@grim said: Jesus, 14 years ago…

Violentblue's Steamed Gin @ HD

We were all doing so much experimentation at that stage. most of us didn't have the science of it like you guys, a lot was 'what if'.

Everything from equipment, to methods, to recipes.

@Smaug's thumper pumper was out there too.....

@olddog brought the bubble plate column to the hobby world not long before we all started StillDragon and doing all our product development.

Was a cool time to be part of the hobby.

Jesus, 14 years ago…

Violentblue's Steamed Gin @ HD

Not at all wrong.

I tried this once using 100% neutral, on accident. Had the water valve off, so I was pumping only neutral. Flavor profile was very different. I let it run through for a while for the hell of it. Then, not sure why, I decided to just try running water alone, and just running steam through.

The end result was very similar to the normal run. Each individual component was not. The steam distilled botanicals ran white like milk. But as soon as the distillate hit the very-high-proof first runnings, it went clear again.

Toyed around a few times after that, but when I ran a vapor proof that was identical to my batch distilled product proof (before dilution), the end result was identical.

The best I could posit was, the same amount of alcohol vapor and water vapor is passing through the botanicals. The only difference is that they are constant through the run, and not starting high/ending low/vice-versa.

Assuming you are getting full extraction of your botanicals, why would the end result differ? it all comes down to grams/liter of extract and oils per finished bottle. Keep those ratios identical, it won't matter.

You could get very fancy with this setup, and vary the ratio of water/ethanol through the run, in a manner similar to a step mash in brewing. (5 minutes at 190 proof, 5 minutes at 150 proof, 17 minutes at 130 proof, 8 minutes at 110 proof, etc, etc). You could actually get better consistency doing this than you ever could with a batch still.

@Bolverk said: Croz, I think that's how Smaug is recommending doing it. Standard CM upper configuration but the dephleg is tuned to only let heads past and then condensed in the product condenser.

Yessir. Definitely helps when calculating actual processing speed.

Steve Todee in Bastrop has settled in on 2L of heads collection for every 100 gallons of beer as an example. Not saying this is "thee" be all concluded ratio btw. It's going to depend on everyone's sensory awareness perception.

@Smaug, I see what you're saying. What I'm thinking is a cross between the GNS continuous and the LM function of the spirit side of your continuous... like a boka but you're taking off the heads at the LM port and your good spirit out of the bottom of the column.

Yeah, I'm trying to work out a few things too including:

Reboiler HX appears to be a 4" x 750mm shell in tube

column

• Diameter? looks to be about 8"
• Length? about 1m?
• Plate type: sieve vs bubblecaps?
• plate count: 4?
• how do the level sensors control the operation of the unit?

condensor

• looks like a 4" x 1m long
• How does it work to take off heads? Are they drawn off via the vent? As there's a flame arrestor I don't think so as it'd also need a condenser for the heads take off
• Is the line from the bottom of the condenser to close to the bottom of the column the condensed hearts?

Pumps

• What would be the specs of the 2 pumps used for feed in and product out?
• Do they run continuous or does the level sensor(s) turn it on/off?
• I think they'd need to be ATEX/IECEX = $$$
• Could double diaphram pneumatic pumps be used instead of electric?

At 2gpm that comes to 462lph! 96% GNS has a boiling point around 78C so the Preheat HX & reboiler don't have as much work to do compared to running a 10% wash which needs to get to about 93C. So trying to work out what I could do with my available power

The operating principal is pretty simple:

1. There is a single pass boiler (not reboiler), which vaporizes near 100% of the feed, injecting that vapor into the column, which operates like. thumper. There are no plates in the column.
2. Vapor moves up into the condenser, gets condensed, and flows BACK INTO THE "column". You can see the drain located about one quarter/third from the left on the top condenser. It flows to the bottom of the column.
3. Column liquid volume is controlled by the level sensor, which operates the discharge valve. There are two pumps, a feed pump, and a discharge pump.
4. There is a temperature probe on the vapor vent from the top condenser (port at the top, far end). There is a PID (visible ABB unit) that controls the water flow through that top condenser using the vapor vent temp as process input. The set point controls the amount of input that's removed as vapor.
5. The discharge pump drains from the bottom of the column, through the heat exchanger, out.

