vessel system I got off eBay that consists of a 20 gallon mash tun, 20 gallon kettle and 30 gallon HLT (could interchange the HLT and kettle to suit batch size but initially starting out with a 20 gallon kettle). I did a boil off test last night for an hour and found I lost 6 litres - I guess it could be a bit less than 6 litres as I ran the whirlpool as I was bringing the water to a boil to circulate it so will have lost a few hundred ml in the pump and pipework. I did this by bringing 40 litres (kettle about 60% full) upto the boil, boiling for 1 hr and allowing to cool overnight and looking on the sight glass to see water loss, which was 6 litres.

As an aside, I noted that once I had brought the water to a boil and noted the level, then again at the end of the 1hr boil and read off the loss it was closer to 5 litres, so not sure to assume the boil off loss as observed with cold water or hot?

This sounds a bit high, certainly higher than I was expecting. Does it sound right compared to others experience?

My only datum is my 35ltr Brewzilla that had 3 litres/hr in the default equipment profile in Brewfather which seemed to be about right in my experience. I'm assuming if I were to swap the 20 gallon vessel for the 30 gallon vessel then I'd be looking at another couple of litres of boil off due to the larger diameter of the vessel.

Thanks
 

Agentgonzo

Landlord.
Joined
Aug 23, 2021
Messages
791
Reaction score
849
As an aside, I noted that once I had brought the water to a boil and noted the level, then again at the end of the 1hr boil and read off the loss it was closer to 5 litres, so not sure to assume the boil off loss as observed with cold water or hot?
Water contacts as it cools. So the same amount of water will have a greater volume straight after the boil then it will after it's cooled.

You can probably look up the density of water at different temperatures to figure out the percentage it shrinks.

6L boil off from a 20 gallon batch seems a very small amount to me
 

DocAnna

Scottish Brewster
Supporting Member
Joined
Aug 24, 2020
Messages
1,363
Reaction score
2,678
Location
Scotland
Sorry to be the geek, there is only a very small change in water density between cool ground temperature and near boiling about 4%
1656422375559.png

The other measures that need to be taken into account are that the vessel will expand which increases its volume, and makes the volume look lower than it is on a fixed scale than it would be otherwise. However your volume loss appears to have increased from leaving it overnight, which very likely reflects a further 1 litre loss through evaporation as it cooled.

The loss per hour is a function of the vigour of the boil, the surface area of your boiler, your altitude/atmospheric pressure, whether you cover the vessel during the boil, and vapour pressure where you are boiling it. I suspect the latter point means that an outside boil on a windy day would lose more than that inside a home or shed, and will vary with ambient temperature too. More importantly for calibration is that the loss from boiling wort is lower than for boiling water, this is because the boiling point of water is lower than of wort, more so with higher gravity beers. This means the vigour of the boil will be less with higher gravity beers, and will progressively reduce as the beer concentrates later in the boil even when the power input stays the same.

Bringing this all together means that measured boil off with tap water will always be an over estimate, and should be estimated from the difference between volume at near boil and shortly after boil.
 

hoppyscotty

Landlord.
Joined
Oct 13, 2020
Messages
507
Reaction score
234
Thanks all...and I don't mind a bit of geekery at all!! I don't mind the boil off rate being slightly over estimated as it means the compensation in the recipe will mean I am covered and wont come up short. Makes sense about all the variables associated. I did replicate exactly the process and conditions I would normally do on a typical brewday so hopefully I achieved a like for like as far as the process goes and will have to live with all the seasonal variables I normally tolerate as I brew in my garage with the garage door open.

Surprised that the readings should be when the water is hot i.e. the lower reading as we tend to target volume in the fermenter so once you've cooled the wort before pitching into the fermenter you're going to lose a bit of volume so this apparent volume loss must be factored into the calculation somewhere? It appears so from experience.

Also if wort has a higher boiling point than tapster then maybe I should be setting my boil temperature to something like 101 degrees C?

