Hail The Mighty Thumper

[haggy]

Thanks for asking. Nice to see you are interested. I have used your hobbybrennen pot still calculator and my pot model calculations agree with yours exactly.

The heat loss in the thumper for each run time increment is taken into account by the following:

I calculate a latent heat ( LH ) of condensation / vaporization (calorie/gal) for the incoming vapor ABV and the outgoing vapor ABV. The incoming vapor always is weaker in alcohol, so its latent heat is higher. Then, if there were no heat losses, a heat balance gives calories in equals calories out:

cal/gal (LH pot vapor) x gal pot vapor in = cal/gal (LH thumper vapor) x gal thumper vapor out

So the no heat loss thumper outgoing gal vapor amount would equal the ratio ( > 1 ) of the the pot to thumper latent heats times the incoming vapor amount.

But with heat losses, the outgoing vapor would be less. This is seen in all of the run data I have studied. I examined run data from many real cases ( you have seen my posts ) and there is a heat loss factor that is common to most all runs. After an initial period of time to equilibrate, the factor is fairly steady as the run progresses. I multiply the factor (probably the most important part of the model) times the no heat loss outgoing vapor to get an actual outgoing vapor amount for each increment of time.

It took a lot of study and trial and error, but I think I have found a realistic heat loss factor common to most pot/thumper runs. Yes, all thumpers are different. But by using some of the model options, you will vary the heat loss factor and this allows one to fine tune the model to their thumper. The initial condensation option, for example, adds some cooling to reduce the vapor outflow, so I reduce the heat loss factor by a certain amount.

If you have good complete data on some thumper runs, please send it to me and we can check out my model. We can work together if you are also developing a pot/thumper model.

[Hügelwilli]

For the heat loss I probably would use the heat exchanger formulas and look what heat transfer coefficient gets the calculation to realistic results. I tried a long time to calculate liebig type condensers, it is impossible because our condensers doesn't have fixed heat transfer coefficients, because they are condensers and coolers at the same time, but in this case, I think it would give realistic results. This in such a way calculated heat loss would take into account that a faster distillation with more energy would cause less loss compared to the heat input. And it would take into account the difference between ambient temperature and thumper temperature, what means, the loss is not much during heat up and grows more and more during the distillation.

What I would like to watch at a thumper simulation would be, how exactly parameters like charge abv or heat loss show their influence on the fill level. This would be a good movie for me and perhaps there is something to learn. But it's not on my to do list. I have never ran a thumper and in my area none uses thumpers. If someone wants to build a thumper in my area, probably the reason is watching the moonshiners TV series.

[haggy]

That is a good thought, I could calculate a heat loss due to the difference in thumper and ambient temperatures and subtract that from the heat brought in by the pot vapor. This would reduce the heat available to vaporize the thumper liquid and reduce the outgoing vapor rate. Adds another degree of complexity and variability to the model. The heat loss method used now is ok, but does not show variability with outside conditions like temperature, insulation, wind etc.

I could determine the amount of heat losses estimated this way which fit the run data. Have to study up on what affects convection and radiation heat transfer. The amount of exposed area is also a variable. I will give it a try.

The thumper temperature rises from about 85 C to 98+ C . For ambient temperature at about 20 C, the delta temperature is 65 C vs 78 C or about 20% higher at the end of the run. This could reduce the exit vapor rate some depending on how big the heat loss is vs the vapor heat input. The thumper vapor exit rate also decreases with time due to the pot vapor rate decreasing with time.

[haggy]

At the suggestion of Hügelwilli, I have updated the model to include a different calculation for heat losses from the thumper.

For heat loss by convection, use the equation:

Heat loss ( Btu/hr ) = hc * A * DelTemp

hc is the heat transfer coefficient, BTU/hr/F/ft^2
A is exposed surface area of the thumper, ft^2
DelTemp is Thumper Temperature minus Ambient Temperature, degrees F

We need to input the total volume of the thumper and a value for the hc, which will stay constant over the run. For DelTemp, we already calculate the thumper liquid temperature ( assume this is the metal temperature also ) and use 68 F (20C) for ambient temperature. We could input the ambient temperature also, later.

This calculated heat loss was subtracted from the heat input by the pot vapor to give a net heat to vaporize the thumper liquid. The thumper distillate amount was then calculated from this net heat input for each run increment.

Heat transfer by convection depends on the velocity of the surrounding air. The hc for free convection of air can range from 0.1 to 170 Btu/hr /ft^2/F. For an air velocity of 7 ft/sec, hc is about 4. We do not know the air flow, but this is the hc that seems to fit the data. A run by Chauncey had about 3000 Btu/hr heat loss ( 25% of the pot vapor heat input ) using this value of hc.

