Huff the magic dragon lived in a still....

[Alchemist75]

This one has been touched upon in passing I believe but upon searching I found little on the topic: dephlegmator huffing. I had assumed for a long while there that the variation I was seeing in my take off at high reflux was an artifact of running a narrow column and was just how things were....until recently. I stopped using a separate stripping riser and now I only use my column set up either packed or un packed, saves having to switch out lines and whatnot. I decided to fiddle with the reflux on a stripping run, trying to compress the nose end a bit by holding high reflux for the first couple hundred ml (I make cuts on both runs). While I'm not clear how well it actually compressed the fores I did observe variation in take off speed at high reflux, unpacked column so no surging or flooding possible. Slow drip speeds up to a flow then back to a drip. I knew at once that the likely scenario was shock cooling in the dephleg. I did a rum spirit run a couple weeks back and I had scrubby stuffed up in the RC and got a nice steady drip at high reflux but I didn't make the connection until last night, I assumed it was a fluke. I had always wondered about this one, if the dephlegmator could huff, not much was written on it though and it seemed a non issue. Well, maybe you've already beat me to it on this one but if you haven't, it's definitely a thing. Right, put scrubby in the condenser, always, moral of the story.

[Saltbush Bill]

dephlegmator huffing

Ive never seen a Dephleglamator huffing ..or heard of one for that matter.
The only time that my plated column will surge is if its being over powered and the plates are flooding. If that is happening its a sure sign that I'm running it wrong.

[Alchemist75]

Right, I'd never heard of it either though I wondered. Now I can actually say it's a real thing. Makes sense, why wouldn't it occur, same principals apply. It may be, Bill, that it's a phenomenon unique to CM designs which utilize a tube and shell dephleg. The vapor hits the coolant and collapses forming the vacuum. It wouldn't be visible in VM, LM or Boka due to the way take off is regulated. The fact that having a scrubby in there fixed it drives the nail in the coffin. Might lend credence to the idea that counterflow on a dephleg is just as important as with the PC.
One thing I noticed is that the cyclic surging it caused didn't seem to affect the proof of the take off, only the rate at which it was flowing. That's why it wasn't a big deal, just a head scratcher. I'm learning all sorts of new things running my strips through an unpacked column.

[Alchemist75]

This also goes back to my thread on CM condenser efficiency, having a more under powered design might correct the problem as well. Either that or having heated coolant flow to the dephleg, that might fix it too.....

[DetroitDIY]

I had huffing on my old liebig on a pot still. Solved it with a copper coil down the center who's OD was the the same as the ID of the liebig vapor tube (thanks to good direction from other members here).

On my flute I now run, I was not getting effective knock down of vapors out the PC. (5) 1/2" tubes in a... 14"? PC. Did the same trick as I did with the liebig to increase the turbulence and surface area, worked like a charm.

Made me wonder if I should do the same for my dephleg, but I haven't gotten around to doing it yet. Stands to reason that it should increase the efficiency there too.

Regarding the huffing, I'm thinking a narrow PC (my liebig) is most prone to it due to the long length with limited cross section. An effectively wider PC... think (cross sectional vapor path) / (vapor tube length) should make the PC more able to draw in "make-up" air to counter the condensation. Not blasting the boiler power and overwhelming the shotgun PC with too much vapor mass flow I'm sure is helping me too.

But in both of those PC scenarios, vapor is condensing and leaving the boundary of the system (our still). In the case of the dephleg, there is initially more expansion from liquid to vapor than condensation, which should push some volume of gas (air) out of the still. Then when you're in full reflux, effectively no mass is crossing the boundary of the system, so there should be no huffing possible. Then in partial reflux (imagine now a smaller boundary around the still where product leaving the dephleg is leaving the system), there is some take off and some condensing going on in the dephleg. Given that a dephleg typically has a lot of cross sectional vapor path and a short length (an effectively SUPER wide condenser relative to the two PC options I mentioned above), and it's in partial reflux, huffing in the dephleg seems very unlikely.

As the dephleg approaches less reflux and more pass through, then there's less condenstaion, and less make-up air required. That section starts to look like any other segment of your still between the boiler and the PC, where there's relaively little condensation and thus no need for make-up air and thus no huffing.

Just thinking it through, it makes sense to me that people don't experience huffing a the dephleg. Not unless it's grossly mis-designed for how the still is being run. I'm NOT trying to say that your design is "mis-designed" Alchemist, but just that I can't rationalize that the fluctuation is a function of huffing. It makes me wonder if Saltbush is more on the right track, and it's a function of how you're running it.

[Kareltje]

I liked the title.
Reminded me of old days.

