Shotgun & deph. tube & baffle design guidelines

[zapata]

* Edit Too Long/Not Gonna Read Summary:
IF you want to optimize a shotgun design:
1. Space tubes 1.25 - 1.5x their diameter center to center
2. Use coolant baffles
3. Baffles should be equally spaced. Minimum spacing 50mm; max spacing equal to shell diameter.
4. Ideal baffle cut is easily calculated based on baffle spacing and shell ID.
EG for a 2" shell: equally space (4) 1/2" nom tubes or (7) 3/8" nom tubes. Place a baffle every 2". Ideal baffle shape is a disk with 37% sliced off, ie mark 18.5mm from edge of a 50mm disk and slice it off. (Changing one of these paramaters may call for changes to others)

I just remembered I worked all this out a few years ago. I shared it elsewhere, but it may see a bigger audience here. Apologies if this info is already here.

So I was designing a shotgun, and wondered if I should use baffles, how many, where to place them, how much of the flow should they block etc. Also how to layout the inner pipes etc. Ended up not being too hard to figure out, at least it wasn't when Harry's library was still up. Everyone raise a toast to the memory of tastylime and all the work Harry has done for us!

I've been accused of over complicating many things. These are industry and engineering guidelines. Use them if you wish. But it does make a difference. At the time I was researching this I used a rather advanced engineer's heat exchange calculator to compare a non-baffled shotgun like you have seen many times to one with proper engineering. The engineered one could handle a couple kw more than the standard one. My opinion is if I was gonna put tubes and baffles in a shotgun, I wanted to know exactly where they SHOULD go. Here's the answer.

All the following comes from The Heat Exchanger Design Handbook edited by G. F. Hewitt.

Reminder, these designs are known as shell and tube heat exchangers, the outside bigger pipe is the shell, the inside pipes are tubes.

Tube layout
Pitch is the measurement from each tube center to center.
Pitch:Tube OD ratio should be 1.25 - 1.5
example, for .5" OD tube use .625" - .75" pitch
Note: this is a tighter layout than some designs use, more spacing is just a waste of the shell (or ease of fabrication). But this tight spacing lets you fit in more tubes OR use a smaller shell.

Layout
I don't have a concise analysis of this. I used a 60* (* meaning degree) angle between my tubes, because that was the only way I could fit the number of tubes I wanted. I can say that 60* will cause more coolant restriction than say 30*, 45* and 90* layouts.

Baffles

I know from experience that pressure drop through a shotgun is low enough to be of no concern to me, so the following uses a segmental baffle. If you have a weak pump or some other reason to worry about coolant flow, the disc and donut style baffles will cause less restriction. (But somebody else will have to share the research on them!)

Baffle spacing
Baffle spacing must be equal to use these guidelines.
Minimum baffle spacing is .2 ID of shell or 50mm whichever is greater.
Maximum baffle spacing is ID of shell.
Eg: for a 2" shell, space your baffles 2" apart.

Baffle cut

This shows a specific recomendation of the ratio of baffle cut to the ratio of baffle spacing:shell diameter. Use the SBC line, since your coolant will not be changing phase.

Use graph and ratio of baffle spacing:shell ID (Lbc:Ds in chart) to get percentage (Bc)
Use formula Bc = (Lbch/Ds)100 to determine how far from baffle edge to cut off your slice.

I'm using a 2" shell shotgun, so I calculated a 50mm Lbc for a 50mm Ds, giving me a Lbc:Ds = 1.0
I read that as recommended Baffle cut of 37% as a percentage of the shell diameter.

Extrapolating that to Lbch (length of baffle cut height, in mm; aka, how far do you measure in to cut off a portion of that baffle plate)
rounding the Ds to 50mm, and shooting for 37%
Bc = (Lbch/Ds)100
37 = (Lbch/50mm)100
18.5 mm = Lbch
So make your 50mm baffle plate and slice off 18-19mm.
Like this drawing BUT here I read the chart wrong, and used the CV area for a 42% Bc = 21mm! (I just don't want to make a new drawing and I lost the correct one in a HD crash). Besides, I cut it with jigsaw so sub mm accuracy wasnt really possible after all!

