jheising

Fun with Spiral Prismatic Packing!

39 posts in this topic

1 hour ago, Silk City Distillers said:

What size Raschig rings are you using?  What's the column diameter/height?  Interested as I'm working on a 8"x20' column, and have been hunting for packing.  The fact that you didn't see much difference between SPP and Raschig is actually positive, as it's much easier to source.  On the hobby side, those guys tend not to be so concerned about fouling, but with a big column, the packing is never leaving the column, and needs to be effectively cleaned in place.

 

When I say we didn't see much difference, I meant for our particular use. We're actually using the rings to slow down liquid more than to provide surface area for reflux. And we only use it in one part of our still. I'd say if you were using the packing as it was meant for—more surface area, higher flow rates— I think SPP would still be superior.

We only looked at SPP initially because it seemed like a nice way to continuously produce a lot of packing very inexpensively. Rather than cutting pipe all day, I imagined turning on a machine in the evening and waking up the next morning with a mountain of packing. But since then we've recently found some good ways to produce a fair amount of raschig rings in a short amount of time, so the SPP became less interesting. Although I do think we may revisit it in the future for other parts of our still.

And you're right about the "effectively cleaned in place". We run our continuous still for months at a time so we can't bring it down to clean. If you're curious we're using 1/2" nom. type M pipe. So the rings end up being 0.625 x 0.625. If you willing to trade money for time, Wisconsin Stamping (http://www.wisconsinstamping.com/raschig-rings/) has been a good source for us in the past and they can make them in pretty much any size.

Share this post


Link to post
Share on other sites

This is my attempt at translating what Cezary Trzciński, the mastermind behind GENIO Stills wrote in Polish:

@jheising - Yes, we do have "some experience" with SPP. I myself have been working with it for several years.

You're right, it is important whether it is "SPP" or "SP" or Raschig Ring ("RR"), but even more important is matching of appropriate PARAMETERS with the packing.

The shape is important - provided that the filling has the right conditions to take advantage of the shapes qualities.

An example with cars (sorry - but I like them very much): If we have a great, powerful, most modern engine in the car (the engine in this context will represent column packing - it performs the main job) - but if you do not give the engine good fuel, or give very little fuel or a lot, or unevenly - then even the best engine will be worth nothing.

It is the same in columns - fuel for the column vapor is generated from the boiler. It must be constant, repeatable, measurable and quickly adjustable (therefore best conditions can be achieved through electrical heating). But the most important MUST BE THE RIGHT AMOUNT.

 

Packing (SP, SPP, RR) will work well only if they are kept in proportion with THREE FACTORS:

 

The first factor:

The surface area of evaporation of the liquid in the tank and the volume of the liquid in relation to the specified heating power "produce" a predetermined amount of vapors - which must "pass" through the packed column. Too little gas - the column will not be "hydrated" - which will result in loss of column stability.

Too much gas - will flood the column - total distillation"flop".

 

The second factor:

Another very important factors are the dimension of packing, including the diameter, height, wire thickness (wall), angular offset of packing (SP, SPP, RR). The packing has to be adapted to the diameter of the column and the column height. We have a packing PARADOX:

 

(a) You must have the largest surface area (contact) with the vapor generated from the boiler - or as much wire "shoved" in the column. Then we will be able to thoroughly purify the alcohol.

(b) On the other hand, vapor should have least amount of "obstacles" to allow it to reach the top of the column so it can be condensed back into liquid alcohol. So it would be best if the column was empty inside.

Therefore, to minimize resistance, but maximize the area of contact with the vapor, "experiment" with SPP began. And again:

Too "packed", dense, small packing (SP, SPP, RR) - little vapor passage and column will flood or get "clogged".

Too loose, too big packing (SP, SPP, RR) - not enough HETP - "dirty" alcohol produced.

 

The third factor:

The material from which packing is made (SP, SPP, RR) - we must remember that inside the column there is an "aggressive" environment - ethanol combined with very high temperature - the packing material must be resistant to such conditions (changes in the surface structure of the wire, ceramics, etc.) - the weaker the material, the more resistance will be created ("porosity" formation - much larger drops are formed on the packing and they "block" the flow of vapor - and again, "flooding" the column - unstable and bad work. Here it is very important to maintain cleanliness - almost laboratory purity (leftovers from vapors, scale deposits out of water, oxidation etc.). Further the material is very important because of the "pressure" that is created by column packing. Consider the amount of weight applied by column packing on packing at the bottom of the column. In addition, during the distillation process when vapors travel through the column - packing is constantly "twitching" and becomes more concentrated - again causing a reduction to vapor passage and "flooding" of the column.

