Efficiency discrepancy from measured pre / post boil

[Meddo]

G'day all,

This seems to me to be a fairly basic question but just want to check that I'm not overlooking anything.

I'm currently using the Brewer's Friend web software to design and reconcile my brewing numbers on a recirculating 1V system. This latest brew is my first since I've added volume markings to my kettle to enable me to accurately reconcile my equipment parameters.

The problem I've got is that the recorded numbers are giving me different efficiencies for pre-boil and post-boil measurements. My understanding is that these two efficiencies should be the same.

Specifically, I measured a pre-boil volume (after mash-out) of 30.5 litres at 95 deg, which according to the BF calculator is equivalent to 29.4 litres at 20 deg. The refractometer measurement (ATC) was 11.2 at ambient (28 deg) which with a wort correction factor of 1.02 gives me SG 1.044.

Post-boil volume was 22 litres at 21 degrees, which a simple boil-off calculation says should be at SG 1.059, however a measurement with a calibrated hydrometer (cal. at 15.6 deg) gave actual SG 1.054. The refractometer reading for this sample was 13.1.

Now the only sources of error that I can see are that either I mis-read the volume markings on the kettle for the pre-boil volume (possible since I haven't labelled the marks yet, but unlikely), or the refractometer is giving me a funky reading pre-boil. I weighed the FV sample and measured the kettle deadspace loss and so I am comfortable that the post-boil volume and SG are reasonably accurate.

Based on 34.7 litres total strike volume BF is giving me conversion efficiency of 106%, pre-boil efficiency 83%, post-boil efficiency 77%.

So my question now is whether there are any other potential sources of error that I am missing here? For my next brew I will double-triple-quadruple-check all the volume measurements, and take a hydrometer reading pre-boil to reconcile the accuracy of the refractometer. Is there anything else that could have gone wrong in my measurements, or any other efficiency losses that I have missed that would imply that my pre- and post-boil measurements are actually correct?

The only other possibility I can see is if all the margins of error happened to align correctly to give such a large discrepency (+/- 0.5 litres on volumes, +/- 0.01 on refractometer correction factor, +/- 0.001 hydrometer readings).

Anything else?

Thanks in advance,

 

[mtb]

I had this issue some time ago and determined it was the BF calculator I was using - this one http://www.brewersfriend.com/hydrometer-temp/
Have cooled my samples to room temp before measuring and haven't had a problem since.

 

[Lyrebird_Cycles]

Firstly: the ATC on a refractometer only works for sucrose solutions. It does not work accurately for wort. To get accurate results make sure your wort is at 20 oC when you measure it.

Nextly*: a conversion efficiency of 106% is immediately suspect: something is wrong somewhere.

Lastly, did you really boil off 7.4 litres? That's ~ 25% boiloff which seems very high.

* I din't think "nextly" was a word either. Turns out it is.

 

[Meddo]

Thank you both.

mtb: You could be right there, I'll have a look whether the adjustment factors from Carlton match what BF is saying.

LC: It seems that it would be pretty hard to confirm and keep a refractometer sample at 20 deg when ambient is 28+ deg, given how small / thin that sample is, unless you know any tricks? It will probably be easier for me to cool 100 ml for use in a hydrometer sample, at least that way there's enough fluid to use a thermometer to confirm the temperature at time of measurement.

What is a reasonable real-world range for calculated conversion efficiency? I assume that slightly higher than 100% calculated efficiency is possible if the max potential extract is underestimated, although I suspect this shouldn't affect anything by more than +/- 1%. I agree that 106% seems suss for my system, especially since I didn't sparge so I doubt that I would be achieving particularly high conversion efficiency relative to some other systems.

This is only my fifth AG and since I couldn't measure kettle volumes previously I don't really have a body of work to look through for trends.

Appreciate your assistance.

 

[Lyrebird_Cycles]

What is a reasonable real-world range for calculated conversion efficiency? I assume that slightly higher than 100% calculated efficiency is possible if the max potential extract is underestimated, although I suspect this shouldn't affect anything by more than +/- 1%.

I think either I've made a mistake or Brewer's Friend has, I can't work out how their "Mash Conversion Efficiency" calculator is supposed to work.

Specifically, I can't see where it takes account of the water absorbed into the spent grain. If you don't account for this it is indeed possible to get over 100% efficiency but I think the number is then meaningless.

 

[mtb]

I'd say Brewer's Friend has the problem due to a) my own issues with BF's calculator, and b] my complete lack of issues with your posts on AHB

 

[Meddo]

I think either I've made a mistake or Brewer's Friend has, I can't work out how their "Mash Conversion Efficiency" calculator is supposed to work.

Specifically, I can't see where it takes account of the water absorbed into the spent grain. If you don't account for this it is indeed possible to get over 100% efficiency but I think the number is then meaningless.

The mash conversion efficiency is calculated on the total water added to the MLT including sparge if done (not in my case). The pre-boil efficiency then uses the same SG reading but accounts for water lost to spent grain by using pre-boil kettle volume.

Edit: corrected 1V-specific terminology.