Radiation does not seem to be as significant. A correlation using copper emissivity and an 8 gal thumper gave about 50 Bu/hr heat loss.

I ran several cases on the model and found the hc which fit the gal produced data at certain vapor %ABV during the run. An hc of 4 to 4.5 BTU/hr/F/ft^2 fit most run gal and %ABV data. The time the sample was taken was more sensitive to the hc used than the gal / ABV data. Only tested a few runs so far. One run was not predicted as well as before, but other results were not very different from the previous way I estimated heat losses. The calculated heat loss as a percent of the total vapor heat in is given in the new results table. The % heat loss increases as the run progresses. The table has expanded ( too much info? ), it also includes the pot vapor rate and drip rate out the condenser and kW along the run which can change with the run time (later).

The Excel spreadsheet tables in the file below give the heat loss updated calculation for runs by Chauncey and Yonder. The results are fairly similar to the results from my previous post of January 2, 2021. The heat losses are about 25% to 50% of the vapor heat input. The Chauncey thumper was insulated, single hole thumper and gave lower % heat loss. Note: These two runs did not report the run time when the samples were taken. This needs to be taken into account also.

I need to check out more runs that include the time of the sample with this update to finally confirm it. The thumper % ABV vs time graphs of the two runs are given below and show smooth transitions over time. A small decrease in pot kW was needed to line up the data at the same run time as before.

With this update, several other variables can be tested with the model, as Hügelwilli suggested.

Chauncey3Yonder3.xlsx

(19.3 KiB) Downloaded 67 times

The Chauncey run. Pot 13 gal at 8% ABV -- Thumper 6 gal at 13% ABV
Data: Ran to 20% ABV. Made 3.5 gal at 43% cumulative ABV

The Yonder run. Pot 5.5 gal at 10% ABV -- Thumper 1 gal at 10% ABV
Data: Ran to 37% ABV. Made 1 gal at 63% cumulative ABV

[Kareltje]

What I would like to watch at a thumper simulation would be, how exactly parameters like charge abv or heat loss show their influence on the fill level. This would be a good movie for me and perhaps there is something to learn. But it's not on my to do list. I have never ran a thumper and in my area none uses thumpers. If someone wants to build a thumper in my area, probably the reason is watching the moonshiners TV series.

I do not know enough about thermodynamics to understand your remarks. So I removed them. No pun intended!!

Bold from me: How the charge abv and the heat loss influence the file leveL?? I do not understand the question. I doe not decide the fill level of my thumper by the charge abv and the heat loss.

The reason I made one was not the moonshiners tv series, but the need to distill a very heavy fruit pulp without scorching. (Although the mere idea was indeed from Moonshiners and HD.) More like a steam distilling than a thumper distilling. It worked very well and later I discovered the use as a second still or a extended still. Or a temporarily additional source.

I am not sure if I belong to your area, but I use a thumper on very various occassions and I am very happy with it. I must admit: we are very few in Europe!

[Hügelwilli]

Bold from me: How the charge abv and the heat loss influence the file leveL?? I do not understand the question. I doe not decide the fill level of my thumper by the charge abv and the heat loss.
I should have written "liquid level" or "filled level" probably. I meant I want to see how the liquid level changes during a distillation. Obviously something like a fan cooled thumper would fill more and more. But a deeper insight would be interesting. Also but not only to find out what circumstances lead to a over-flooding thumper.

I am not sure if I belong to your area, but I use a thumper on very various occassions and I am very happy with it. I must admit: we are very few in Europe! Yes. With "my area" I meant something like non-English-speaking middle Europe. You are the only one coming into my mind who distills with a thumper there.

[haggy]

Kareltje,

I have seen in the model that the charge ABV and the heat loss can affect the thumper fill level.

I will do some tests and report some results in a few days after I find the best hc ( heat transfer coefficient) that fits the data. Maybe then we can determine why and how they affect the thumper fillage.

[Kareltje]

Bold from me: How the charge abv and the heat loss influence the file leveL?? I do not understand the question. I doe not decide the fill level of my thumper by the charge abv and the heat loss.
I should have written "liquid level" or "filled level" probably. I meant I want to see how the liquid level changes during a distillation. Obviously something like a fan cooled thumper would fill more and more. But a deeper insight would be interesting. Also but not only to find out what circumstances lead to a over-flooding thumper.

I am not sure if I belong to your area, but I use a thumper on very various occassions and I am very happy with it. I must admit: we are very few in Europe! Yes. With "my area" I meant something like non-English-speaking middle Europe. You are the only one coming into my mind who distills with a thumper there.

Oh, I see what you mean.
And in that case I belong to your area.

Kareltje,

I have seen in the model that the charge ABV and the heat loss can affect the thumper fill level.

