Jump to content

Condenser Design/Size/Build Question


Stumpy's

Recommended Posts

I am currently working on building a new condenser for my pot still. I recently purchased a "retiring" distillery and am not happy with the condenser design that is currently attached to the pot still. The current design is a 10" tube with a 4" tube inside. The 4" tube has baffles to create turbulence for the vapor. Based on the size of the still (700 gallon, 75" boiler diameter, 16" column that has 5 plates and is 5' tall) I do not expect that condenser to be too terribly efficient....not to mention that it is rusting inside!

I want to build a shell-and-tube condenser. I have read the forum topics on this matter but didn't really find what I was looking for. I was wondering if I could bounce some ideas/math/assumptions off you guys.

Based on my calculations and assumptions, I believe I will need about 155,000 Btu/hr heat input to finish the run in a reasonable amount of time. One big assumption I used was that the overall heat transfer coefficient for this type of heat exchanger was 175 Btu/(ft^2*hr*F). Am I way off base here?

So doing the math based on the calculations and assumptions I show in the attached file, I find that I need 29 sq-ft of heat transfer area, at minimum. Breaking that down based on material I can buy, here is what I am thinking. 8" shell diameter, 1/2" tube diameter, 5' tube length, with the area of tubes being 33% of the area of the shell. That leads me to having 85, 1/2" tubes in an 8" shell which would give me 56 sq ft of heat transfer area. The factor of safety comes out to 1.94 so the condenser is a little over-sized to be safe.

Based on what is shown in the attached file and the text above, do you all think I am in the ballpark? I want to get your experienced opinions to find out if I am blowing smoke up my own butt before I go and spend a small fortune on this thing!

Please let me know your thoughts...I look forward to hearing them!

Adam

Heat Exchanger Design.pdf

Link to comment
Share on other sites

Just curious, how did you calculate your 155K BTU input? Is that before or after you remove a lot of heat from the system via dephlegmator?

Also, are you using copper? 175 heat transfer coefficient seems high for stainless steel? Also, that coefficient is itself variable depending on the temperature differential between shell and tube. If you're figuring an average 30F differential, is your coefficient value still in the ballpark. Another factor is the vapor velocity of the vapor entering the condenser. If you're using a narrow lyne arm it could be possible for your vapor to be moving too fast for your condenser (a problem I've experienced).

I'm no engineer, but I've often tried (and continue to try) to engineer systems using similar calcs, but I've found it's almost always more predictable to extrapolate from data you get from similar real world systems rather than calculating from scratch.

Link to comment
Share on other sites

Just curious, how did you calculate your 155K BTU input? Is that before or after you remove a lot of heat from the system via dephlegmator?

Also, are you using copper? 175 heat transfer coefficient seems high for stainless steel? Also, that coefficient is itself variable depending on the temperature differential between shell and tube. If you're figuring an average 30F differential, is your coefficient value still in the ballpark. Another factor is the vapor velocity of the vapor entering the condenser. If you're using a narrow lyne arm it could be possible for your vapor to be moving too fast for your condenser (a problem I've experienced).

I'm no engineer, but I've often tried (and continue to try) to engineer systems using similar calcs, but I've found it's almost always more predictable to extrapolate from data you get from similar real world systems rather than calculating from scratch.

I came up with the 155k BTU/hr from the specific heat equation (Q=m*Cp*dT). For the mass, I assumed the weighted average of water and ethanol, based on a 10% wash...same for Cp. I used 212-70 for my dT and 5 hours for time. That should be assuming no heat removal from dephlegmator.

I will be using copper for this particular build and will do an iteration in stainless after cash starts flowing! I was planning to have a vapor chamber at the top of the condenser to introduce swirl...similar to air induction into a cylinder in a Diesel engine.The narrowest point in my lyne arm is about 4.25" ID...hoping that the vapor isn't too terribly fast through there.

Thanks for the input!!!

Link to comment
Share on other sites

If your figures are correct, the condenser would need a 12 1/2 ton chiller to operate.

I doubt anyone out there needs that size chiller to perform that duty. I may use that size on the system only to perform the mash cooling, but it would at partial load at the condenser cooling.

The condensation of the water+ethanol mix is much more complicated that the simplified version you used, so be wary of extrapolations.

I was wondering why aren't you just purchasing a condenser from a reputable company that have people who have worked in the field for many years?

Anyway, good luck to you and feel free to call me if you would like to discuss cooling strategies.

Regards,

Mike

Link to comment
Share on other sites

If your figures are correct, the condenser would need a 12 1/2 ton chiller to operate.

I doubt anyone out there needs that size chiller to perform that duty. I may use that size on the system only to perform the mash cooling, but it would at partial load at the condenser cooling.

The condensation of the water+ethanol mix is much more complicated that the simplified version you used, so be wary of extrapolations.

I was wondering why aren't you just purchasing a condenser from a reputable company that have people who have worked in the field for many years?

Anyway, good luck to you and feel free to call me if you would like to discuss cooling strategies.

Regards,

Mike

Mike,

I figured my water/ethanol weighted average was a pretty BIG assumption so I weighted it a little further toward the water side to build in a little buffer. That is why my specific heat is so close to that of water. As I am reading through my old fluid dynamics textbook I am realizing how much I have forgotten over the years!

I suppose I like doing things the hard way. I have a little free time on my hands and love fabricating...not to mention that we are in the middle of equipment setup for a new "bootstrap" distillery so I am trying to stretch every dime as far as I can right now!

