4" packing options

[Anthoney]

Not quite. It does make the scrubbies more efficient, once you sieve out all the little bits of chaff from your cut up scrubbies, but not in the same realm as SPP. Poor man's compromise but still good.

[stilly_bugger]

Hi guys,

going to put a 4" packed extension on top of a bubbler to see if i can produce a reasonable vodka with 1 run through what i guess you'd call a hybrid bubbler 8-)

Rather than a packed section above or below the plates, you could try packing the plates. From what I've read, this improves the efficiency of each plate. If you give it a crack, it would be good to hear how it works in comparison to a separate packed section.

:handgestures-thumbupleft:

[Anthoney]

That was interesting. It is effectively packing above a plate, albiet done repeatedly. This seems to be the same as the packing above a bubble ball some people have had good results with and I am trying to build if I ever get my flanges sorted :evil:

To answer the question you posed in that thread; the reason they don't just build a packed column is that the bottom of a packed column is not very efficient because of the low ABV of the vapour. It only kicks in properly further up. The plate rectifies this by providing higher ABV in a short hight. Hence the bubble ball packed column hybrid. using multiple units of this stacked on top of each other should improve efficiency vs hight even further. Don't know where the point of diminishing returns is and too preoccupied with other problems to sit and work it out by looking at the efficiency of packed columns at various ABV and comparing that with what you have after each plate/packing section. I am pleased to learn that you can just fill the ball with packing though. Never thought of that myself. Blindingly obvious in hindsight. Just need a screen on the downcomer to keep the packing on top. Stop you seeing what is going on with the caps though so a sight glass becomes moot.

Carrying the thought of moot design features further, having the ball filled with SPP removes the need for slits or mesh on the caps. You increase the hight of the puke gap and the SPP provides the necessary. Yes?

You could fill the bottom of the ball too. Should give a slight boost to the ABV reaching the plate but not a very efficient use of spp if it is expensive to you or in short supply. Otherwise it makes good use of the space that is there anyway to get the most from the height you have. Not a lot of space in my particular ball so I probably wont bother or just use scrubbies. Most of the space is above the plate which I may well pack after watching it work without packing to be sure I am happy with the bubble caps and just to see it go at least once. So I would have to use mesh or slits at least on this one. Maybe once I am confident I know what I think I do and can build what I think I can I will just move to big puke gaps and SPP. Or keep a mesh to hold the SPP at bay. That way the surface area is more than just the gap between the upcomer and the cap which would be the case with no mesh and SPP filling beneath the cap. Not sure.

As the thoughts continue to trickle in on this while I write I am beginning to wonder if stacked balls full of packing could be the ultimate short/fast neutral column. Smaller balls as you go up I think. All filled with SPP. Only need a downcomer screen on the bottom ball to keep the SPP out of the boiler. Just a short bit of column at the top for the moving coil VM head. Hmmm

Ah no wait; aren't you just creating flooding zones then? If the spp is submerged in a bed of liquid? Or is that ok on a plate? Head hurts now :think: Someone else chime in and help.

[Anthoney]

Thinking on not sure if just having two or however many balls first to get the packing up to a good efficiency then enough column to reach neutral wouldn't be a lot easier to get right. Dunno now.

[stilly_bugger]

...aren't you just creating flooding zones then? If the spp is submerged in a bed of liquid? Or is that ok on a plate?

From what I can gather it's okay on a plated column so long as you don't get excessive entrainment (flooding passing from a plate to the one above, eventually flooding the column).

In plated columns, the liquid phase is the dominant phase (in packed columns, the vapour phase is the dominant phase). The improvement to a plated column with the addition of packing is due to the increase in surface area (or sites) on which upcoming vapour can interact with the liquid on each plate.

So if column efficiency (largest increase in % abv over shortest column height) is the name of the game (which it isn't if you're producing flavoured spirit), then then most efficient plated column is going to be the one that has the greatest amount of liquid and surface area per cm of column height while remaining within the column's band of 'normal operation'.



For all plated columns (sieve/perf plate, bubble cap, valve) this means having the most liquid per plate (high downcomers) and the most surface area (packed plates) that doesn't induce excessive entrainment.

At least, that's what I gather. As always, happy to be corrected.

[Anthoney]

That sounds perfect but without analysing the graph, I will, how do you determine where the boundary of excessive entrainment arises?

[Anthoney]

I see with a bubble plate the viable area would be most if not all of the area below the excessive entrainment and downcomer flooding red lines.

No problems with weeping or dumping.

might be a minimum vapour to liquid ratio to get the caps to bubble against the backpressure of the liquid and atmosphere.

But how do you calculate the units on the graph so you can work out where the limits are?

Is it just trial and error?

Seems like there should be an equation?

In fact the graph was surely generated by an equation? it shows curves that follow the type of shape you see in mathematical relationships. Not a linear or straight line graph. But an interactive distribution between the vapour and the liquid. Fluid dynamics I guess.

Hmm sounds hard.

[stilly_bugger]

But how do you calculate the units on the graph so you can work out where the limits are?

Trial and error (empirical testing) would be the simplest way to work out the limits of a particular column. It would be very difficult to calculate the actual limits of a column on paper, and not much help if that calculation doesn't match how the column actually runs.

The graph itself just visually represents the operating limits of a sieve tray. It's deduced from the different variables at play in a sieve tray column — column structure and components, principles of vapour/liquid interaction, etc. There's probably a formal equation stating the relationship between these different variables, but it's not listed in the article. If you got your hands on a copy of the book Distillation Operation by Kister, it will probably go into more detail. Try your local library ;-) .

