StillDragon® Community Forum


Be part of our community & join our international next generation forum now!

In this Discussion

Super Dephlegmator Build Issues

edited November 2014 in General

I'm building something like a Super Dephlegmator 4" SS. Firstly I wanted to make 2 inputs and 2 outputs for water. But then I thought about building 1 input and 3 outputs for water. This (as I'm imaging it) would allow me to have 3 different water levels in the dephlegmator.
For small powers, first lower output, for medium second, and third on the top for max power.
Am I thinking right way, or it will not behave like i imagine?




  • Hi Woxy,

    I am unclear if anyone has built a dephlegmator as your diagram illustrates?

    StillDragon North America - Your StillDragon® Distributor for North America

  • Seems like it should work as long as the input doesn't over power any of the individual outlets. It would probably be easier just to throttle the flow rate if you want less reflux for lower powers though...

  • ONE OR TWO INLETS ONE OR TWO OUTLETS and a needle valve to control the water flow and no need for the other outlets... can't see a need for any additional flow or water levels...especially on just a 4" column


  • Since you are investing in the stainless in the first place you may as well build it so that it has the highest flexibility possible. Built large enough to handle your maximum power. 1 inlet and 1 outlet plus sufficient number and sized baffles. Combine that with the ability to control the water flow down to your lowest power requirements. That's what I'd be doing.

  • edited November 2014

    @Smaug said: Hi Woxy,

    I am unclear if anyone has built a dephlegmator as your diagram illustrates?

    Hi Smaug, it is just my concept, I haven't found any one that built it.

    @brewsmith , @FullySilenced , @TheMechWarrior,

    I'm planning to use 37 x12mm OD size, about 30cm long tubes in 114,3mm OD tube. It will give me about 4182,48 mm2 of pipes to 9498,5 of 110,3mm ID tube. So it will be 44% for pipes to 56% for water. Maybe it should be more difference, i.e. 30% for pipes to 70% for water?

    With the 3 levels idea, I was hoping to have less water in the dephlegmator, an with less water comes less thermal capacity, and that causes quicker temperatures change.

    But as all You say, it's better to have max 2 in/out, use full water capacity, and just regulate water throughput (this type I'm using in a pot still). Thermal capacity is not a problem ?

    The dephlemgator is not finished, so I can change number of pipes in it, can You veryfiy my calculations?


    682 x 632 - 55K
  • I'll have a look at your design when I get a chance, for the moment here's a few of my thoughts on shell and tube design.

    I'm not suggesting my shitty drawings are Gospel or anything close to it but it does give you a clear image of what goes on in MY mind when I look at a shell and tube heat exchanger.

    I've had a couple of these made:


    Here's some thoughts about why baffles are good if you are planning on running anything less than full water flow:


    1221 x 546 - 125K
    800 x 740 - 125K
  • Oh my, I've never seen so many pipes packed so tightly in a dephlegmator before.
    Not saying its wrong, just different.
    A dephlegmator is usually built to be a little inefficient (fewer and bigger pipes) vs very efficient (many small pipes) because, until relatively recently, it was never meant to put a column into total reflux. But in the last few years the trend has been to reflux the column at 100% to "equalize" and stack the fores and heads at the top of the column before allowing product to be taken. Get them out of the boiler and into the column, so to speak.
    I find your build very interesting @woxy and will keep an eye on it as you develop and test it out.

    At first look it seems over built but the very worst than can possibly happen is it costs you a few dollars more to build and it will certainly conserve water. The water path between the pipes will be tight so some pipes could get warmer than others. I'd lose a few pipes just to let the water circulate better.

    Good stuff. That layout would also make a very nice product condenser for a bigger still. You draw that? It looks like @LawOfOhm's work.

    Are you thinking to use 3 valves on the outs? Open the top valve for full reflux, middle for removing fores and heads and the lower valve for hearts?
    A dephlegmator like this might be useful for those of us that are plagued by warm water in the summer and ice cold water in the winter. With warmish water you need a huge water flow or a bigger condenser. With really cold input water a smaller condenser is preferred and a precision needle valve is a must.
    I've found really cold input water to be the hardest to work with. The tiniest movement of the needle valve and I get too big of an adjustment and Plan B is called on where I try to balance hot exit water from the product condenser with ice water from the mains to feed the dephlegmator.

    Now I have an @olddog controller so that is not a problem for me anymore but many folks don't have such convenience and your concept could offer a viable alternative. At the very least, it is a novel concept and I hope that you pursue it.

