Old traps for new players

[PeteB]

My official status on this forum is "Advanced Member" but I am just a beginner. I only had limited experience in other distilleries before starting production in my own distillery 7 months ago.

Despite having a science degree and having spent countless hours in a lab, I was becoming very frustrated with variable readings on my hydrometers when measuring the same batch.

I eventually noticed liquid droplets sticking to the sides of the hydrometer stem above the liquid level. This would slightly increase the weight, incorrectly indicating higher alcohol.

Even though I regularly wipe the hydrometers after use, there was still an oily film preventing liquid running off. I had to thoroughly clean with detergent to get rid of the oil film. With a bit of experimenting I found I was getting up to 0.3 ABV variation (0.6 Proof) caused by the liquid droplets. I now also lower the hydrometer carefully so it won't go much below the assumed reading, that way there is no liquid film on the glass above the meniscus.

It is probably very obvious I did not attend distilling school!

Are these things taught at distilling school?

And here is another trap I found the hard way.

When diluting spirit I was finding problems hitting the target. Someome on this forum recently said "blending is a bitch" I WAS in agreement.

I know how to do the calculations so I was sure that was not the problem.

Recently I added the correct amount of water, I stirred thoroughly and re-tested immediately with thermometer and clean hydrometer. The alcohol concentration was still too high which meant I needed to add more water. Did another calculation and added a small amount of water to correct. When I checked a few days later the alcohol strength was too low! Too much water added. The proof was probably correct the first time.

Has anyone else found this problem?

Initial enquires suggest the mixture does not stabilise for some time.

I now add the calculated amount (by mass) of water and leave for a day or so. I am hitting the target dead centre.

[will]

Is that taught in distilling schools? It is in some, but you might not catch it - it goes by pretty fast.

For those of us who are stateside, exactly that procedure is in 27 CFR 30.23 of the Gauging Manual.

I'm curious, do your calculations include a correction for the contraction that occurs when ethanol and water are mixed?

[Gwydion Stone]

When diluting spirit I was finding problems hitting the target. Someone on this forum recently said "blending is a bitch" I WAS in agreement.

I think that was me.

In the words of Erik Martin at Artisan Spirits: "Proofing Is A Bitch."

When I checked a few days later the alcohol strength was too low! Too much water added. The proof was probably correct the first time.

This could be due to insufficient mixing at the time. You'd be surprised how pockets of water can float around in there, un-blended; especially in a large tank. You really have to churn it up to get it fully blended and get an accurate reading.

[delaware_phoenix]

One problem you may be running into is not letting your temperature stabilize before testing. If you have taken spirit from a place that is colder or warmer than the ambient temperature where you are proofing, the temperature in changing in the proofing bottle. If you don't let it stabilize (which may take 5, 10, 15 minutes) you won't have the right temperature and so even if you have the right proof from the hydrometer, you won't have the right correction factor for temperature. And you'll be off.

Also stirring as Gwydion mentions may not be sufficient.

Also if you have spirit in the tank and withdraw your sample, after adding water, the liquid in the spigot may not be will mixed and may not accurately reflect the proof in the tank. Withdraw a good amount, say a liter and pour it back into the tank. Now draw another liter and take your sample from that (say 250 ml) for proofing or testing the proof. Now you have a much better chance of having an accurate sample.

Check the accuracy/resolution of your digital thermometer. If it's only good to ±1º C, you can't get a correct value as you might be off nearly 2º F. Many digital thermometers will get to only 65% of the accurate reading with 2.5 seconds and may take up to 5 times that to get to 95% of the accurate reading. So you need to let that temperature reading stabilize.

Also don't forget that adding water generates heat and may cause the temperature of your spirit to rise, so another reason to let the temperature stabilize when taking the proof.

Temperature was my biggest problem when first proofing.

[PeteB]

I didn't want to put too much in my first post on this thread, I often have trouble with all my typing vanishing for no apparent reason.

Thorough mixing is certainly necessary. And as Delaware Phoenix points out the mixing produces heat, it is exothermic. The mixture slowly warms up then cools again to room temperature.

The thermometer and hydrometer will be lagging a little behind, maybe at different rates which would confuse any attempts at corrections.

I did a bit more experimenting and did a small dilution in a glass cylinder.