I will do some tests and report some results in a few days after I find the best hc ( heat transfer coefficient) that fits the data. Maybe then we can determine why and how they affect the thumper fillage.

Now I see what is meant by fill level I concur: ABV and heat losses affect the fill level. Also the starting temperature of the content of the thumper. (I always have to preheat the content of the boiler to prevent quenching of the flame by condensing combustion water from the gas!)
The lower the starting temperature and the greater the heatlosses, the longer it takes to reach boiling temperature and so evaporation in the thumper. And the more high % ABV distillate will remain in the thumper.

I made a model to find out what happens in boiler and thumper and although I did not seriously try to validate the model, the insights I got from playing with it were very helpful.

[Kareltje]

I can not make a film, but I can make the runs visual. All with the same boilerload: 10 L of 10 %. In this picture thumperloads of 5 L and 10 %.
Than: in both blocks 30 % loss on top and 50 % loss on bottom. Upper block starts at 20 dg C, lower block at 60 dg C in thumper.

Left of the two graphs are volumes, right ones in %ABV.
I suppose the texts can be readed by you, although the are in Dutch. I can of course translate them if you like.

Working on a block of starting %ABV in thumper of 20 %.

[Kareltje]

Same setup, but this time with a starting %ABV in the thumper of 20.

But of course this are modelruns, not real ones. Although I find the very helpfull, they are not exact replicas, of course.

The model can warn me against dry boiling of the boiler and overflowing of the thumper.

[haggy]

For the heat loss I probably would use the heat exchanger formulas and look what heat transfer coefficient gets the calculation to realistic results. I tried a long time to calculate liebig type condensers, it is impossible because our condensers doesn't have fixed heat transfer coefficients, because they are condensers and coolers at the same time, but in this case, I think it would give realistic results. This in such a way calculated heat loss would take into account that a faster distillation with more energy would cause less loss compared to the heat input. And it would take into account the difference between ambient temperature and thumper temperature, what means, the loss is not much during heat up and grows more and more during the distillation.

What I would like to watch at a thumper simulation would be, how exactly parameters like charge abv or heat loss show their influence on the fill level. This would be a good movie for me and perhaps there is something to learn. But it's not on my to do list. I have never ran a thumper and in my area none uses thumpers. If someone wants to build a thumper in my area, probably the reason is watching the moonshiners TV series.

OK, I have gone through most all the posted runs that I have studied and have found a heat transfer coefficient (hc) to fit each run data. The hc does vary with different thumpers. The hc results varied from 1.3 to 6.6 Btu/hr/ft^2/F . Attached is an Excel file "Finding hc" giving the results. I do see some correlation of the hc with the Total Gal of the thumper, and maybe with a very low fill level.

I have added the hc and heat loss calculation { hc* A * (T thumper-T ambient ) } to my pot/thumper model. The A is a calculated surface area of a cylindrical thumper. Model results showed that most (93%) of the runs required a 35%-40% heat loss in order to fit the gal and ABV product run data. That is why my original model using a 40% factor for heat losses fit the data well.

As you requested, running the model at an increase in energy (pot kW) would send more vapor and heat to the thumper and make the % heat losses less. The model does show this now. An example is given in the Excel file "Faster Distillation" below. There is some change in the ABV of the end product. The thumper ABV curves of the two cases are given below. The model results also show that the % heat losses increase during the run due to a larger temperature difference between the thumper and ambient.

The charge ABV can affect the fill level as Kareltje has shown. I will post some results like this in my next post.

So, we have a more realistic updated pot/thumper model now. The hc used may not be totally accurate but does fit most run data examined and is close.

The hc determined should really be called an overall heat transfer coefficient U. The values found are consistent with typical steam radiator coefficients.

Here is the Finding hc results file:

Finding hc.xlsx

(18.49 KiB) Downloaded 86 times

Here is the spreadsheet information comparing runs at different speeds of distillation.
The pot charge was 10 gal at 12% ABV, the thumper charge was 5 gal at 12% ABV in an 8 gal thumper.
The pot power was raised from 2.1 kW to 3.2 kW and heat losses decreased from an average of 52% to 35%.
The faster run had less of a chance of overfilling the thumper. The slow run could overflow if not stopped at 20% ABV.

FasterDistillation.xlsx

(18.1 KiB) Downloaded 75 times

Here are the slower 2.1 kW thumper ABV graphs:

Here are the faster 3.2 kW thumper ABV graphs:

[haggy]

One more post here and then I may be done for a while. Would like to see more interest in this undertaking and more people downloading and reviewing all the information that I have put into the attached Excel files. I will certainly respond to questions or suggestions, but the model is good and realistic enough now.