  • @TheMechWarrior, WoW :) thanks for such nice drawings, I'll study it later carefully :)

    @Lloyd, Yes, I'm thinking of use 3 output valves. Now just considering little larger one input with 3 smaller outputs, but it's just loose thought. And Yes, this is my drawing, I'm not trying just to copy others work. For acknowledge attached file.

    I'll study @TheMechWarrior drawings and rethink my build, and let You know what have came out of all this :)

  • edited November 2014

    By altering the water level in the dephlegmator to control it, the variable being controlled is efficiency. By using a lower water level, you are forcing the dephlegmator to be less efficient.

    What you might find, is that on the lowest water level setting, you actually need to use a higher flow rate than you need on a higher water level setting, to maintain the same level of reflux. Which means your overall cooling system will be less efficient, or you may need to use more cooling water than otherwise necessary.

    Ideally, you use the largest heat exchanger you can fit or afford, and it will allow you to use an overall lower flow rate, at a higher delta T.

    The idea is interesting, but if you are going to spend the money to do something unique? Baffles.

    Baffles, on the other hand, are an awesome idea. What Mechwarrior says? Yeah. That!

    tmw - With regard to tube spacing in the HX, be sure that your tubing is arranged in such a way that you can insert a temperature probe IN the HX. Oh, and that tube spacing is pretty dense, baffles would absolutely be required, and I suspect you would need a specialty manufacturing shop to be able to pull that off. Those welds are TIGHT - two welds within what looks to be 1/8th at the middle. Although if you can fab as well as you can put drawings together, you'll clearly have no problem. Me? I'd make a whole lot of mess and it would end up with a trip to the scrap yard.

    I suspect you may not need as many baffles as in the top drawing, when the HX is oriented vertically, I suspect due to the temperatures we use, there would be significant convection within the HX body. Yes, there would still be hot spots, but I suspect not nearly as significant as just static flow would lead you to believe. Probably be a bigger problem horizontally, as shown.

  • edited November 2014

    Probably ideal thing would be one long spiral "baffle", but as for me very difficult to make. Something like this, but with the pipes going through the side of spiral. I've seen something like that but from copper. Not remember where I've seen it.


    431 x 210 - 12K
  • Is like trying to reinvent the wheel. Designs. Have been proven in this application for 200 yrs or there abouts.

  • A member called tickle did a dephleg with a outlet half way, from.what I remember it worked well for him, but I don't think he uses it anymore?. From all the testing I've done I find fewer larger tubes better than more smaller tubes.

  • @Lloyd makes a good point about the tube density. Also agree with your really cold inlet water temp comment...which has got me thinking again!

    @grim for that HX shown above it had temp and pressure monitoring on both the product and water side. It was used on a ultrafiltration membrane unit in a biotech plant. Baffles can be relatively easily and cheaply laser cut. Or even cheaper by the hobby builder. I'm with you on the tube spacing for the hobby builder.

    @FullySilenced agree with what you said

    @woxy, I couldn't sleep last night thinking about your idea! It took all my strength NOT to get out of bed at 01:30 and jump on the computer to draw up your plan and why I think it has merit. Your design has roughly the same tube packing as my 1.5m heat exchangers...meaning it will have a HELL of a lot of kick down power (more than you'll ever need on a 4" rig). Downside is that in this particular application you also need the finess of fine control for use as a dephleg to allow you to fine tune your product output. For this you need efficient cooling water flow through your HX as I've illistrated above via baffles...OR...your concept of multiple take-offs to reduce the efficiency of the HX down to a level that allows you the fine control you desire.

    More than one way to a solution. Personally I'd like to see you complete your build and post your thoughts and results along with pics.

    Here's a drawing of what kept me up at 01:30 this morning:


    800 x 565 - 54K
  • I have thought about this plan in the past but the concensus seems to be that it is easier to control the temperature in a full dephlegmator, instead of the flow rate through a 1/2 full dephlegmator.

    I would be tempted to use 2 inputs. One with fine control on it to use during the run, and one with max flow control to rapidly put you into 100% reflux.

    Just my opinion though.

  • edited November 2014

    If you are going to go all out, hell, why not make flow in true countercurrent?

    TMW - I'm still not convinced, it throws everything I've ever learned about HX on it's head. What's the downside of using the entire HX volume, which will be higher efficiency? Two things keep prodding me in the eyeball, by using less of the HX you'll need to compensate with a higher flow rate, a lower input temperature, or potentially both.

    Why would you want to put yourself in a position where you need more flow, and colder input water?

    What am I missing? Do you lose controllability due to the relative size of the HX in comparison to the load? Do you gain that back when only using the lower-efficiency partial volume? Suspect that could be the case. The other potential downside is that if you HX is so massively oversized, the lowest flow rate necessary to maintain a set reflux might still be too fast (inability to reduce reflux rate past a certain minimum point). This might manifest itself in a cold HX taking a very long time to heat up, thus keeping the column in 100% reflux for significantly longer than necessary.