Everything in the "distillation" stream is product, there are no bottoms (as this is GNS), the only thing removed is a very small amount of "heads" as vapor.

@richard said: Seems excessive. Say 125mm diameter x 30 = 3.75m. Normally a minimum of 1.2m packed section is sufficient plus RC etc. on top. I could relate to this if it were plates at pitches of say 100mm and 30 plates were the requirement.

Thanks. Yeah I can't remember what that rule of thumb was.

@Smaug said: No sir. The diameter to height ratio for columns with random packing materials.

Seems excessive. Say 125mm diameter x 30 = 3.75m. Normally a minimum of 1.2m packed section is sufficient plus RC etc. on top. I could relate to this if it were plates at pitches of say 100mm and 30 plates were the requirement.

Both methanol and ethanol will react with long chain fatty acids found in meat or vegetables to form the precursors of diesel fuel (FAME/FAEE) - basically just long chain esters. They don't all come across as heads products, the longer chain stuff smears across hearts, tails, and beyond.

I could see locking up some methanol in longer-chain esters, and being able to cut that out as tails.

The protein stuff, that stumps me. I need to study aldehyde chemistry more, aldehydes will react with proteins and amino acids.

Stuff like:

Understanding interactions among aldehyde compounds and porcine myofibrillar proteins by spectroscopy and molecular dynamics simulations @ ScienceDirect

Understanding interactions among aldehyde compounds and porcine myofibrillar proteins by spectroscopy and molecular dynamics simulations

porcine myofibrillar proteins = pork meat

Among flavour compounds, aldehydes have a significant role in flavour development of meat and meat products [9], [10]. Some studies describe the influences of aldehyde characteristics (such as structure and concentration) on their interactions with protein, and some investigations concluded that aldehydes react with amino acid residues of proteins via irreversible covalent binding [20], [21]. Other studies reported that hydrophobic interactions were the predominant force responsible for the interactions among proteins and aldehydes [22]. Based on these studies, both reversible and irreversible binding modes appear to be the major contributors to the interaction among proteins and aldehydes, depending on the specific chemistry and structure of the interacting molecules. Nevertheless, information on the conformational changes that occur in MPs upon aldehyde binding remains limited, and the mechanisms of aldehydes binding to MPs remain unexplored.

Probably aimed mostly @grim.

Is there any evidence to suggest that having fats in the kettle (in this instance a petchuga styled chicken in the vapor path or in the kettle charge) would have a reactive relationship with heads constituents in such a way that it would sequester perceived heads?

Here is the dude's assertion:

About serious cabbage soup.

My colleague carried out a laboratory analysis of a simple distilled sugar mash with a meat composition and vegetable oil IN ONE TIME WITHOUT SELECTING THE FIRST DROPS.

• Aldehydes 68.1 mg/dm3 (norm – 10-350)
• Ketones (acetone) 0.5 mg/dm3 (norm – 0.1-2.34)
• Esters 337.5 mg/dm3 (norm - 10/50 -1500)
• Methanol 0.0005% vol. (norm – 0.05)
• Higher alcohols (fusel oils - 1658.6 mg/dm3 (norm - 500-4000/6000)

THINK: IN ONE TIME WITHOUT SELECTING THE FIRST DROPS! These are impressive results that are better than the results of double or triple distillation.

The colleague promised to continue to conduct comparative analyzes on oil seeds and meat products. There is a suspicion that the distillate is purified not only by fats, but also by proteins. He must find out.

A short packed section with some structured copper mesh is what most of the hobby guys here use on top. I do a 4" x 510mm with mesh screen, clamp and gasket and i sell a heap of them.

@BorisTheMidget said: thanks everyone for your responses!

punkin - can i ask for a little more info on this one? When i spoke with the manufacture, they we indicating that 4 plates would be enough for a high quality vodka. To help me learn (i am not trying to be rude), do you believe this is not the case, and could i ask why?

Ok firstly, have a look at anyone making a premium vodka. 16ish to 30 plates would be the tool they use. 30 being the theoretical max. Anyone making vodka with a 4 plate column is just not really making vodka or having success marketing a premium vodka spirit unless they are at least charcoal filtering.