So, how do changes in the charge ABV of the pot and thumper affect the fillage of the pot and thumper? How do they affect the thumper distillate amount and ABV? Let's see what the pot/thumper model says.

I suspect experienced distillers know this stuff already, so this is for the novice.

A key concept is this: As the pot or thumper liquid gets richer in alcohol ( %ABV increases) , the vapor boiled off is richer in alcohol. The latent heat of evaporation decreases as the vapor gets richer in alcohol. Therefore, more gallons can be boiled off for a given heat input as the liquid charge %ABV gets richer. So, the starting pot ABV will affect the amount boiled off and therefore affect the fillage of the pot. The thumper fillage is affected by both the pot and thumper initial ABV and gallons.

I ran the model for four cases with changes in the pot and thumper %ABV charge. The cases had different power inputs to the pot and all used a 10 total gal thumper. No extra reflux.

Case 1. Pot 10 gal at 10% ABV -------Thumper 5 gal at 10% ABV Strip Run
Case 2. Pot 10 gal at 40% ABV -------Thumper 5 gal at 10% ABV Spirit Run
Case 3. Pot 10 gal at 10% ABV -------Thumper 5 gal at 40% ABV Spirit Run
Case 4. Pot 10 gal at 40% ABV -------Thumper 5 gal at 40% ABV Spirit Run

We will look at the effect of these changes only on fill levels and % ABV made. There are many other considerations like where do heads and congeners end up may actually dictate which spirit run case is best. Many forum posts cover this topic.

The spreadsheet results for all the runs are in the Excel file below.

Case 1 is a standard strip run. Low wash /mash charge 10 %ABV, no problems in either fillage. Could strip down to 5-10% ABV vapor from thumper and make about 3.7 gallons low wines at 43% ABV. Can make several of these runs and dilute with water to make the 40% ABV low wines charge for the next spirit runs. Most distillers run a higher power kW ( 4.5 kW for this 2 hour run ) and get about the same ABV results quicker, about 0.2 gal more each run and a little less thumper fillup. Lot of alcohol to reuse.


Case 2 with the highest (safe) 40% ABV low wines at10 gal in the pot. Watch out for the pot here, because more vapor is generated early on ( compare the ml/min of the pot with Case 1 ) and the pot fillage decreases faster. In fact the pot can empty before the thumper gets to 10% vapor ABV. Would need to stop and refill pot at about 160 min if you want to get more alcohol out. This is a usually a slower spirit run, and you would stop the run when tails start (at about 45-50%ABV?). But, the thumper has filled to almost the maximum level, so the 2.4 kW is a minimum setting. The pot has about 1.5 gal in it, so ok there. About 5 gal of 100 proof after dilution is made stopping at 50% ABV. Should run this a little faster ( to 2.9 kW ) to get more product out and less thumper fillup. Recycle and rerun the remains.

Case 3 has 5 gal of 40% low wines in the thumper and 10% mash in the pot. Less total alcohol than Case 2 and another spirit run. There are no problems with depleting the pot or filling the thumper in this case. Can run this faster than Case 2, about 3.3 kW used here. Might stop at about 50% ABV distillate out and make about 5 gal of 100 proof after dilution. Then recycle and rerun the remains.

Case 4 has 40% low wines in both the pot and thumper. A power of 3.2 kW was used. This run has the most initial alcohol charged. Stopping the run at 50-55% ABV distillate comes very close to emptying the pot but ok on the thumper fill level. A lower kW would slow down the run too much and the pot could empty. Might be good to add to the pot during the run or run faster with less heat losses. Can make 8 gal of 100 proof after dilution here. Going further another 25 minutes without refilling the pot could empty the pot. The thumper fill level is ok most of the way, 8 gal after initial fill up then about steady or decreasing since the thumper distillate rate is a little higher than the pot vapor rate. Recycle and rerun the remains.

The thumper distillate flow ml/min can be measured during a run. This is proportional to the power input to the pot. The model shows that increasing the power to the pot has the net effect of decreasing the thumper fillage. Knowing the minimum ml/min rate from the model results can help find a power setting to keep the thumper from over filling.

So, those are some of the things to expect and watch out for in each case.

I ran another case. What if we ran a 12 gal charge in both the pot and thumper with 40% ABV low wines in each 15 gal keg? This is like the posted run by Beerswimmer. He ran 13 and 10 gal charges at 40% in 15 gal kegs. The 12 gal charge in the thumper might be too much. Any problems here? Go to the spreadsheet Case 5 and find out.


So, I am done for a while unless we have some comments/ questions. It is time to stop and watch my saved TV episodes of "The Curse of Oak Island" and "Moonshiners".

ABVeffects.xlsx

(33.33 KiB) Downloaded 92 times

[Hügelwilli]

Interesting results.