I watched closely and noticed each time I stirred the mixture, a lot of very fine bubbles were released. Bubbles in a solution will increase its volume and hence REDUCE its density. But at the same time lots of these bubbles were sticking to the bottom of the hydrometer causing lift, giving the appearance of INCREASING density. Spinning the hydrometer got rid of the stuck bubles but not for long.

Also, as Will says above, when diluting ethanol with water, volume shrinkage occurs. ie, 1 litre of alcohol + 1 litre of water equals less than 2 litres of mixture.

The next bit is MY THEORY ONLY. IF shrinkage is not instantaneous, the initial volume will be larger which means the density is lower. A hydrometer will sink further, mimicking a higher %alcohol

As the volume shrinks gradually the density will increase. MY THEORY ONLY

Maybe this is fact! Does anyone know?

[Jake Norris]

Cutting proof can be a real bear... and I know nothing is worse than over cutting and having to go harvest more spirit to bring you proof back up to the desired level. To avoid this issue I go slow! I start cutting early to provide plenty of time to cut with out rushing. I make proof measurements on the cask strength, adjusting for temperature, and do the math to cut to desired proof and then cut to 80% of my target proof. Let it marry for a day or so, and then remeasure and repeat. I make sure that I have strong circulation in the tank and I have even built a bubble screen to cause extreme agitation. Alcohol absorbs water (hygroscopic reaction) during marrying as well as create heat(exothermic reaction), so I make sure to take new readings of volume, proof and temp each time I effect the spirit. I don't add more water until I get the same proof reading 3 times in a row.

Some fun cliches I like.....

- Measure twice cut once.

- Haste makes waste.

- When in doubt leave it out.

[fotoski]

Someone correct me if I am wrong, but I have been told that the best way to ensure accuracy is to proof down using weight.

X amount of water always weighs the same regardless of temperature. Y amount of alcohol always weighs the same, regardless of temperature. X weight of water + Y weight of spirit = X+Y weight at any temperature (below boiling).

If you convert your calculations to weight, there should be no fluctuations, and you are only dependent on the accuracy of your measurements and your proper use of formulas to convert to weight.

[PeteB]

Someone correct me if I am wrong, but I have been told that the best way to ensure accuracy is to proof down using weight.

X amount of water always weighs the same regardless of temperature. Y amount of alcohol always weighs the same, regardless of temperature. X weight of water + Y weight of spirit = X+Y weight at any temperature (below boiling).

If you convert your calculations to weight, there should be no fluctuations, and you are only dependent on the accuracy of your measurements and your proper use of formulas to convert to weight.

I totally agree with you Drew. I am sure Will is also a fan of mass measurements.

The only volumetric device I use in my distillery is a 60 mL syringe for filling sample bottles.

Unless a distillery and its liquids and measuring containers are all at "standard temperature" I think it would be difficult to get accurate measurements.

It is possible but not that quick to calculate liquid expansion or contraction, but how does one allow for expansion of the measuring vessel?

[PeteB]

Some fun cliches I like.....

- Measure twice cut once.

Jake, are your blending vessels small enough to place on platform scales?

I don't want to sound rude, but I see why you would like to "measure twice cut once". With your technique you appear to measure about 12 times and cut about 4 times.

And from other snippets on this forum I am sure you are not the only one who does it that way.

Cutting is so easy on scales.

Once I am sure I know the corrected proof of the spirit, I put the numbers into ALCODENS, pour in the calculated mass of spirit then make up to final mass with water. As long as initial proofing was correct I hit spot on every time.

If you are a bit wary of overshooting then hold back a little of the water then test the next day. You should find you can then toss the rest of the water in.

If for some reason you did put too much water in then it is very easy with ALCODENS to work out how much spirit to add to correct it.

Does anyone know if it is possible to put a screen grab into a post? If so, I can show you how to do the calcs.

ps I am not an Alcodens salesperson, I purchased a copy.

[Rich Morgan]

I totally agree with you Drew. I am sure Will is also a fan of mass measurements.

The only volumetric device I use in my distillery is a 60 mL syringe for filling sample bottles.

Unless a distillery and its liquids and measuring containers are all at "standard temperature" I think it would be difficult to get accurate measurements.