Think I'll just plug in 5 litres relative to the 'hot boil off rate' and go with it initially and see how my first couple of brews turn out.

Thanks.
 

Agentgonzo

Landlord.
Joined
Aug 23, 2021
Messages
791
Reaction score
849
Sorry to be the geek, there is only a very small change in water density between cool ground temperature and near boiling about 4%
So even in a 30L volume, 4% is 1.2L, which is significant compared with ~3-5L of boiloff for these sized batches, and 1.6L in the OP's 40L test batch, which is roughly the 1L disparity they're seeing.
 

DocAnna

Scottish Brewster
Supporting Member
Joined
Aug 24, 2020
Messages
1,363
Reaction score
2,678
Location
Scotland
So even in a 30L volume, 4% is 1.2L, which is significant compared with ~3-5L of boiloff for these sized batches, and 1.6L in the OP's 40L test batch, which is roughly the 1L disparity they're seeing.
Except that the 1 litre variation in the OP is in the wrong direction. Cold volume of 40 litres, to a hot volume of 41.6 litres, lose 5 litres (hot) to boil to take to 36.6 litres, allow for 4% expanded volume, yields 35.2 litres. So the loss from cold to cold would be 4.8 litres, but was observed to be 6 litres, ie the loss was somewhere else, likely evaporation after the boil.
 
Joined
Nov 13, 2013
Messages
4,559
Reaction score
3,740
I'm just in the process of setting up and establishing the equipment profile for a 3 vessel system I got off eBay that consists of a 20 gallon mash tun, 20 gallon kettle and 30 gallon HLT (could interchange the HLT and kettle to suit batch size but initially starting out with a 20 gallon kettle). I did a boil off test last night for an hour and found I lost 6 litres - I guess it could be a bit less than 6 litres as I ran the whirlpool as I was bringing the water to a boil to circulate it so will have lost a few hundred ml in the pump and pipework. I did this by bringing 40 litres (kettle about 60% full) upto the boil, boiling for 1 hr and allowing to cool overnight and looking on the sight glass to see water loss, which was 6 litres.

As an aside, I noted that once I had brought the water to a boil and noted the level, then again at the end of the 1hr boil and read off the loss it was closer to 5 litres, so not sure to assume the boil off loss as observed with cold water or hot?

This sounds a bit high, certainly higher than I was expecting. Does it sound right compared to others experience?

My only datum is my 35ltr Brewzilla that had 3 litres/hr in the default equipment profile in Brewfather which seemed to be about right in my experience. I'm assuming if I were to swap the 20 gallon vessel for the 30 gallon vessel then I'd be looking at another couple of litres of boil off due to the larger diameter of the vessel.

Thanks
On a home brew scale 10 to 15% is the boil off rate so you are in the ball park, I lose 5 litres on my 70 litre kettle and 4 litres on my 40 litre kettle, owing to the difference in surface area.
 

RoomWithABrew

Landlord.
Joined
Apr 19, 2021
Messages
1,001
Reaction score
622
Location
Paremata New Zealand
@DocAnna
Great info, a lid placed at the end of the boil would have prevented the added evaporation loss.
Sorry to be the geek, there is only a very small change in water density
Sorry to be the pedant
The change is 4% as you say, adjective small not required.
You have really got me thinking about that vessel expansion aspect though. With a different shaped vessel perhaps with a portion jutting into the kettle that expansion in would compensate for the expansion of the rest of the kettle out.
Not suggesting a new design just a virtual perfect vessel, guessing the perimeter jutting in would need to equal the remaining circumference or perimeter of the hypothetical kettle.
Thank goodness we don't have to be that accurate to make beer!
 

DocAnna

Scottish Brewster
Supporting Member
Joined
Aug 24, 2020
Messages
1,363
Reaction score
2,678
Location
Scotland
@DocAnna
Great info, a lid placed at the end of the boil would have prevented the added evaporation loss.