There ya go. The shotgun I designed (2" shell, 7 x 3/8" nominal tubes, 60* tube pattern, 2" baffle spacing, 37% baffle cut) worked great, handles all I've asked of it. Of course I didnt actually build a non-baffled or randomly baffled one to compare it to.
I can't praise The Heat Exchanger Design Handbook enough. Couple thousand pages of everything you never knew you wanted to know.

[Saltbush Bill]

I just have to ask why ? Why complicate something so simple to such a degree. We are trying to cool hobby size stills here not nuclear reactors. The unbaffled condensers people are building for these stills will already knock down much more vapor than needed without going through all of the rigmarole of including baffles.

[LWTCS]

Have to agree with Bill.
By definition the dephlegmator is a "partial" condenser.
You are going to run across dozens of previously split hairs and hairs yet to be split. And that's just on the coolant consumption side of the equation. Then there is the behavioral side that may or may not affect the quality / profile of your finished product.
Most here build according to their budget and available time. Then learn to run their apparatus as optimal as their point of view will allow.

On the other hand,,if you're enjoying the journey,,,, WTG and good luck with your research.

[zapata]

Sigh. Ok. Nevermind. I should've just slapped something together that looks like a random shotty on the internet. Maybe put a baffle or 3 in there somewhere, some people do that, right? Just guess how big to make that baffle because there can't be an easy way to predict what works best. How should I space my tubes? How many x size tubes should go in the shell?

Pop quiz, if you wanted a shotgun, say for a beer stripper, how would you make all these decisions? You HAVE to make the decisions, or copy someone elses. But even then, I'd come on here, see a shotty with wide spaced tubes and one with tight spacing, so which do I copy? Do I copy the no baffled shotty, the 1 baffle, or the 3 baffle? They're all on here.

How is this any different than say wrapping a liebig with copper wire for turbulence or crimping the inner tube? Or calculating the right sized VM takeoff, or calculating vapor speed in a column? Sure you don't HAVE to do any of that stuff, but some people do and if you have questions about any of it there are actual answers.

[zapata]

Maybe I'm just too wordy and make things seem complicated? How 'bout this.

IF you want to optimize your shotgun:
1. Space tubes 1.25 - 1.5x their diameter
2. Use coolant baffles
3. Baffles should be equally spaced. Minimum spacing 50mm; max spacing equal to shell diameter.
4. There is a 30 second calculation to determine ideal baffle size.

[ShineonCrazyDiamond]

I love my PC shotgun with baffles. Both the ones I have include baffles. 2", 21" long, (4) 1/2" tubes. One I made, one I had made for me. Stripping 5500w, no more than a trickle, slightly more when running my CCVM in tandem.

Did I need the baffles in this application? No, I could have done without and used a little more water. But one day when I go up to 9k watts our so, guess who doesn't have to wonder if the shotgun will have to be remade?

I believe a PC shotgun, in particular, should always be designed the most efficient as possible. One condenser to liquefy them all . It was literally no real extra work. Instead of 2 end caps, I made 4, and cut 2 of them right in half. Slide them on, alternating, and spot tacked them on. Easiest part of the build, really, that was the biggest bang for the buck in ROI efficiency.

I didn't over think the details of the most optimal design, though. Picked a length, had an extra 21" piece lying around. Drilling out the plates, I just doubled my cuts. No biggy. Figured 4 baffles were better than none, whether I needed them or not.

Each their own, glad you got something working for you. That shotty will do anything you tell it to.

[LWTCS]

So the baffles with only 4 thru tubes doesn't sound nearly as efficient as simply using a smaller diameter thru tube and installing more tubes (on the product shot gun).....See what I mean? But that is more of a pain in the ass to build, or more specifically may exceed some folks skill level and impair their ability to give themselves the instant gratification that so many newbies seem to crave.

Or can pack the larger thru tubes with scrubber material to promote more surface area contact.

Efficiency has a couple few faces with respect to distilling.

Lots of guys that have been around for a while have done it more than once. That's how they figured out what works best and also created their own point of view.

[zapata]

SCD, obviously you know your condenser performs, she sounds like a beast. So lets take this proven condenser as a case study and see if the formulas agree.
Tubes
You say 1/2". Is that true 1/2" OD (sold as 3/8" nominal)? Or can I assume they're 1/2" nominal (true OD 5/8")? If so, then 4 tubes is the optimal number. If they are equidistant from each other and the shell you have a Dt:pitch ratio of about 1.25. Perfect. You could have rammed one more in there (depending on shell ID; DWV and type M would hold another, L and K wouldnt) BUT it would have messed up the flow dynamics. Analysis: you nailed it.