Also consider that less plates than that can certainly render high proof. Lets take a 10 plate column for example (or perhaps even less). If you are able to achieve 95% abv with 10 plates it will not be as "clean" as the same 95% driven through 16 plates or more. Why? Because you will need to run at a higher reflux ratio with the shorter column to maintain proof. The higher reflux ratio means that you will be dragging up "extra" flavor congeners. Particularly with the 4 plate column. Higher plate count with lower reflux ratio is the best way to render the cleanest spirit possible.

The 4 plate column does not at all have the ability to make a proper modern vodka. It is a tool that does much better making flavored spirits like rum , whiskey and brandies.

Now,,you can certainly do multiple runs to clean the spirit up to the extent possible. But in doing so you will loose product and spend additional resources with respect to labor hours and utilities. Now your "basic" vodka costs more to produce than a more desirable whiskey. This is why so many folks source NGS for their vodka. The least desirable spirit costs the most to make. In other words, you can very frequently buy NGS cheaper than it costs to make it unless your distillery is outfitted correctly.

The equipment will never cost more than the labor to operate the equipment over the service life of the equipment. If you want to make a premium vodka, buy a dedicated vodka still.

Use the short plated column for your flagship flavored spirits and put together a smaller dedicated vodka still. It will likely be less expensive than putting together one of those "jack of all trades , master of none" type stills.

Punkin will likely be along to fill in the gaps.

EDIT: Ah wait. Are you a hobbyist? If so the production cost concerns don't have the same priority. You can make "vodka" on any still.

@SLOPEmeisteR said: grim crozdog thanks for your input. I'm a bit happier now knowing 380lt boiler with 8" Dash Pro can run spirit with 15kW, giving me loads of capacity... for now. Simple dimensions check, if you can help. What's the floor to superstructure opening height on the 380lt boiler? I've got the column (standard 3 plate config) at 1160mm and extra sections at 203mm from the web shop, so can work out what fits!

The single wall is right around 1100 + or - mm tall at the column connection.

The Electric Jacketed version is roughly 1219 mm + or - at the column connection. The conventional steam kettle should be just about the same as the Electric Jacket.

Hi @punkin - I hear you on volume/work load! Can always charge less than 380lt, but mostly it'll be a full charge to maximise the capacity and volume output. Maybe smaller charge for trial runs, but I've got the 100lt/4" for that anyway.

Hi @crozdog - I'd prefer to go 8" if I think the powers enough, commercially it makes sense.

Heat up time is what it is really, max power available, how long will it take. We all work with that.

The only metric I have to select the column diameter is vapour speed.

Looked again at my existing 100lt/4"Dash (3 pro-cap plates), vapour speed is actually 16.03 on the run at 4.4kW. Your running 380lt/8", vapour speed is 12.75 on the run at 14kW.

I could do 380lt/8", vapour speed at 13.66 on the run at 15kW (leaving 5kW for the house load; max 20kW incoming main).

I do run my 4" quite hard then at 16.03 vapour speed, I use a good bit of reflux on this to maintain product flow. With 8" running at 15kW the heat to cooling ratio would be changed i.e. less reflux cooling at this lower vapour speed to maintain product flow. Is that the right thinking?

Really my big concern about going up to 8" was will I be able to power a decent enough flow rate, sound like I can albeit with a lower vapour speed than I'm used to on the 4".

Also 100lt to 380lt is almost quadruple, 4" to 8" is quadruple area. Roughly speaking then my heat-up/run time will be around the same for the same wash.

When I say sugar wash I mean molasses, demerara and granulated sugar blend, so there are some interesting things in the heads/tails and I do want to barrel age. Currently ageing on Still Dragon supplied oak woods, done one whisky cask aged for 8 months and it was outrageously good and sold out fast!

If you're running a reflux column, you don't need to do a stripping run first. You can do it in a single pass.

If you can let us know more about the wash, your rig and what you're trying to achieve, It'll help you get better answers.

yeah 12l in a 100l still isn't enough. If you're keen on doing strips before spirit runs (typically done when running a pot still) you can save and combine several strips of the same wash to get the required volume for a spirit run to ensure that not only do you cover the elements at the start, but importantly at the end of the run they are still covered.