Sorry to be the pedant
The change is 4% as you say, adjective small not required.
You have really got me thinking about that vessel expansion aspect though. With a different shaped vessel perhaps with a portion jutting into the kettle that expansion in would compensate for the expansion of the rest of the kettle out.
Not suggesting a new design just a virtual perfect vessel, guessing the perimeter jutting in would need to equal the remaining circumference or perimeter of the hypothetical kettle.
Thank goodness we don't have to be that accurate to make beer!
The expansion of a vessel due to heat and reduction in fluid level can easily be observed when you pour boiling water into a mug. within the first few seconds you can see the fluid level visibly drop as the mug expands a little. This is more obvious with a pottery mug than a porcelain thin walled mug as the time to heat up is fractionally slower which gives you time to see the change.
 

RoomWithABrew

Landlord.
Joined
Apr 19, 2021
Messages
1,001
Reaction score
622
Location
Paremata New Zealand
That volume " loss* after pouring also partly due to the greater drop in temperature seen with a pottery mug. The pottery mug has more mass so will cool the water more and hence the thermal coefficient of expansion ( contraction in this case of liquid ) will be more.
I'm not convinced the majority of volume loss seen is due to mug expansion, more liquid contraction.
 

DocAnna

Scottish Brewster
Supporting Member
Joined
Aug 24, 2020
Messages
1,363
Reaction score
2,678
Location
Scotland
That volume " loss* after pouring also partly due to the greater drop in temperature seen with a pottery mug. The pottery mug has more mass so will cool the water more and hence the thermal coefficient of expansion ( contraction in this case of liquid ) will be more.
I'm not convinced the majority of volume loss seen is due to mug expansion, more liquid contraction.
Fair point, and I suspect both factors are at play. The speed of change and degree of change certainly seems more than the just over 1% contraction which would be anticipated with the first 20 deg temperature reduction. We need a PhD student with too much time on their hands to measure this experimentally 😁 .
 

RoomWithABrew

Landlord.
Joined
Apr 19, 2021
Messages
1,001
Reaction score
622
Location
Paremata New Zealand
@DocAnna
I have spent the last couple of hours looking at it. No way am I going to try and show my working typing on a mobile. I'm over here in UK and away from keyboard.
Summarising for a 65 litre vessel of stainless steel volume change between 20 Celsius and 100 is 155ml.
Height change would be 0.48mm, diameter increase 0.3 mm.
So I'm pretty sure we couldn't measure that.
Incidentally if mug 9cm high and 8cm wide made of polythene which has much greater coefficient of expansion than SS.
Capacity would increase from 452.39 to 474.45 cm3 that would be a height change of mug of 1.44 mm between 20 and 100c. But we'd likely fill it with the larger figure. So level would appear to drop though if your 452.39 at boiling temp went into mug and it stayed at that temp.
Interesting on cooling volume of liquid would reduce by 18ml but vessel would shrink by 22ml. So actually would appear to have more in it by 0.79mm.
But if mug was SS vol increase of mug with temp at 100 would be 1.75ml.
If you could do it you'd see a 0.072 mm mm drop in level of the liquid.
As the liquid and mug cooled to 20 then you'd see a 3.18 mm drop in level.

The coefficient of expansion for a pottery mug was too complicated for me.
However in real terms we fill a mug to a level with hot liquid which is different to brewing where we bring a volume from cool to hot.
 

hoppyscotty

Landlord.
Joined
Oct 13, 2020
Messages
507
Reaction score
234
This is why it doesn't make sense to me to measure the boil off rate hot vs cold as you want to take into account all the variables in the system. All I'm interested in is what volume I end up with in the fermenter at circa 20 degrees C. I observed a cold to cold volume reduction of 6 litres vs a hot to hot volume reduction of 5 litres so the additional volume loss of the cold to cold observation will be a combination of liquid contraction, vessel contraction plus some additional evaporation as the liquid cooled from boiling to room temp plus some volume loss in the hoses from the vessel to the pump and back to the whirlpool port, so all of these factors need to be taken into consideration when I'm looking to achieve a target volume in the fermenter...though in reality I'd recover the volume in the hoses.
 
Top