[Alchemist75]

Well, it's not flooding, I'm sure of that. I've deliberately flooded the column when I first built it to test heat input parameters. Flooding is always preceded by an audible gurgling followed up with a big gush of lower proof hot liquid. I initially attributed it to surging lower down in the column but when I observed it exhibiting that same behavior unpacked I knew that couldn't be it either. The dephleg itself isn't flooding, putting scrubby up there would have exacerbated the problem if it was, it fixed it instead. The id of the dephleg is 3/4" which I don't believe is too narrow for any condenser though perhaps narrower than some. Bumping isn't the issue, if I dial back the reflux it holds a steady stream no problems. So unless there's some other reason I'm missing here it surely looks consistent with huffing, why else would the scrubby keep it from happening? I'm not challenging what you say here but if it isn't huffing then what is it?

[Alchemist75]

Well, I'm doing a spirit run tonight, we'll see if I can get a repeat performance. Repeatable results are what we like.

[Alchemist75]

So I popped the scrubby in the condenser and started the run. While it seemed diminished there was still take off cycling happening.
Let it never be said that I won't admit to my own tomfoolery:
I got the brilliant idea to pull out one of my super sensitive lab thermometers and monitor the temperature of the water leaving the dephlegmator.......yeah, temps bouncing around by as much as 5 degrees. DOH! Will be getting a pressure regulator shortly. Goodnight and thank you for participating in my duhtastical rant. Ahem.

[kimbodious]

good catch!

[still_stirrin]

A water reservoir and pump will help stabilize the coolant variations for you.

For my small (16 quart) gin still, I use a sink basin and a small pond pump. When the reservoir gets warm, I drain and replace a portion of the water. Throughout a run, I usually have to “turn over” the water 2 or 3 times.

So, help me understand your “dephlegmator”...is it a single (concentric) vapor tube inside a water jacket shell? Or, do you have a series of tubes in the shell? Also, how long is the vapor tube inside the jacket?

These questions will tell me a lot about your design and whether or not it is appropriate for the purpose you’re intending it for.

As DDiy suggested, multiple vapor tubes inside a water jacket (shotgun style) are more effective for duty as a CM reflux condenser. The larger vapor flow area helps maintain proper vapor velocity (without acceleration of the mass) while extracting heat. So, a dephleg should not be longer than 10 pipe diameters (of the vapor tube).

But, the length of the vapor tubes becomes a design concern too as too long will not only cause flow loss (resistance) but also can lead to complete vapor collapse, or sometimes called “shock cooling”. That is, or can equate to “huffing”, especially if in a product condenser. I think you expressed this as “overly efficient deflag” in your other thread.

Proper design of a still is more than just throwing some parts together. Each component has a specific function and as such, a design performance objective. Consideration of the requirements will help optimize the design for the purpose intended. I know...this makes me sound like an engineer...but, guess what...I AM an engineer, with years of experience, including process flow systems design. So, design matters. A properly designed and built still will run easy and stable.
ss

[StillerBoy]

Proper design of a still is more than just throwing some parts together. Each component has a specific function and as such, a design performance objective. Consideration of the requirements will help optimize the design for the purpose intended. So, design matters. A properly designed and built still will run easy and stable.

In total agreement here..

That is why commercially build still are not the best to buy.. they lac the design finesse to make them operate efficiently.. yah they are called still, but are they efficient.. not from what I've seen in operation..

Mars

[Alchemist75]

Still: so the dephlegmator is a 4.75" tube and shell style with an array of through condensers built into it so it has internal cooling surface area as well as the cooling of the walls. The vapor path is 3/4" id with a 1" id jacket surrounding it, basically a "west" style liebig with the addition of multiple internal cold fingers. It'll knock down everything I can throw at it. It behaves exactly the same as my old external coil design except it's about half the length. Despite the cycling of take off rate I can count on the design to pull about 94% at 550 ml an hour (I finally clocked it last night). That's slow but it is a 1" x 17" packed section after all. It exhibits that cycling either packed or un packed at higher reflux though i believe I may have diagnosed the "problem", if such it could be called, last night.

[Alchemist75]

And it does appear that stuffing scrubby in it does reduce the cycling affect, don't know why but it helps.

[still_stirrin]

... The vapor path is 3/4" id with a 1" id jacket surrounding it... with the addition of multiple internal cold fingers... but it is a 1" x 17" packed section after all...

OK, so it’s a 4-1/2” long concentric 1” over 3/4” with several cross-tubes. And your column is 1” ID with scrubby packing.

I think that your cross-tubes complicate the concentric’s operation. Perhaps if you disconnect all of the cross-tubes (except maybe just one), then you may find a more stable throughput.

Part of your problem is the vapor velocity speeds up through the section causing minimal condensation so you increase the coolant flow causing rapid condensation (vapor collapse). Reducing the cross-tube effectiveness will keep the vapor from condensing so quickly. But, the vapor velocity is fast because of the small vapor tube. If you could resize the condenser with 1-1/2” over 1”, you would have more uniform velocities through the column and the reflux condenser. In fact, if you could increase the column diameter and the vapor tube diameter as well, you’ll get a much more controllable process.