Hey Folks,

I have ~80L of wash for a 100L still, which is now at the point it's ready to go through a stripping run. Using a wash calculator, it looks like we'll end up with about 15L of spirit post stripping run. This is my first time, so i'm guessing as i've got the reflux column running, we'll be able to collect (after foreshots / heads / tails) around 12L of usable spirit for spirit run.

Question is, given that i've got a 100L still with about 12L for the spirit run, what should i add back into the collection so i can re-run it through the still? i'm worried that if i pop that 12L in, its not enough to cover the element at the bottom and i'll bugger the heating element. From memory, i think it needs about 5 or 10L at the bottom of the still to cover the element.

thanks!!

@SLOPEmeisteR Have a look at the calcs page

I used it to calculate your heat up time for 380l with 15000W & got an answer of around 2.5 hours to get a 10% wash to 90C @ 90% efficiency.

That page also has a vapour speed calculator ;-)

IMO the 380l is a perfect match with 8" columns. Remember diameter = speed. As a commercial operation, time = money. Smaller diameter = Longer runs = increased cost and lower production volumes :-(

I don't know why you say you can't get the power to run an 8". I have a 380l with 4 x 8" plates & use 21kw for heat up but cut back to 14kw for the run.

Good cuts come down to how you run the rig (eg power input, reflux ratio, timing, if you do a strip then spirit run etc). If you're going to be barrel aging you don't need the hearts cut to be as tight because you actually want a bunch of the congeners in heads (& late tails with rum) for flavour (but you won't get much of that on a sugar wash).

FYI, adding a few more plates will only take a little extra time to heat up. The longest time is getting the wash up to temp, getting the column to temp is pretty quick once the wash is boiling.

As you're operating in a pretty cold environment, don't forget you can insulate the boiler & column to improve thermal efficiency & hence overall performance of the rig.

Hi @grim,

Yeah I've got space and power limits to work within. I've converted my garage at home to a nano-brewery/distillery, fully HMRC licenced and selling product online and markets. I'm amazed at the quality (as are customers) I'm producing rum at on single pass through 3 plate 4" Dash, wow!

I put 12 kW solar on the roof last year, but that's only good for 8 months of the year, so working with limitation of single phase mains. The mains will do around 20 kW but I need to leave some for the wife to run the house! I recon 15 kW is a safe limit for the brewery/distillery.

This upgrade gets production capacity on rum alone into significant figures, with literally no overheads (no rent, rates, etc). As I approach capacity on the upgrade kit the business will be well placed to go for a much larger upgrade, commercial premises, and have the books to raise capital for investment. It's a plan to grow with as low a risk as possible.

Hence a bigger boiler at 380L than perhaps a 6" column requires, but I can't get the power for 8" and a 380L charge on 6" column will give a nice long run with long clear cuts I think. 380L boiler will never go a miss in a craft distillery. In the sunnier months (in Scotland, don't laugh!), I'll have up to 32 kW to play with depending on house load scheduling, this would let me run the upgrade still and my existing 100L 4" Dash... thinking maybe stripping runs on the 6" then run smaller still as a rectifier to produce vodka (might need to go 4 or 5 plates each on 4" and 6" to hit vodka).

Current 100L boiler has 5 kW heating and takes around 2 hours to heat up my 4" Dash with 3 ProCap plates. Then around 4.5 hours at 4.4 kW with dephlegmator on pretty strong (cold Scottish water) I get a good steady flow through my cuts taking rum hearts around 81% from a 9.5% sugar wash. So I'm running with vapour speed around 15.9"/sec, right? I figure I want to replicate this as I scale up.

Thinking going with a 380L boiler with around 15 kW heating, this should take around 3 hours to heat up. I'm thinking a 6" Dash with 3 ProCap plates would take around 9.8 kW to run double flow rate through the cuts with the same vapour speed as my old rig at 15.9"/sec, right? But since the boiler volume is nearly 4 times bigger the time from heat up to end of tails will now be double the old rig, around 8 to 10 hours. Is my thinking right?

If I want to run more plates, say 6 or 8, would the total 15 kW give me enough headroom on power to push up the extra plates and maintain the product flow rate?

How many plates would I need on the new rig to make vodka, would 8 be enough and would the headroom of 15 kW of power be enough to push this through the plates with a reasonable product flow rate?

Plates and power, the balance of life!