Case 2 seems to overfill the thumper. The run with 40%abv in the pot and only 10%abv in the thumper. Looks like to have low wines in the boiler and wash or backset or dunder in the thumper is not a good idea from this standpoint.
Do it some distillers here this way? Or is it a general rule that the thumper charge should not be weaker than the pot charge?

Case 4 with 40%abv in both pot and thumper emptied the pot before the thumper had distilled off all its alcohol.
Is this verified by practice, that you cannot get all the alcohol out, if you have high abv in both pot and thumper?

Those are real questions. I don't know much about thumpers.

[Kareltje]

One more post here and then I may be done for a while. Would like to see more interest in this undertaking and more people downloading and reviewing all the information that I have put into the attached Excel files. I will certainly respond to questions or suggestions, but the model is good and realistic enough now.


So, I am done for a while unless we have some comments/ questions. It is time to stop and watch my saved TV episodes of "The Curse of Oak Island" and "Moonshiners".


ABVeffects.xlsx

I understand your dismay. It is no fun trying to invent something and getting hardly any respons.
I have seen some discussion between you and Hügelwilli and was flabbergasted about the details of heat transfer and other technical aspects. My model was just based on evaporation/condensing heats and mass and volume transfers. And just guessing heat losses and tray numbers.

I am not really interested in developing a model for this forum, as I do not think my model has the right characteristics to be used as a tool in this forum (or in hobbybrennen.ch, for that matter). I use my model, thanks to your input, for my own good and there it suffices.

The use of the thumper, as discussed on this forum, seems to be very limited by the rules of thumb, as if this thumb is sacred.

But for example a member like Truckinbutch does not obey this rule and some others too use thumpers that are smaller than 1/3 or larger than 1/2 of the boiler size. Whith good results! Even Tennessee thumpers of less than 1/10 of kettle size are used with success.

Than you have got another discussion partner in Hügelwilli. In Europe thumpers are hardly used, at least not in western Europe. I am not sure about eastern Europe: they seem to use thumpers there, or maybe slobberboxes, I am not sure.
But Hügelwilli seems to work among the same lines as you do: heat transfer coefficients and all kinds of other physical and chemical constants. So I guess you could have a worthy discussion partner in him.

And I would really like to lurch!

If I try and run one or more of your cases, I will report. Sincerely yours.

[The Baker]

Hi,
I can understand lack of response can be daunting.
So I will repeat my reasons...
I have no idea what you are talking about!
But that does not mean I don't respect your work, quite the reverse, I am glad
that people with a lot more knowledge and probably ability than myself are working on the
technical side of our hobby. It can only be to the benefit of all of us.
Thanks and keep up the good work.

Geoff

[haggy]

Interesting results.

Case 2 seems to overfill the thumper. The run with 40%abv in the pot and only 10%abv in the thumper. Looks like to have low wines in the boiler and wash or backset or dunder in the thumper is not a good idea from this standpoint.

I picked a set of cases to illustrate different ABV options. The 40% ABV used is a maximum value where we can see the effects.

On purpose, for Case 2 I used a low 2.4 kw for the pot where the thumper was close to fill. Notice the high 53% heat losses. The run is better at 3.5 kw power where the heat losses are 36% and maximum fill is 8.4 gal.

Yes, the boiler can empty before the thumper. You have to know when to stop the run before that or refill the boiler during the run. Usually you reach the tails and stop the run before the boiler empties.

Do it some distillers here this way? Or is it a general rule that the thumper charge should not be weaker than the pot charge?

I do not know if that is a general rule. Many posts have equal ABV in both vessels, and there are many posts where the thumper only has water in it just to cover the inlet pipe. I suppose you can run less in the thumper if you know when to stop the run, before the pot empties or when you reach the tails. The 40% pot ABV is the worst case.


Case 4 with 40%abv in both pot and thumper emptied the pot before the thumper had distilled off all its alcohol.
Is this verified by practice, that you cannot get all the alcohol out, if you have high abv in both pot and thumper?

Yes, the pot can empty before all the alcohol is out of the thumper. Some people will refill the pot with wash or water during the run to make sure it does not empty. If they do not refill, they stop the run and recycle the contents of the thumper.

Here is a post by Beerswimmer that illustrates some of your questions.

Re: Thumper talk
Post by Beerswimmer » Mon Oct 14, 2019 1:08 am
Was doing the spirit of 3 gens of the BWKO, and was going well about 3/4 through it and.......smoke started coming out the PC :esurprised:

I learned a lesson on my new rig, the boiler had better NOT run out of liquid! I learned that I basically did it backwards, I loaded the boiler with 40% abv low wines and the thumper had way lower % than that. So the boiler ran dry while the thumper still had plenty of juice left in it. I'm about a gallon and a half-2 gallons short of expected :|

Next time I may lower the % of the boiler charge to 30% for the spirit run to be sure that I don't scorch again.