I always try to keep the vapor flows steady as possible and that means a uniform cross-sectional area in the vapor flow path to the extent possible. Whenever you narrow the path, the vapor must speed up and that reduces dwell time... the time needed for the condenser to do its magic.

So, without a redesign/rebuild of your system...try eliminating the cross-tubes in the condenser section. And perhaps add a small copper scrubby stuffed into the condenser section (keep it very loose) to help cooling efficiency, in lieu of the cross tubes.
ss

FYI - the scrubby in a condenser increases the surface area that the vapors can condense onto. The scrubby conducts heat into the vapor tube walls, which then transfers heat to the coolant. As the vapors contact the scrubby and gives up its heat, once condensation begins the condensate continues to pull more heat from the passing vapors. The scrubby surface area acts as the catalyst to the condensation.

[still_stirrin]

And your small diameter still examplifies the control instability due to the dynamics of its size. Flow rates vary and the small scale features make small variances significant with respect to the processes.

As an example, 4” diameter stills are very stable to operate compared to a 2” still, even if the designs are equivalent. Likeswise, commercial stills 6” or larger are comparatively more stable with operations and product quality. It’s all about the “economy of scale”.
ss

[Alchemist75]

So in a sense I was right about the huffing. Funny. The cross tubes cannot be disconnected, they are literally inside the condenser. The design was taken from here:
http://www.artisan-distiller.net/phpBB3 ... a+cold+tip" onclick="window.open(this.href);return false;" rel="nofollow
Kina's "cold tip", I'd post the diagram he drew up but it doesn't belong to me so I'll just post the link. He didn't invent it but that's where I got the idea. I was trying to reduce needed condenser length so I figured I'd try the more compact design featuring greater surface area. It works alright! The next build will be 2", I have lots of scrap that I could use to do another 1" mod but I can't be bothered with trying to upgrade something this narrow. The whole thing operates in a narrow range, I can control reflux pretty well but obviously power input has to be kept low. It's an interesting experiment in novel design, it kinda blows the top off the idea that a 1" can't be done not to mention the HETP it gets at such a short hight. I wonder if I can pull that same effect on a 2"......
Anyway, the fluctuations in temperature coming from the dephlegmator told me at least one aspect of what was going on though you seemed to confirm my initial diagnosis to some degree. Thanks for adding your insight Still, you always do a great job of breaking it down in plain English.

[still_stirrin]

... I wonder if I can pull that same effect on a 2"...

So, if you want a CM in 2", try building a dephlegmator (shotgun) with four or five 1/2" ID vapor tubes. Make the condenser 5 to 6" long with the vapor tubes 4 to 5" long. You won't need baffles in the water jacket if you put the inlet and outlet on opposite sides of the outer tube. Make the water inlet the top port and put an external pipe on the outlet to route the water flow back to the top of the condenser section, and if you make it higher than the inlet, the condenser will always be full of water ... gravity (Bernoulli's law) ... you know. TC the top and bottom and you can stick this in your 2" column. You should be able to pull 180 proof at least from this condenser depending on the column height and packing selection.

If you need more RC knockdown power, flatten some copper, twist a spiral and stick it into the vapor tubes. The vanes will increase conduction surface and improve condensation power. You can solder a small piece of Romex solid copper wire on the strips to hold them in the shotgun and you can easily remove them for cleaning and/or storage.

I believe you'll appreciate the 2" still. It'll be so much more "user friendly" without all the control problems the small still has.
ss

[DetroitDIY]

If you need more RC knockdown power, flatten some copper, twist a spiral and stick it into the vapor tubes. The vanes will increase conduction surface and improve condensation power. You can solder a small piece of Romex solid copper wire on the strips to hold them in the shotgun and you can easily remove them for cleaning and/or storage.
ss

Or you can try skipping the flattened copper spiral and go just to the Romex helix in the vapor tube. Here's the quick fix I did along that line for my 5 tube shotgun PC. Go to the posting from Sept. 29. It worked well for me. Was getting uncondensed vapors out the PC before this, no problems at any power input after this.

viewtopic.php?f=17&t=65335&start=150

[Alchemist75]

Appreciate it is an understatement sir. It's difficult to convey the Glee I experienced when I went from running two gallons of wash using a lab boiler in a sand bath painfully, paifully slow, having to contrive a method of getting copper in the path with all ground glass connections and just getting ok product to being able to strip 4 gallons in 45 min and then being able to produce 3 times the amount at very good quality on a spirit run that took less time! Now that was a run on sentence but I think you get my point, the current gear I run was about 10 steps forward. Glad I learned to solder copper and steel. A 2" "hybrid" that can pull 93-94% in a quarter of the time with good cuts and double the volume will be better than good in my world. It's a progression from highest difficulty to the lowest.

[Alchemist75]

Awwww, the drip rate cycles a bit at high reflux? Helluva lot better than doing a stripping run with a ceiling speed of a liter per 4 hours lol.