    Outside of my pay grade for sure! :))

  • The other benefit of running this as a true countercurrent is that you get the benefit of additional mass transfer through the HX tubes as more liquid will condense towards the top of the HX, and not only at the bottom.

    As the drips fall down the wall against the vapor rising upwards, you get additional mass transfer and increase in the ABV of the vapor, aka "Dephlegmation".

  • Ok this is just my take on this. There is a good reason for being able to put the column into 100% reflux, and this functionality is unwanted during the run.

    So there are a few things to think about. The HX needs a decent flow rate to avoid hot spots etc, and low flow rates are counter productive. So what options are there.

    1. Use 2 dephlegmators. Both for full reflux and just 1 during the run. This gives you the reduced volume you wish (to increase sensitivity) and enables you to maintain a higher flow rate.

    2. Use the 1 big dephlegmator, maintain the higher flow rate, and reduce the sensitivity by increasing the temperature of the coolant flowing into the HX.

    Now I am not sure if anyone uses this method (2) on small scale columns at the moment, but it may be an option. Temperature control inside the dephlegmator is important, but so is sensitivity and response time.

  • edited November 2014

    Recirculate the hx? Or recirc using a standard 3-way temperature mixing valve. In this manner you could maintain a high internal flow rate at a relatively narrow temp range.

    Response time is not so important, the heating load seen by the reflux condenser actually changes very, very slowly during the run. The only time responsiveness comes into play is when the operator changes settings. My proportional valve/controller setup actually changes very, very slowly during the run. My new larger proportional valves take 2 minutes from full close to full open, the old valves took 1 minute from full close to full open. No difference at all in how they act.

    I still want to see someone come up with a method for true countercurrent reflux at reasonable flow rates, that would be "super".

  • edited November 2014

    Not sure i can add much to the conversation, but this is how i get counter flow to my defleg.


    800 x 540 - 71K
  • Bloody Hell!

    I've solved the problem!

    Will switch my 6" rig from CM to VM; simple as that, cooling water management issues a thing of the past.

  • @mikew - what if you put some sort of sight glass/air trap where the orange elbow is? Some way to remove air from the system and trying to pump it down?

    How about some sort of reversing valve like the design on a heat pump? It could be manual or solenoid.. they have 2-in and 2 out, and a sliding piston changes the flow direction... run it for one minute with cooling flowing up to flush air bubbles, then reverse.. You could not use those valves directly, but the design is intriguing..

  • I would imagine it would require some sort of automatic air vent and check valve to work correctly, no?

  • the reversing valve could replace the air vent and check valve, you flow the system with coolant going up to flush air, providing your return line is small enough to have high enough flow rate to flush air bubbles (I bet there is a buch of science on that, LOL), then when reversed, you just have a closed system unless you introduce air somehow...

  • edited July 2015

    Hi, I've made it finally :) 4 baffle plates in dephlegmator as @TheMechWarrior said.
    Didn't run this yet, as I'm waiting to make new or modify my existing beer keg.
    So here it is, my own, based on Crystal Dragon idea, Red Crystal Flame :)
    It's beta version and it needs some finish touch, parrot, new or modified old beer keg, eventually electronic controller.




    310 x 800 - 36K
    532 x 800 - 75K
    800 x 532 - 75K
  • edited July 2015

    @woxy, interesting, but I can not see the advantage over our Crystal Dragon™. If this build is more than a private project (and it pretty much looks like that), you are in the wrong place btw.

    StillDragon Europe - Your StillDragon® Distributor for Europe & the surrounding area

  • As I've spoke with LLoyd it's just and only private project (no worry about copyrights ). Advantage of this, will came out as I run it. Different plate design, different and more advanced dephlegmator. Rest is obvious.

  • @woxy, as long as this is about a private build, it's OK. Of course you are welcome here to exchange ideas, but we can not allow any competition trying to sneak in. Although we are all open minded this forum primarily is about StillDragon®. It would have been easier to determine if we could see any StillDragon® parts in your pictures. ;)

    Your Place to be >>> <<< Home of the StillDragon® Community Forum

  • Sorry for that, These parts are my build, of course based on crystal dragon, but my built! Ie, the plate has 6 risers vs 5 in CD 4"., and few other changes .

  • What is the expected benefit from the baffled dephlegmator?

  • Minimizes entrained hot spots by creating turbulence.

    StillDragon North America - Your StillDragon® Distributor for North America

Sign In or Register to comment.