So, I am not the expert in this area. There are many in HD who are by their distilling experience. All I know is what I have read in many, many posts and what the model and science tells me. Hope this helps you understand more about the thumper operation. You have to search the Thumper forums in HD for more information. I can get you the source of some of the best posts.

[Chauncey]

Pretty cool sheet ima have to play around with it. Thanks

[Kareltje]

The problem of boiling the boiler completely dry I solved in real life.
I have a thermometer between boiler and thumper and when that shows a 100 dg C for some time, I know all alcohol is transferred. Then I can either lower the heat for the boiler until I just keep up the turbulence in the thumper and start heating the thumper. Or I can disconnect them completely, close the entrance pipe to the thumper and go on distilling with the thumper as a boiler.
Can do the same when I fear overloading the thumper.

[haggy]

Pretty cool sheet ima have to play around with it. Thanks

Chauncey,

Thanks for replying and thanks for posting good complete run data.

You know that I have used your posts, they have helped a lot. I have updated the model and used it with your run data of July 1, 2020 in the "Maximum thumper charge ..." forum. You got a lot of good responses there, your nice setup is similar to several others.

Attached is an updated Excel file of your run and some nice graphs of the pot and thumper. I did not know the final run time of the run, maybe it was shorter for the 3 kw power I guessed. So I include in the file a faster case at more pot power, 5 kw to show the benefits of a faster strip. Then I also added another run to the file using 13 %ABV in both and estimated what might be the maximum charge you can run in a strip. Easy to run a lot of cases on the computer.

Strip runs start with lower ABV in pot and thumper so they are a little more forgiving in emptying the pot. Also, you are probably running them fast now. Spirit runs can be a little unforgiving.

How are you running your strips and spirits now? If you could give me some current info via this forum or in a PM, I would appreciate it. I would be glad to run some new cases for you also. Model is not perfect but it is close enough.

Best Regards

Chauncey22.xlsx

(23.78 KiB) Downloaded 107 times

Here are graphs of the last case,13 gal pot and 11 gal thumper both at 13% ABV.

[haggy]

Sorry, but I just realized not everyone has Excel software on their computer to download the Excel files I have been posting.

So maybe I can get you the info another way.
Let's look again at the last Excel file "Chauncsey22.xlsx". Next is the spreadsheet of the first run in the file.
You see on top the input information, then in the columns are all the key variables during the run. Options on top right. Pot info is on the left, thumper on the right. Time is on the left and the pot vapor rate and thumper product ( distillate ) rate are given on the left and right. Estimated pot and thumper temperatures are in the middle. The new thing about thumper heat losses is on the left, and this can vary with kW power input. I have blocked out the run conditions which matched the run data.

Then the second case where the power is increased from 3 to 5 kW.

The last case has 13% ABV in both pot and thumper with a larger charge in the thumper.

Hopefully, you can see the goodness of the model in being able to understand what is happening and comparing different run options. The model gives the estimated % ABV out the thumper, but in no way implies where cuts should be made. But it can give you an idea of when the pot might empty or the thumper might overfill and what the %ABV might be when you stop the run. Plus a lot more info like the ml/min out the thumper predicted which can be compared with actual run data.

So, maybe this will help everyone get a better understanding of what is happening of the process.
Comments and your run data for comparison are welcome.

[Chauncey]

Yea the numbers definitely seem on point i just havent had time to mess with it yet. Definitely a useful tool

[haggy]

I am back. Have been watching this thread and see a lot of lookers but no comments . There must be some interest judging by the amount of lookers. The pot/thumper calculations have been updated in several areas since my last post, and are as real as I can get them.

YES, I know that run experience on a still's operation is the best option, and you do not need a calculator if you have much experience on your still. So maybe this is more of a teaching tool for what is going on in the pot and thumper.

YES, all still's are some what different in operation, that is a reality. But, once you have zeroed in on calculating your operation using the heat loss option in the model ( the heat transfer coefficient U ) , the amount of initial reflux option and the kW power used, you can get a good representation of your run operation. Then you can test different operating conditions and see if they offer any advantages. Things to improve might be more output, higher ABV, shorter run times, longer time at high ABV, thumper and pot fill levels, etc.

YES, as I have said before, the cuts can not be determined with the model, you must do that during the actual run. But the model gives you %ABV coming off at all times, so at least you know when to look for the tails starting. The model is good at detecting when the big drop in ABV starts so it can warn you when this may happen.