 

And only once the three conditions are met, then we will be able to achieve a well-functioning rectifying column. But not a fully 100% perfect column (but I write further).

 

In reality, here in a "quick" manner I only showed the "tip of the iceberg" - the complexity of column packing (SP, SPP, RR) is very broad and the decision should not be reduced only to how to produce SPP or SP. Here, first, you should be asking WHAT SPP or SP I need.

In our company for many years we experimented with the most varied packing. Through trial and error, and that is how we got some results, for example:

Our company through my trials and errors on hundreds of test processes, achieved and utilized over the past 5 years packing with parameters congruent to very specifically sized tanks and appropriately fined tuned heating power. And still, in some cases column flooding happened, or something else is not working as it should and so on. So even though the essential THREE CONDITIONS WERE SATISFIED 100%, from time to time there is something else that was not as it should be.

Then (as a manufacturer) we have no influence on cleanliness of our customer’s column packing, the physical condition of the columns and operator’s control of the columns, but we can have an impact on heating - to this conclusion I have come a long time ago, but only now after more than 2 years of trial and error we managed to bring to achieve full automation and automatic control of electrical heating (that is how long it took us to record results of the most varied conditions in the form of algorithms for subsequent programming). For example, on the production capabilities of the column and the appropriate column performance (although the parameters were chosen ideally) the customer still has an impact (and that is not all): how much wash is in the tank, what is the density (thickness) of the batch, what is the alcohol content of the vapors, what is the packing "compaction", how clean is the packing, we also have to consider the conditions of the environment’s temperature and pressure, and finally one of the major ills I noticed in the US - what is the actual voltage supplied?  All these variables affect the stable operation of the column, and we were able to overcome them just by creating adaptive heating to compensate for all these factors automatically.

Sorry for my "chaotic" language - in addition unfortunately my English is at a level of a "tourist" visiting your country - that is why I’m obtaining translation assistance from my friend and GENIO representative for Canada and Western US - Adam.

Kind regards to all current and future GENIO Still owners. Cezary Trzciński.

Share this post


Link to post
Share on other sites

Thanks Genio, that was indeed helpful. There is a lot of useful knowledge and experience in Poland and Russia which is not always easy to tap into!

Equation 7 of the following paper explains Langmuir's derivation of an equation quantifying all of the relevant parameters for successful fractionation.

It's heavy going unless your maths ability is up to it (I have a SmartMaths pal who broke it down into small bits for me initially!)

http://bado-shanai.net/Map of Physics/mopLangmuirEvaporation.htm

It's an equation that "grows" on you and FWIW, the thing that I had most trouble getting my head around was "the flux of vapor molecules in kilograms per square meter per second onto the interface" which is just another way of saying that "whatever you do, you will change the rate of exchange between the liquid and gas phases and that the mass of the total exchange, per second, will be directly proportional to the gas/liquid interface area. That exchange can be from liquid to gas (positive = evaporation) or from gas to liquid (negative = condensation).

Intimate realisation of the meaning (at least) of this equation is probably fundamental to being able to design improvements in any fractionation process.

Edit: I almost forgot...

What isn't so obvious is that if either liquid or gas phase is moving with respect to the other,  via Bernouilli's equations, there will be a pressure change and the rate of molecular exchange will change (Vapour Pressure in both phases will change, therefore the flux.)

Which is why your wet face feels cooler the faster you cycle along (increased evaporation and thus energy loss from liquid to the gas phase)  until the water has gone, and then your face gets warm again.....non-constant gas or liquid material flow in a continuous still will lead to a similar potentially wide ranging variation of the fractionation expected (steady state destroyed).

In a pot still, the effect is just as unmanageable, but much harder to pin down. Which can lead to all sorts of misunderstanding of basic fractionation principles......

 

 

Edited by MythBuster
footnote added

Share this post


Link to post
Share on other sites

So good to see someone doing this properly. It doesn't matter that prototypes are hand-crafted, as long as they are able to collect useful data.

My own are made that way too (I am a well-equipped model engineer, with solid background in computing hardware and software too).

I tend to use Arduinos as cost-effective prototyping compute engines. Can't quite see if yours is Arduino too?

I can get a 16MHz Arduino for about 5 Euros (£3 UK) from China, post paid!