You probably have a 90* layout pattern, which I won't comment on because I don't think you can fit 4 tubes in any other pattern, and tbh I don't remember the differences in geometry because mine were fixed as well.
Baffle # and spacing
4 baffles is clearly better than no baffles. Assuming they are at least roughly evenly spaced then they are a bit further than optimum, so ideally you would have had 3 more. Analysis: not perfect, but much better than terrible. Let's say 4/7=57% but the improvement is not linear so lets curve that to a modest 70% optimized.
Baffle cut
Given the spacing and number of plates you chose, 50% plates are in the neigborhood of the ideal cut. I had to guesstimate that because the graph doesn't exceed a spacing:Ds ratio of 1 while your ratio is about 1.75. Analysis: my eye says you nailed it.

Sounds like your experience corroborates the formulas fairly well. You optimized 2/3 and moderately optimized the 3rd. It's possible to quantify performance potential if truly optimized , but the heat exchange simulator is laborious and there is no real point since you're happy already. But if we say you aced 2/3, and 70% optimized the 3rd your made up score is 90% optimal.

I find it insightful to be able to analyse a proven piece of kit and understand WHY it works well, and how to replicate or possibly slightly improve on it.

[zapata]

LWTCS,
For sure more smaller tubes would be "more efficient", up to the point they are dangerously small. Just going on hobbiest opinions, some feel 1/2" nominal is the smallest that should be used due to vapor speed (not an opinion I share in a PC) or plugging danger in a potential puke. Personally I'm comfortable using 7 x 3/8" nominal even with a puke risk, but I'm not aware of a definitive source for how small is too small. I certainly would not go down to 1/4". In that case even if somewhat arbitrary, SCD's decision to not go smaller than 1/2" nominal is perfectly justifiable and can be optimized.

Going the other direction though we can definitively say 1/2" is better than 3/4". You can only fit 3 tubes instead of 4. Also they would be packed too tight with a Dt:pitch ratio of 1 (ideal 1.25 -1.5). Meaning there would be too little (literally no) space around them to allow for optimal flow. 5/8" nominal is rare but exists, can fit 4 tubes in a 2" shell, but would have a pitch ratio of 1.1. Too tight. So the only options in a 2" shell that are safe and ootimizable are 1/2" and 3/8" nom.

Mesh can be added to any shotgun or not. It definitely increases surface area and efficiency, but can restrict puking and theoretically smear. To use it or not is not within the scope of shell, tube and baffle specs.

[Saltbush Bill]

But that is more of a pain in the ass to build, or more specifically may exceed some folks skill level

I couldnt agree more, this is were I have a problem with threads like this one.
Newbies come along and read this , then start to believe that they cant build a still because of all of the extra complexity. Others with more experience who should know better jump on the band wagon and reinforce the idea that you MUST have baffles. None of this does the hobby any favors ,it just puts newbies off building their own and into the hands of the commercial builders.
I don't care if or how many baffles people choose to run, I just don't like the idea that people might begin to think that you MUST have them.
Id personally rather encourage people than discourage them.
For what its worth my own un baffled product condenser is 2 inch diameter, 17 inches long, and has 5 x ½ inch vapor tubes. On an average spirit run it could be half that size and still do its job.
I have used it for stripping at 15L an hour / 3.9 gal an hour, its capable of much more given more water flow.
How many of us in the hobby distilling world really need a condenser that will produce more than that?

[jedneck]

Mine is 4 tubes of 1/2 id in a 2" shell 19" longs NO baffles. Sofar i hit it with 6500 watts and no problem knockin it down. Shouldnt be a problem at 9000 after new controller completed.

[zapata]

Saltbush, I hear ya. Sounds like this thread ain't for you. It wasn't started by or targeted to noobs either. Every hobby has different levels of involvement, which isn't necessarily a reflection of experience or quality.

I really don't see how it's bad for the hobby? The reason I posted this now was a thread yesterday where someone else said "build a 2" shotgun, and put some baffles in it, at least 3". Not calling him out (he's not wrong) but the at least 3 sounds like a shot in the dark. I'm just adding numbers and design theory to a conversation that has been part of the hobby for years. I started a new thread because the comment that prompted it was off topic enough, I would've posted a link to a thread on the subject, but all I could find were dozens of questions with the best answers being "I don't know, but this is what I did". Search the site for "baffle" "shotgun tube spacing" or "shotgun questions" you'll find them too.