I do not know when or if anyone will have access to this model, or even how many people will want to use it. But, several have said they would like to have it. Someone else has to get onboard here and promote this Excel model and convert it to the HD calculator format for all to use. Maybe the only thing that will happen is that some one PM's me their operating conditions and some run data, and then I feedback to them the model verification. We could then look at various operating condition changes that we agree on.

Here is one good thing that keeps me going. A Site Administrator sent me a PM and said he likes this kind of thing, thanks and keep it up. So I will.


An Example Spirit Run

So, here is another example of the current model simulating existing run data. I will use data from a post by wpkluck on March 10,2021 in the thread "Yield - UJSSM Spirit Run Update".

This was a spirit run with a 5 gal pot charge at 40% ABV and a 1 gal thumper filled just above the end of the inlet spout with about 0.2 gal of 35%ABV. Takeoff gal amounts and ABV at several times during the run were given in the post. The run was stopped after about 40% ABV in the distillate, and at the end of the run about 0.75 gal remained in the thumper. Heads, hearts and tails were estimated by taste and ABV, but the heads and hearts might have been smeared a little. A 2 hour pot heat up time to the start of boiling is estimated. The first product gal was made at about 165 min after the start of the pot boiling. The other data times are referenced back to this time.

The following is my Excel model spreadsheet for the run. The specified cuts from the data ABV are in different colors. The U value of 3.9 and 700 watts power were fit to the data and no initial reflux was needed. Initial ABV is very high due to the starting pot conditions and a thumper at high ABV and low fill. Many run variables as they change with time are detailed in the spreadsheet.

Here are graphs of the key thumper run variables calculated by the model. First is the ABV of the thumper liquid and distillate as they change during the run. The distillate flow rate initially is about 28-29 ml/min. The distillate data ABV results are given in the filled squares. You can see how high the liquid ABV gets during the run and the rate at which both ABV's drop. At about 400 minutes the run is stopped at about 40% distillate ABV.

Next the calculated thumper volume and distillate gal made are plotted. The filled triangles are the distillate gal made data. The thumper fill calculated by the model at the end of the run is about 0.8 gallons and might get to about 0.9 gal maximum during the run.

Having this base case, you can explore different run conditions with the model to see if better results ( like more gal made or higher ABV made or a longer time at a good ABV or the thumper max fill ) can be obtained. Runs at higher or lower pot kW ( giving higher or lower ml/min output ) can be examined. Maybe a lower pot ABV does not fill the thumper as much. What if wash or water were used in the thumper? These are ideas for another post, maybe.

[Berserk]

I think it would be very interesting to use the model to see what the differences would be when running with different volumes and ABVs in the thumper, like wash/water/low wines.

Also, I wouldn't mind taking a look at the model and see if I can convert it to JavaScript.

[haggy]

Thanks for the reply and offer.
I have studied many different thumper combinations and they do make a difference. One or two real run posts that I examined looked at different ABV's but not volume. I will post some results for a few different cases next week.

I can send you the Excel version of the model. I have to clean it up first, many extraneous comments and unused info and blind paths. Will take a few days.

We have sent PM's to each other previously, and from your posts I know that you are trustworthy.

[Kareltje]

I have no skills in transforming the model in other computer languages, but I am glad to hear from you again and I will certainly try out the model, once it is available.
Good work haggy!

[Mr Sippy]

Excellent campaign, haggy!
I am delighted there is interest in perpetuating the thumper model. Much thanks as well to Beserk for the generous gift of time. A prime example of Forum spirit at it's best and I'm sure others would echo the same sentiment. Looking forward to a 'calculator'.

[haggy]

I think it would be very interesting to use the model to see what the differences would be when running with different volumes and ABVs in the thumper, like wash/water/low wines.

My fellow Swede Berserk ( I am half Swedish ) asked to see what the pot/thumper model predicted for different thumper charges, the gals and %ABV. So I ran a matrix of conditions with the model and report here the results.

A pot charge of 10 gal and 13% ABV wash was common for all thumper conditions, like a strip run. Maybe next I can be a spirit run with much a higher pot charge. Then I ran a matrix of thumper conditions: three ABV amounts, 0% water, 13% wash and 40% low wines. I ran four different thumper volumes, 0.6, 2, 5, and 8 gal so we then have four T/P ratios, 0.06, 0.2, 0.5 and 0.8. The thumper gal to pot gal ratio ( T/P ) is an important variable. So we have a 3 x 4 matrix of run conditions to evaluate and see what the differences are.

All run conditions were at 2 kW and with some reflux ( Reflux Option set to 1 ). If a thumper has more reflux, the ABV results would be higher. A 41% heat loss was used and held constant. This is about the average value seen in run data. The results can be adjusted a little for stills with higher or lower heat loss. It was seen that a different kW did not change most results, only affected the takeoff rate and run time. If the pot gal is changed from 10 gal to another fill, the results were similar if the T/P ratio is kept the same.

