First test of my thumpermodel

Distillation methods and improvements.

Moderator: Site Moderator

First test of my thumpermodel

Postby Kareltje » Mon Sep 04, 2017 5:23 am

Earlier this year I bought a secondhand still. It was not the promised 15 L but only 10 L, but it had some parts I did not yet have: a helmet and a watercooled line-condenser, so I was happy with it. I soon made some adaptions to give me more freedom and later I made a one-hole-thumperlid on it.
Happy with my new toy I immediately used it to still a very havy wallnutpersscakeferment without any scorching.
Triggered by the question: what does a thumper actually do, I tried the model from the homepage, but at my computer it did not work, so I decided to build my own. By now I have it running and I think I have corrected most mistakes and failures, so it was time to test it against reality.
The job at hand was cleaning some bad tasting neutral by running it through water, as Pikey told Pintoshine said would work. Two runs: one with cold water and one with hot water in the thumper.

Here I must add, that due to the round bottoms of my boilers I always have to fill the boiler with preheated wash. If I put the boiler cold on the stove, the water from the burning gas condenses, flows down on the boiler and quenches the flame. This stops when the contents are warmed to about 40 dgr C, so I heat my wash always to at least that tempreature.

The model predicted a measurable difference between both runs, so I put it to the test. I filled the boiler with 2 ltr 83 %ABV plus 6 ltr water and the thumper with 6 ltr clean water.
I measured and calculated the normal parameters, like volume and %ABV of product per sample and in total. I also took readings of the temperatures coming from the boiler and coming from the thumper. And I maesured volumes at the start and at he end in both boiler and thumper. I made the following graphs of the results: the left one are the volumes in ml, the right are the %ABV, the x-axis is time since start of the run through the thumper, judgied by the sound of it.
Koud-warm2.jpg

The difference is clear, I guess.
When the thumper starts cold, the incoming vapour has to heat up the contents and does this by condensing, thus filling the thumper with both warmth and alcohol. If than the thumper is hot enough, it starts to produce vapour from an alcoholic content.
When the thumper starts hot, the incoming vapour needs only little time to heat up the contents enough to start vapourizing and so the alcohol content is still much lower.
As boiler and thumper are right beside each other and the ID of the connecting pipe is only 13 mm I estimate an almost immidiate transfer from boiler to thumper. Furthermore I assumed the transfer is linear. The first assumption might be true, the second probably not entirely.
Both boiler and thumper were left with 0 %ABV, but there was a unexplained loss: I started both runs with 1.660 ml alcohol and a total of 14 ltr, but ended with 1.440 ml alcohol and 13 litres total volume!

I tried the parameters in my model to get the best compliance with my observations, and it came to this:
Model-hoofd.jpg


Starting with the known data: content in kettles and energy of the stove and adjusting the supposed number of trays and the loss in the thumper.
The results were these graphs:
Koud-warm-model2.jpg


In these results we see, in contradiction with the real runs, no loss af alcohol and volume.
There is a tremendous difference in time, though: 40 steps of 2 minutes is only 80 minutes as opposed to more than 4 hours in real life.
This can be compensated if we assume an efficiency of only 33 % of the gasheating. If that is the case, we need steps that are three times as long, so we get 40 steps of 6 minutes. Loo and behold: that is about 4 hours too!
I think an efficiency of only 33 % of a gasfired stove is not really uncommon: in previous runs of my potstill I frequently got these kind of figures.
So the first test is passed, if you ask me.

Any comments are welcome!
User avatar
Kareltje
Distiller
 
Posts: 1591
Joined: Tue Feb 09, 2016 4:29 pm

Return to Research and Theory



Who is online

Users browsing this forum: No registered users and 1 guest