As a newb my eyes glazed over all the calculations in the Complete Distiller, but I was reassured that there were calculations someone was doing. Over the years I came to intimately know my rigs and thanks to many patient explanations mixed with a solid dose of "no, YOU figure it out", well now I can. I know a lot of the hows and whys, and have a pretty good idea how to find the ones I don't. I LIKE that part of the hobby.

You don't. Thats cool. I don't like grappa, even a 25 Euro glass in Milan tastes like UJSSM fores mixed with Cuervo Tails. But maybe you like it? Thats cool, theres room for nerds and craftsmen, grappa and single malt. I bet for every newb you think would be turned off, theres another just like me intrigued and inspired that thinks, I may not understand that, but I want to, and I will.

You have no interest in optimizing shotguns. Noted. Again. But I ain't hurting the hobby. The fact that I am a hobbiest, learning and applying this as a hobby, and sharing it on a hobby board, where other people are doing and talking about the same shit for years shows this IS the hobby for some of us. I won't say casting bad-for-the-hobby stones is bad for the hobby, but please reconsider that approach. A simple "mine works just fine without any of this" works just fine it kinda misses the whole notion of the thread but a valid data point and opinion nontheless.

[zapata]

Jed, good to see another comfirmation of a successful shotty using the tube sizing and spacing suggested by the the book. Clearly nothing wrong with your shotty. In the context of this thread though, if it had an optimum number and size of baffles it could be substantially lighter and shorter, or could handle sunstantially more power when your next next controller is built....by North Korea

I do want to make it clear, I'm not saying anybody's successful shotties are flawed. Just that there are guidelines to easily maximise a given condensers heat exchange capability. Whatever shotgun someone may build, you can maximize its potential in 5 minutes with $5.

[LWTCS]

Enough 1/4" thru tubes in a 2" shell can offer more available open area as well as surface area than a mere four 1/2" thru tubes in a 2" shell.

Do you need that at the hobby scale? Prolly not.

Btw, not bashing at all. Good to see folks push forward with their education. It's just for a few of us, some of this is a bit of re inventing the wheel. Please do continue. Any perceived counter argument is merely an attempt at making sure that folks understand the breadth of the subject matter.

It's like golf. The game revolves around whacking that little white ball with a stick. Very simple concept. Hit the ball into the hole. Super simple.
But it is also far more complicated when you start to install optimization. Far more nuanced.
So it can be as basic,,or as complicated as you want it to be.

[jedneck]

I have no problems with trying to improve efficiency. But not evryone is handy with tools, many never soldered before startin this hobby. Build what your skills say you can and next build get more complex/harder.

Remember if your not learnin you might as well be dead

[Twisted Brick]

Zapata,

I can’t thank you enough for posting your analysis. It was actually your shotgun coolant baffle design on MD that I first read through, and when planning my first build was exactly what I was looking for.

I'm a complete novice, and for my proposed 3” pot still, I knew I wanted to build a shotgun. Your thread easily had the most comprehensive analysis applicable to the hobbyist that I found on any of the distilling sites. I also knew I wanted a design that would not require upgrading down the road.

Dimensions of my shotgun: 2” x 21” copper shell with 4 x 1/2” (nominal) vapor tubes. 7 baffles spaced 2.5” apart with 38% baffle cuts. With the help of endplate/baffle dies, the build came together easily, especially for someone new to soldering. From your 'blueprint", I feel I was able to optimize all of the critical design features and ended up with a condenser that should serve me well for a long time.

My shotgun build thread:

viewtopic.php?f=87&t=67972

I don’t have anything (except reports on HD) to compare it to, but in operation, the condenser appears to be a workhorse. My first ever 7gal UJSSM strip went 1hr 40min (60 ABV down to 20ABV) using a 30k btu burner wide-open, and the condenser never hiccuped or gave up a wisp of vapor. My 30 gal reservoir temp rose from 73F to 118F over this time, which I think translates to condenser efficiency and a need to either increase my reservoir size or figure out a way to cool the condenser exit water.