Here is the matrix of results, a lot of information. I will graph some of the main results so we can see how things change.

This first graph is for the effects of the variables on the Initial Distillate ABV. The x axis is the thumper charge ABV and the curves are for different T/P ratios. The data points are for water, 13% wash, and 40% low wines. The Initial ABV naturally increases as thumper charge increases but acts differently at thumper charges below 13%. The curves bend down except for the low 0.06 T/P ratio. Using water in the thumper (far left on the curves) has a big effect on the ABV for different T/P ratios. If water is used, only fill the thumper just to cover the spout.

The next graph is for the effects of the variables on the (hearts?) Cumulative ABV at 60%ABV out the spout. The Cumulative ABV increase smoothly as thumper charge increases. The lowest T/P is flatter than the other curves, the Cumulative ABV does not change much. Again, if water is used, only fill the thumper just to cover the spout.

The next graph is for the effects of the variables on the Cumulative ABV at 20%ABV out the spout. This could be the low wines made. The Cumulative ABV increases as thumper charge increases. The lowest T/P is flatter than the other curves, the Cumulative ABV does not change much. Above T/P of 0.2 , the values increase significantly with thumper charge. The values crossover at 13% thumper charge. Below 13%, high T/P is worse; above 13%, high T/P is better. Again, if water is used, only fill the thumper just to cover the spout.

Next is the Distillate Gal made down to 60% ABV out the spout. A linear increase for all T/P ratios with the slope increasing at higher T/P ratio, more gal of alcohol in the thumper to start.

Next is the Distillate Gal made down to to 20% ABV out the spout. A nice neat linear increase for all T/P ratios. With very low T/P ( not much alcohol to start with ) the amount made is not increased much over the range of start %ABV.

Not much is really new here. More alcohol to start with gives more alcohol that you end up with and a longer time to get to the end of the run. Water in the thumper is a little bit different. BUT, these matrix results put numbers on the cases and can be used as a reference. I made curves for down to 60% and 20% out the spout to see what ABV and gal are made close to the end of hearts and tails if you use this data for a one and done run.

The ml/min and time of to the end of the run are given in the matrix table. End Time for the run is a function of the kW used and total amount of alcohol in both the pot and thumper. The % ABV and gal results would be about the same if a different kW was used, just the time of each step and the end time would change. You could just ratio the new kW used with 2 kW to get a new time. A 25% increase in kW to would get a end time shorter by about 20% ( 2/2.5 ) if heat losses are the same.

The thumper maximum fill gal along the run time is given in the matrix. The range is about 20% more than the starting fill gal for highest T/P and about 80% more at the lowest T/P. A higher ABV in the thumper gives a slightly low max fill.

Also if we started at a different pot gal than the 10 gal here, you could ratio the gal results for a different pot gal to the 10 gal. So for a 5 gal pot at 13% ABV, the gal results would be half as much at each T/P ratio.

And the % ABV results should be the same as given in the matrix for different starting pot gal and kW as long as the T/P is the same and heat losses are the same and the pot is at 13% ABV to start. The matrix results might be used for all 13% pot ABV runs. I will test this out.

Next, we should probably see the results for a spirit run starting with 40% low wines in the pot instead of a 13% wash. More work.

So, if you get all of this you should be working with me in developing and confirming this pot/thumper model.

[haggy]

I have one more graph for the above post. This is how the ABV out the spout changes at different thumper to pot (T/P ) ratios during the run time for 13% ABV in both the pot and thumper. The Pot charge was 10 gal with a 2 kW power input.

[Kareltje]

NICE!
I will try the same runs with my own model.

About the little effect of different kW's on various effects: I tried once with a pot still different amounts of kW and so different amounts of cooling. It was expected that very slow cooling (as I have with a spiral in the air) needed more energy as it has a longer period of heating. To my surprise there was no difference: very slow or very fast running, i.e. with high and low firing, so with high (water) and low (air) cooling, were only different in time of running and in debit. Amount produced vs energy needed were the same for all three runs.

You now show that what goes for a pot still also goes for a boiler plus thumper.

[haggy]

Yeah, good point, but it is more complicated with the pot plus thumper. These runs for simplification had the same heat losses for all cases during each run. That is one option in my model. Results are about the same. And that is when there would be little effect of different kW's. Maybe I should have been more clear on that.

The other more complicated option in the model is to input a heat transfer coefficient and the area of heat transfer and calculate the temperature difference and you then can calculate heat losses ( U * A * (T thumper-T air ). When you change kW for this option, the heat losses will vary with the kW and during the run time. And then what goes for a pot still is not exactly the same for the pot plus thumper. The kW will have an effect.