 

Around 1980, I wrote the software to fully control a 500,000 litres per week (Heinecken) lager fermentation room.

It ran on a 2MHz 16-bit computer with  32K(!) of ROM/32K(!) of RAM....

In case you haven't discovered the "trick" yet.... when prototyping and you need many buttons/switches, it is a nuisance putting a lot of wiring together to get them linked to your control computer. I use an IR keypad, serially linked to an Arduino, Grab a Public Domain script to get it working, and than have all the prototyping buttons I need in a nice, easy-to-use keypad.

With no trailing wires because of the IR link.... it works very well indeed.

Probably not recommended for Production use, but for prototyping a HUGE time-saver!

 

I've also found an Arduino especially useful for hooking together MULTIPLE DS18B20 devices. Really low-cost and really effective telemetry.

I typically hook the Arduino up to a laptop and collect huge amounts of VERY useful temperature-profile data on my fractionators. All with respect to time!.......

Good to have discussed with you.

Warm regards.

 

 

Share this post


Link to post
Share on other sites
21 hours ago, MythBuster said:

It's heavy going unless your maths ability is up to it (I have a SmartMaths pal who broke it down into small bits for me initially!)

http://bado-shanai.net/Map of Physics/mopLangmuirEvaporation.htm

It's an equation that "grows" on you and FWIW, the thing that I had most trouble getting my head around was "the flux of vapor molecules in kilograms per square meter per second onto the interface" which is just another way of saying that "whatever you do, you will change the rate of exchange between the liquid and gas phases and that the mass of the total exchange, per second, will be directly proportional to the gas/liquid interface area. That exchange can be from liquid to gas (positive = evaporation) or from gas to liquid (negative = condensation).

Interestingly enough, Langmuir originally derived this for solid sublimation, when studying failure of light bulb filaments at GE.

Share this post


Link to post
Share on other sites

So how does this translate into real world performance?

A year back I immersed myself in the mathematics as well as spending a considerable amount of time reviewing literature, paying specific attention to commonly used industrial methodologies.  I'm no mathematician by training, but I didn't think it was an insurmountable task if one was patient and inquisitive enough.

My conclusion, after all of that work, was that minor improvements in system efficiency didn't necessary translate into major real-world benefits at all.  In fact, what could be considered serious efficiency issues could be easily "corrected" through what were relatively minor changes to geometry, or specific operational parameters.  For example, even minor changes in column height/theoretical trays would easily compensate for what would be poor Murphree efficiencies.

You don't need to look much further than the hobby community to see this translated into real world performance. Even still designs that shouldn't at all work theoretically still see passable performance.  Realistically, you could fill a column with any compatible material and get passable efficiencies.  Anybody who has built a still themselves, or tinkered, would attest to this.

And all of this is somewhat moot if we're talking about distillation of products other than neutral alcohol, as by their very definition, these are products who are created through inefficiency during distillation.  The fact that many very good products can be created on a pot still, or through double distillation on a pot still, really is significant evidence that looking to maximize efficiency doesn't necessarily translate to specific benefits to beverage distillers.

Probably the most striking realization I came to was that to achieve high levels of tray efficiency, you needed to design a system to operate in an incredibly narrow range of operating parameters.  This was critical.

However, these constraints on operating parameters would place significant limitations on the distiller to run various types of products, or to adjust the operation parameters to suit a specific product.  So while they would give you ideal performance in an industrial arrangement, where nearly zero variability is expected, in the beverage world, this means a still that only does one thing well, when run in a very specific way.

Just my two cents.

Share this post


Link to post
Share on other sites

Don't get me wrong, I love technology simply for technology's sake.  I see Genio's machine as being a work of art, and not just a machine.  I also fully understand the need for competitive differentiation in the marketplace.

Just playing devil's advocate, because it's more fun to take that position.

Share this post


Link to post
Share on other sites
7 minutes ago, Silk City Distillers said:

So how does this translate into real world performance?

A year back I immersed myself in the mathematics as well as spending a considerable amount of time reviewing literature, paying specific attention to commonly used industrial methodologies.  I'm no mathematician by training, but I didn't think it was an insurmountable task if one was patient and inquisitive enough.

My conclusion, after all of that work, was that minor improvements in system efficiency didn't necessary translate into major real-world benefits at all.  In fact, what could be considered serious efficiency issues could be easily "corrected" through what were relatively minor changes to geometry, or specific operational parameters.  For example, even minor changes in column height/theoretical trays would easily compensate for what would be poor Murphree efficiencies.