This entire site is about optimization of equipment and processes and I applaud you for elevating the level of understanding that will likely become standard build specs for anyone inclined to build a shotgun. I’m pleased to have taken advantage of your efforts, because whenever I look back and realize I could’ve accomplished something I was originally too scared to do, I get nothing but regret.

Thanks again. Because of you I was able to build my ideal shotgun my first time out and I'm sure there will be others.

[zapata]

I'd be perfectly comfortable with an ending res temp of 118. No need for cooling or bigger res unless you're gonna do bigger runs or back to back strips. 1 thing for me is I usually pump from the bottom, return to the top of the res so I get a stratified reservoir. So is your 118 the top layer with some cooler on bottom, or is it fairly well mixed?

[Twisted Brick]

Now that you mention it,118F on the top. Thanks for the re-assurance. I'll just stick with it as is and monitor it when I move up to a 10gal batch.

I assume that with a lower flame during a spirit run, it'll be fine.

[ecir54]

It was actually your shotgun coolant baffle design on MD that I first read through, and when planning my first build was exactly what I was looking for.

I would like to read this. Is it available out there?

[Twisted Brick]

I haven't been able to locate it the last few years either. I think it went down with the Modern Distiller site. FWIW, the info presented in this thread is about 70-75% of what was contained in the shotgun coolant baffle design post on MD.

[ecir54]

I haven't been able to locate it the last few years either. I think it went down with the Modern Distiller site. FWIW, the info presented in this thread is about 70-75% of what was contained in the shotgun coolant baffle design post on MD.

think I found it but page wont load, damn.

https://web.archive.org/web/20131016150 ... m.php?f=47

[Hebden]

4. Ideal baffle cut is easily calculated based on baffle spacing and shell ID.
EG for a 2" shell: equally space (4) 1/2" nom tubes or (7) 3/8" nom tubes.

Can I just ask, when the term "nominal" is used, would I be correct in thinking "nominal" simply means the ID of tube?

[Ben]

4. Ideal baffle cut is easily calculated based on baffle spacing and shell ID.
EG for a 2" shell: equally space (4) 1/2" nom tubes or (7) 3/8" nom tubes.

Can I just ask, when the term "nominal" is used, would I be correct in thinking "nominal" simply means the ID of tube?

Correct.

[zapata]

Can I just ask, when the term "nominal" is used, would I be correct in thinking "nominal" simply means the ID of tube?

No. Nom and nominal mean the name we say that is close to one of it's dimensions.
For example. If you go into a hardware store in America and ask for "half inch copper pipe". That is nominal half inch. The label on the shelf says 1/2 inch. We call it 1/2 inch. But nothing about it is .5 inches. The outside diameter is .625 inches. Depending on wall thickness (aka schedule) the inside diameter is either 0.527 (schedule K), 0.545 (Schedule L) or 0.569 (schedule M).

Copper is standardized to have the same OD regardless of wall thickness, and for whatever reason the OD is 1/8" (.125") larger than the nominal size. Which can cause a lot of confusion because for example a nominal 3/8" pipe has an OD of 1/2". In normal usage like plumbing it doesn't matter much, as long as you buy pipes and fittings that are called the same thing, then they will fit together. But since we are designing and fabricating custom gear you need to know the difference. If you look at an endplate design that fits 7 tubes in a 2" shell, the holes in the end plates will be .5" diameter, but the pipe you buy to fill those holes is called Nominal 3/8".

[Twisted Brick]

Can I just ask, when the term "nominal" is used, would I be correct in thinking "nominal" simply means the ID of tube?

No. Nom and nominal mean the name we say that is close to one of it's dimensions.
For example. If you go into a hardware store in America and ask for "half inch copper pipe". That is nominal half inch. The label on the shelf says 1/2 inch. We call it 1/2 inch. But nothing about it is .5 inches. The outside diameter is .625 inches. Depending on wall thickness (aka schedule) the inside diameter is either 0.527 (schedule K), 0.545 (Schedule L) or 0.569 (schedule M).

Kinda like going into any lumber department in America looking for a 'two by four".

[Ben]

Can I just ask, when the term "nominal" is used, would I be correct in thinking "nominal" simply means the ID of tube?

No. Nom and nominal mean the name we say that is close to one of it's dimensions.

Copper is standardized to have the same OD regardless of wall thickness,

This is correct. Ignore my above post. I had a moment of idiot.