 

[Rocker1986]

I'm confused. My pre-boil efficiency i.e. mash efficiency, is always a higher number than my overall efficiency is, usually somewhere between 8 and 10%. Mind you, in Beersmith there is no figure for "post boil" efficiency - it's measured by the amount of wort that goes into the fermenter, not what's left in the kettle. Maybe they'd be the same if I included the 3 litres or so lost to trub, but to me that's not an accurate figure of the process from raw ingredients to wort in the FV, because it is including a volume that doesn't end up in the FV.

At the end of the day when you construct a recipe you are aiming for a certain volume in the fermenter at a certain SG. It makes more sense to me to use these figures to work out the overall efficiency, not include an amount that isn't used.

 

[n87]

pre-boil and post-boil eff should be the same. there is the same amount of sugar. that would be my guess as to why Beersmith doesnt have a post boil.
Whereas the eff changes when you put it into the fermenter cause you have losses

 

[Rocker1986]

Yes having a play just now I have to increase the "batch size" by about 2.5-3 litres to get the efficiencies to be basically the same as each other. I'm used to including all the losses into the fermenter having done it like that for the last 4 years though so it still works for building recipes either way.

 

[Meddo]

Yeah in principal that's correct mtb, BF provides for both post-boil (in kettle) efficiency and overall brewhouse efficiency allowing for trub / kettle deadspace losses. The attached "raw" image shows the measured volume, SG, and efficiency figures at each step for the brew under discussion.

My brewhouse efficiency seems to consistently be somewhere around mid-60's, with about 5% of the missing efficiency being due to leaving 3-4 litres in the kettle (single batches in a 70 litre stockpot currently results in a fair bit of trub wastage). It's on my to-do list to reduce that amount to improve overall brewhouse efficiency.

My interest in getting each of these individual efficiencies identified for my system is basically to get my mash water volumes correct, and to identify where I'm losing sugars along the way, whether that's in poor conversion, high grain water losses, high kettle deadspace losses, or something else.

This particular brew was intended to deliver 23 litres of 1.044 wort at 66% brewhouse efficiency. I actually got 19 litres of 1.054 wort at 66% brewhouse efficiency. So the brewhouse efficiency on its own isn't enough for me to hit my recipe targets. The problem here happens to be that I added a second element to the kettle, so the boiloff rate went through the roof compared to my previous brews...

The other issue for me is just that it's frickin frustrating when the numbers don't add up like they should. As an engineer, that's an itch that I can't leave alone :huh:

It does seem like the problem must have been with the refractometer reading of the of the pre-boil wort. The "corrected" image below shows the efficiencies generated by using a pre-boil SG back-calculated using a simple boil-off calculation from the post-boil SG and volume. The numbers in that case look much more reasonable IMO.

 

[contrarian]

How did you measure the markings on your kettle?

Weight is probably the most accurate way as most jugs etc and even flasks can be out by a decent amount.

I normally just use a ruler and a calculation to measure my volumes. For my pot 1cm is pretty close to 1.6 litres.

If your pre boil and post boil efficiencies don't agree then it is a measuring of some kind. There just isn't anywhere else for the sugar to go!

 

[Meddo]

I used a five-litre Erlenmeyer flask to measure 5 litres and 65 litres into the kettle and interpolated between those points. I confirmed (-ish) using the weight, although I don't know the accuracy of the bathroom scales I used so it was a bit of a close-enough thing. Also confirmed as being pretty close by draining into a 30 litre FV with volume markings. I doubt it would be out by more than about +/- 0.5 litres at full, and any error there should carry through consistently at each step I would assume.

As mentioned earlier I suspect though that there may be some imprecision in my reading of the volumes due to temperature, foam, angle of eyes, meniscus etc. Probably not enough on its own to account for the discrepancy, even with an alignment of the stars type situation. But it may have compounded with errors from the initial refractometer reading to give the large total error. Especially since the actual brewhouse efficiency was bang-on consistent with my previous few brews.

Cheers,

 

[Lyrebird_Cycles]

The mash conversion efficiency is calculated on the total water added to the MLT including sparge if done (not in my case). The pre-boil efficiency then uses the same SG reading but accounts for water lost to spent grain by using pre-boil kettle volume.

Edit: corrected 1V-specific terminology.

Yes I understand that, my problem lies in working out how that figure is in any way useful. I guess as long as it's consistent from batch to batch it tells you something.

 

[Meddo]

Sorry LC, wasn't meaning to lecture you, I'm not that familiar with the literature as to whether BF is using its own terminology for some parameters so thought that may have been where your question lay.

Probably those interim efficiencies aren't all that relevant other than being an alternative way of presenting the data regarding losses through the system. As in, the fact that my calculated pre- and post-boil efficiencies are significantly different highlights easily that there is likely an error in my measurements somewhere along the way (given most replies here are confirming my assumption that there's no possible loss of sugars simply by boiling). The mash conversion efficiency tells me if I've missed out on extracting some sugars by buggering up something in the mash - pH, crush size etc. At least that's my understanding of it? And also a sanity-check, as you said the 106% number does imply an error somewhere in my view.

Cheers,

 

[contrarian]

Efficiency gets confusing because people use the same term for different things.

Understanding conversion is important. As you say it will tell you about any issues with crush, pH etc etc.

The rest of efficiency losses should really be able to be calculated based on losses to trub etc.

There's plenty of factors that can contribute to measurement errors, especially when measuring hot wort! I normally see brew day refractometer measurements as a guide and don't really worry about a few points either way but I'm probably not the most precise brewer around either!