It is possible but not that quick to calculate liquid expansion or contraction, but how does one allow for expansion of the measuring vessel?

Most quality volumetric flasks are calibrated at a given temperature, and come with temperature to volume compensation tables. Temperature effects volume based assay the same as density assay, but to a lesser degree. My only advice is to make temperature calibration and correction integral with your tests. I took an easy path, and purchased hydros and flasks all calibrated for 22C and then set the aircon in the lab to exactly 22C. But even then when I get tank samples that need to be tested in a hurry, you still must temperature compensate.

A electronic scale with the accuracy and repeat-ability to reliably perform ABV tests are expensive, hence most of use either Hydros or IR/UV Spectro.

rich..

[Palmetto Coast]

Most quality volumetric flasks are calibrated at a given temperature, and come with temperature to volume compensation tables. Temperature effects volume based assay the same as density assay, but to a lesser degree. My only advice is to make temperature calibration and correction integral with your tests. I took an easy path, and purchased hydros and flasks all calibrated for 22C and then set the aircon in the lab to exactly 22C. But even then when I get tank samples that need to be tested in a hurry, you still must temperature compensate.

A electronic scale with the accuracy and repeat-ability to reliably perform ABV tests are expensive, hence most of use either Hydros or IR/UV Spectro.

rich..

How specific would an electronic scale need to be? I guess it would depend on the volume being measured. We had looked at a pallet scale from Dissonance, as they are local. They were rated within 0.5 lbs, but he found the reliability slightly better. What scales are others using?

[PeteB]

Rich, sorry, I think I might have confused you.

I was not talking about laboratory techniques to measure ABV or Proof ie by very very accuratly measurement of mass and volume, I do that with a hydrometer and thermometer.

I was talking about cutting spirit for barreling or botteling.

For this I am weighing between 100 and 300 Kg on platform scales to an accuracy of(I have edited this to correct) 0.05 Kg. So at 100 Kg I have an accuracy of +/- 0.05% which our excise people have accepted.

I will also point out that if I tried to weigh only 1 Kg on these scales the accuracy would be down to +/- 5% which is probably not good enough? I must re-read our regs!

My scales need to be re-certified every 12 months.

[PeteB]

How specific would an electronic scale need to be? I guess it would depend on the volume being measured. We had looked at a pallet scale from Dissonance, as they are local. They were rated within 0.5 lbs, but he found the reliability slightly better. What scales are others using?

I think this is where you will find what you need to know in USA

§ 30.44 Weighing containers. Gauging Manual.

[tl5612]

re-visiting an old topic, but Pete, i am looking into going down the proofing by mass route and wondered if you could shine some light on your scales set up (this might sound a bit stupid, but i trying to design something efficient and easy).

indeed, i'll be working on a pretty small scale, and so this looks like it could work quite well. 300kg would be around the maximum for a bottling run.

i've been looking at platform scales with an accuracy of 0.05kg, and wondered what you would recommend?

i ask this mainly because i've noticed that these type of scales all seem to be raised about 15cm off the floor - and so practicality of lifting a full drum/tank up to the platform could be a bit tricky. do you use a forklift in one go (with forkliftable tanks of course), or have some other simple method?

alternatively, i could get a cheaper set of scales that go up to 50kg (perhaps slightly more accurate than the big ones) and weigh everything in smaller containers, before adding to a standing tank... and just add it all up).

any input would be great.

thanks

[PeteB]

It would be much easier to weigh spirit then add the water once. Weighing smaller amounts on those scales will be less accurate. Yes, it can be awkward to get tank onto the platform. I use several methods. If you can tip the tank over a bit by hand then place a block of wood under it then push the tank up straight and slide onto scales. Or lift tank with levers and chock up higher than scales then slide scales under etc. forklift would be safest but take care not to bump the readout. I use up to 200 litre (approx 50 G US) they are a little bit of a struggle but I can get them on by myself with the block of wood. With 2 people it is easy.

[tl5612]

thanks very much.

love the idea of using levers!

what is quite annoying is that i can find plenty of EC trade-approved, 300kg scales with divisions of 50g, but they are only certified to divisions of 100g (i am from the UK). so not sure if this is a suitable option... no doubt when i look harder, i will discover the more accurate scales are very, very expensive.