As for the boundary of 'excessive entrainment' in a sieve tray, the article author only offers an indication of excessive entrainment: unacceptable separation. For us, that would be the unacceptable separation of fores, heads, hearts and tails as tested by a functioning olfaction. A bit more substantially, the author does state that you're well beyond the boundary of excessive entrainment when the inter-tray space is full of spray, a large portion of which is hitting the bottom of the tray above. You'd need sightglasses to check this.

[Anthoney]

So vapour rate is effected by power in and liquid rate is effected by reflux so keep power low to start with a new trial design and turn it up in stages starting with full reflux to find out how much power you can use to drive it. If I pack a ball below a column then there could potentially be no space at all. Why not just pack it solid all the way up and make the most use of all the space? Seems like empty space between trays is mostly a break gap to stop the kind of runaway interaction you mentioned and allow more power to be used. Think the real work is done in the liquid layer and the packing. If a plate can be layered with packing then why not go all the way up and make best use of all the space? I have a purpose built, sensitive, differential pressure sensor to cut the heat if the column starts to flood or puke. That might be enough to go with a design that has no visual feedback on the internal operation. What do you think Stilly?

In practice the physical construction of my ball is such that it would be difficult to pack more than the lower half of the ball above the plate. Perhaps some scrubbies stuffed up below the plate. I have a reflux distribution tray that might make it a bit tricky to get SPP to pack the upper half after it was assembled. Don't want to remove it even with the packing as I want the reflux to be moved to the opposite side as the downcomer as much as is possible with the packing effecting the flow.

hmmm I could fill the upper half, which is the larger half, upside down. Then assemble and invert. That would give me a nearly full ball I think could be good to start experimenting with. Just got to finish my build first :angry-banghead:

Writing as I think like this just goes to show there is a process to most thoughts. The temptation is to edit it to make myself look quicker on the uptake than I am of course. But I don't mind everyone realising I'm not as clever as a carefully edited post could make me look. Not that clever, just more than I really am. I think sharing the process itself has some value.

[stilly_bugger]

I have a purpose built, sensitive, differential pressure sensor to cut the heat if the column starts to flood or puke. That might be enough to go with a design that has no visual feedback on the internal operation. What do you think Stilly?

There's an article on 'distillation tower flooding' that can be found in your local library. It discusses the problems with using pressure changes to measure column flooding — namely, that by the time the pressure drop registers, the column has already started to flood. If this holds true for hobby sized columns, then a sightglass is going to be a better way to monitor column flooding. But, in the absence of sightglasses, a pressure sensor is better than nothing. A particularly sensitive pressure sensor might be even better. I'm not sure.

Why not just pack it solid all the way up and make the most use of all the space?

If you're trying to produce neutral, then packing it solid might be an option. But then you've just got a packed column.

If the rave reviews about the quality of product that you get out of a trayed column are anything to go by (and I think they might be), there's something about how distillation occurs in trayed columns that sets them apart from packed columns. That 'something', I suspect, has to do with the fact that each tray acts as a radical distillation stage — quite different to the 'smooth' distillation that occur in a packed column.

If you pack a trayed column solid, you're increasing the efficiency of the column but you're also significantly changing how it functions. The question is whether by changing how it functions in order to make it more efficient you risk losing the very virtue of the trayed column: staged distillation.

:think:

[Anthoney]

Re the pressure sensor I have a modified version of the one used by the Russian and Polish hobby distillers

http://translate.google.com/translate?hl=et&ie=UTF8&prev=_t&rurl=translate.google.ee&sl=ru&tl=en&u=http://www.svoimi-ru-kami.narod.ru/MeterPress/Meterpress.htm

http://translate.googleusercontent.com/translate_c?hl=et&ie=UTF8&prev=_t&rurl=translate.google.com&sl=ru&tl=en&twu=1&u=http://www.svoimi-ru-kami.narod.ru/Spirt/Rectific.htm&usg=ALkJrhhczCfeqX1uLwtlrNWFNhxlOF2Aow

We are talking of maintaining 3 to 5 mm Hg

Mine controls the heaters where the original simply gives a precise readout. Also mine, well my mate Tom made it for me, uses more modern and easy to get parts. It's a mark2 essentially.

Neutral is my aim so packed is good for me. The reason for the tray is to improve the efficiency/speed vs hight/dia ratio of the column. That is the aim of the hybrid for me.

If I want to make a brandy instead for example I would expect to have to detune it in some way to be more closely determined when I have some actual experience with using it. But removing packing basically.

It all makes sense to me but what do I know apart from second hand knowledge from the work of others?

Which is why I ask opinions from experienced distillers like yourself and anyone else who wants to chime in.

[stilly_bugger]

It all makes sense to me but what do I know apart from second hand knowledge from the work of others?

Which is why I ask opinions from experienced distillers like yourself and anyone else who wants to chime in.

Just to be clear, my experience is with regular pot stills and packed columns only. I have never run a plated column. But I do research them, and peer-reviewed scientific papers are as good a source of info as any if you want to understand how they work. Just as you don't have to drive a car to know how an internal combustion engine works.

[Anthoney]

Good enough for me stilly. It was kind of implied from your comments on plated stills that you hadn't run one. But I'm sure you know more than I do all the same and have the calibration of experience with pots and packed to inform your reading.