You don't need to look much further than the hobby community to see this translated into real world performance. Even still designs that shouldn't at all work theoretically still see passable performance.  Realistically, you could fill a column with any compatible material and get passable efficiencies.  Anybody who has built a still themselves, or tinkered, would attest to this.

And all of this is somewhat moot if we're talking about distillation of products other than neutral alcohol, as by their very definition, these are products who are created through inefficiency during distillation.  The fact that many very good products can be created on a pot still, or through double distillation on a pot still, really is significant evidence that looking to maximize efficiency doesn't necessarily translate to specific benefits to beverage distillers.

Probably the most striking realization I came to was that to achieve high levels of tray efficiency, you needed to design a system to operate in an incredibly narrow range of operating parameters.  This was critical.

However, these constraints on operating parameters would place significant limitations on the distiller to run various types of products, or to adjust the operation parameters to suit a specific product.  So while they would give you ideal performance in an industrial arrangement, where nearly zero variability is expected, in the beverage world, this means a still that only does one thing well, when run in a very specific way.

Just my two cents.

I do get all of your 2 cents (I have seen your other post before completing this, on "devil's advocate"; see your point; and have no problem with that at all!) and it seems to me that hard science can be accepted in two primary ways:-

1. Those who understand it and then bear it in mind as they do their own thing, formulating their own conclusions about the effects of the Scientifically nailed-down physical phenomena. They do tend to call this process "craft" or similar. And I make no mistake: the World needs Craftsmen (and Craftswomen!)

2. Those who understand it and then keep it in sharp focus as they explore ways to apply the characterised phenomena knowledge to progressive and fully-characterised improvements. Generally called Science. It tends to be slower and methodical, but it is only Science which precisely, repeatedly and mathematically characterises the phenomena of daily life.

NB: neither is wrong, nor good/bad.

I'm aware of both approaches and mostly follow (2), but do sometimes "think outside of the box" and play around in the (1) domain.

That said, having seen for myself that careful application of the available scientific knowledge can indeed produce a continuous still which comfortably and repeatedly (to accepted and measurable scientific standards) actually works. I fully realise that conventional distillers NOT so scientifically-inclined place huge doubt on that, even though such systems are commercially available (by using the available Science) as exemplified by Member jheising at website,

http://bunkerstills.com/

Why does (2) especially attract my sort of person?

I am something of a polymath and like to model most designs I get involved with (distilling is just one such avenue.)

I can state categorically that my prototype nanostill, within reasonable experimental limits, follows PRECISELY the parameters of its mathematical model. I also know (for sure!) that unless the prototype is run in line with the parameter limits encompassed by the model, it breaks down and delivers junk. That said, once steady-state is achieved (mine takes 10 to 12 minutes from cold) output remains rock-solid until shut-down. Which can be 7 days ahead, or more!

I'm satisfied that my design MUST therefore by very much aligned to jheising's design m.o., albeit operating at dramatically lower throughput and power requirement.

I have no doubt at all (having already modelled it and prototyped it) that if I swapped my own-design fractionator packing  for "scrubbies" and stretched its length 2 or 3x then it would work JUST as well and deliver JUST as reliable steady-state output.

I do not intend to disclose my packing design at this time, but hope you can see that its precise nature is relatively unimportant to the current argument?

So a successful continuous design is NOT a function of a low HETP column packing. That said, I also have no doubt at all that I cannot keep my continuous still perfectly balanced (steady state) UNLESS I take full account of the second-to-second profile of  the fractionator's temperature gradient, power input, reflux ratio and wash feed rate. Having seen Langmuir's formula, that is completely understandable, so I have designed "to" it.

FYI, I have an outlandishly novel prototype which has no column at all, or trays, relying instead on another means to produce a large area gas/liquid interface. It too is in alignment with its Langmuirian model and it has no packing at all! It is more difficult to "scale down" to NanoStill size, so is on the Back Burner for the time being.......

Conclusion: The successful design of a "flavored-output continuous still" of ANY size MUST, IMHO, have a control system which keeps all pertinent parameters continuosly aligned according to the Known Science.  The design of a continuous "stripper" is an entirely different animal, and relatively trivial.

Do all please maintain the flow, this is an interesting discussion/debate.

I do hope jheising is still about and can join in?

(I'm about to go offline until tomorrow.)

 

 

 

 

 

 

Share this post


Link to post
Share on other sites

SPP is nothing new to industrial applications. I can find white papers back to the 1940's when it was used to fraction tritium. It has the lowest HETP and can provide the greatest number of theoretical plates in a given space.

It is easy to see why it would be great for the production of neutral ethanol. More plates equal greater purity.

But just like trays vs diameter vs liquid depth vs vapor speed vs height of the column, etc,...SPP, size, shape, material has to be applied for the situation.

If you have a limited ceiling height or can dedicate a still to vodka, SPP can get you the highest ABV and or greatest through put in the shortest column. The physical mass of SPP also makes it efficient, less power due to less heat loss.

Most of the time, in practical applications, people buy a packing not suited for their column diameter or height, and wonder why it fails.

How many of us actually try to optimize the liquid depth on our plates? We'll getting the size of SPP right for a column still is more time consuming and costly than redesigning your downcomers.

Share this post


Link to post
Share on other sites

Random Packings, SPP, Dixon rings, Raschig rings, Mesh, Saddles, Pro Pak, Balls (I won't say marbles), stainless, copper, glass, ceramic...they can all be great applied appropriately. But SPP still has the lowest HETP (more plates per height). It's also the most expensive to acquire. But in Stainless Steel it is a permanent solution.

There are even columns using rocks...

For neutral, these packing's are more efficient than bubble caps or sieve plates, size and energy required. A lot of that efficiency comes from the mass of the Steel versions transferring heat so well.

The petro industry regularly uses Structured Packing...another story.

 

2 people like this

Share this post


Link to post
Share on other sites
15 hours ago, Falling Rock said:

Random Packings, SPP, Dixon rings, Raschig rings, Mesh, Saddles, Pro Pak, Balls (I won't say marbles), stainless, copper, glass, ceramic...they can all be great applied appropriately. But SPP still has the lowest HETP (more plates per height). It's also the most expensive to acquire. But in Stainless Steel it is a permanent solution.

There are even columns using rocks...

For neutral, these packing's are more efficient than bubble caps or sieve plates, size and energy required. A lot of that efficiency comes from the mass of the Steel versions transferring heat so well.

The petro industry regularly uses Structured Packing...another story.

 

I have found that especially, and necessarily, in a continuous still, once the fabrication metal (or any other material) of a fractionator packing, or plate for that matter in a plate fractionator, is at the temperature governed by the equilibrium temperature of the gas and liquid phases surrounding it, you can consider the thermal (and Mass) characteristics of the packing material irrelevant  to the thermodynamics of the still operation, in comparison to the heat transfer associated with the molecular exchange solely in the surface area between gas and liquid phases. Once at the desired equilibrium temperature, that packing (or plate) does almost nothing thermodynamically except in its contribution to conducted Heat Losses!

Proviso: If the design relies upon such conductive heat losses to modify the fractionator efficiency/effectiveness, then the REVERSE applies! But in my book, that is poor thermal design and energy wasted.

 

It is correct to assume that SPP's are fabricated from wire which is strong enough to support the likely weight of other SPP's above them, without undue mechanical distortion.

And I repeat... there is more to selection of a packing medium than mere "HETP", which takes absolutely NO account of throughput capability. Whilst there is general acceptance that SPP does provide  very good working solutions, that does not mean that SPP is providing the most cost-effective solution in all or even just those cases.

If a material of 90% SPP's general effectiveness but at 50% of the cost were used instead, would it be considered inferior.....?

Perhaps fractionator packing choice becomes more meaningful, in practice, if one considers the methodology of "tuning" a given length of packed column (usually via precision reflux control) so that the degree of separation can be finely selected for hearts, slightly pre-hearts or slightly post-hearts without having to mechanically alter the column's geometry or physical take-off point(s).

In the case of flavoured spirit output(s), the absolute effectiveness of separation vs column height (HETP) is rather inconsequential compared to the needs of throughput and a wide range of separation tuneablity.

In my own NanoStill case, I selected a column packing which is - compared to SPP's - much more tolerant of wide-ranging reflux ratios which gives me a wide range of component separation tuneability.

Great discussion!

Share this post


Link to post
Share on other sites
6 hours ago, Silk City Distillers said:

@MythBuster You are still talking about packing in the context of continuous operation, outputting final product, correct?

Your device uses a fixed product take-off point?

All of my discussion revolves around continuous distillation.

The Product output point(s) are user-defined. The "Q" (quality factor) of separation at EACH output point is tuneable. Such points can be added/removed.

 

 

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now