I think we should be able to get away with a smaller chiller than 12.5 ton....still not saying my calculations are correct! I think it would take 12.5 ton to remove 155k Btu/hr but we will not need to remove the entire thermal load put into the boiler.

As I sit here and type this, you made me realize that my math assumes I will be condensing every drop of liquid in the boiler, not just a portion of it. So you are right in that my condenser is way too big. Back to the spreadsheet now. Thank you for pointing my brain in the right direction!

Adam

P.S. I'll be in touch when we are designing our cooling loop.

Link to comment
Share on other sites

If you are brave, you can pick up a 600kbtu stainless steel pool heat exchanger for about $500 bucks on eBay, or the smaller 300kbtu for $300 or so (although the 300k looks pretty small). I would imagine it would be difficult to make it yourself for less money. No experience with how well they work, but I've always wondered. They are either a complete waste or money or they best bargain around. You can probably dig up the specs if you dig hard enough. Other nice thing about most of these is that the tube side runs straight through, so you don't need to push the vapor through the shell. Some of these even have the convoluted tubes.

Link to comment
Share on other sites

If you run your coolant through the tubes and your vapor outside the tubes with baffles alternating side to side every 8" or so vertically, your condenser will be a great deal more efficient. The whole exterior shell is heat exchange surface area (the outside air is the coolant) and your tubes have a much larger OD than ID, plus the baffles themselves are heat exchange surface area because they are in contact with the vapor tubes. Of course the heat exchange surface area of the baffles removes fewer BTUs per square inch than the tubes, because the tubes have direct contact with the coolant.

If you are not going to be doing your own TIG welding and if you do not get your copper or stainless directly from the factory that makes it, I am sure that I can build you one for less and it would be TIG welded instead of brazed or soldered. Soldering or brazing this kind of condenser in copper can be a real Bear. Copper sucks the heat in so fast that you will have a really hard time soldering in those tubes without melting the solder on the tubes that you have already done, because the tubes are so close together. There are ways around that but keep in mind that soldering and brazing are inferior to Tig welding. I can build it from copper but it would cost less for me to build it from stainless and you would not have to worry about contaminating your spirits with copper salts. Copper is needed in the vapor path but not in the condenser. Email me paul@distillery-equipment.com It would not hurt for me to give you a quote and I may be able to help you with some of your other equipment.

Link to comment
Share on other sites

Here is another tip. Use good clean water for all of your cooling processes and run the output from your cooling processes into an insulated hot water holding tank. Use the hot cooling process water for making mash and cleaning. The larger the mass of hot water the better it will hold its heat over time. This will save you a fortune in the long run. You get the heat from the water for free and its like you never even used any water for cooling. If you plan on doing tours and want something really nice, I can sell you a beautiful insulated stainless hot water holding tank at a better price than my competitors. However, you can save allot of money by using a polypro tank and some silver blanket insulation. You should always buy polypro tanks from a source that is close to you, like a farm supply or large feed store. They are large and they don't weigh much, so the freight class for them is a really high one, which can sometimes make shipping cost more than the tank itself, if you buy one from a few states away.

Link to comment
Share on other sites

On the Polypro tanks, Paul is right, plus most distributors have a good selection in stock and will help you out putting extra holes in and supplying fittings. Send my over an email and I can send you photos of a distillery in CT with their cooling loop and polypro chilled water reservoir (500 Gal still).

I just have been too tied up to get them on my website, but have been out of town.

Mike

Link to comment
Share on other sites

If you are brave, you can pick up a 600kbtu stainless steel pool heat exchanger for about $500 bucks on eBay, or the smaller 300kbtu for $300 or so (although the 300k looks pretty small). I would imagine it would be difficult to make it yourself for less money. No experience with how well they work, but I've always wondered. They are either a complete waste or money or they best bargain around. You can probably dig up the specs if you dig hard enough. Other nice thing about most of these is that the tube side runs straight through, so you don't need to push the vapor through the shell. Some of these even have the convoluted tubes.

I agree, I build my own. But I do have several "pool " heat exchangers from ebay. get the 360,000 or 600,000 models ( I personally use a lot of the 360,000's). I weld tri-clamp fittings to them and they work GREAT! and the other great thing about them is they are made with convoluted tubing (means the tubing is twisted) thats gives it more heat exchanging area.

get the stainless 304 /316 not the titanium ones.

I have the water on mine coming in from the sides and the vapor in from the ends long ways.

take care.

Link to comment
Share on other sites

Wow, thank you for your opinions everyone! The pool heat exchangers are quite intriguing. I may buy one just to test! They are priced low enough and made from 316, which I like. If all else fails, I can try the pool heat exchanger and if it doesn't work, I can go another route and not have a ton of money sunk. $500 for a 600k BTU/hr condenser seems way too good to be true...

Link to comment
Share on other sites

Stumpy,

It is to good to be true. Don't waste time and money buying that pool heat exchanger. As a condenser it will not give you 600K of cooling for your situation. As a condenser it might do the job in a 100 gallon still, as long as you did not use it for a stripping still.

Also please see my first reply to this topic. I made a typo. It should have read "If you run the coolant through the tubes and the vapor outside the tubes...... your condenser will be a great deal more efficient." It states that now.

Link to comment
Share on other sites

How do you deal with cleaning of the condenser if you run vapor through the shell? Do you have some sort of setup that allows for flooding of the exchanger?

Maybe it's irrelevant for a stripping setup, but seems odd to do this in any other situation. I could imagine all sorts of potential negatives. If the shell is baffled, there would be numerous areas to catch distillate resulting in smearing of fractions. I've never seen a tube in shell with a shell side that was condusive to draining. There isn't any position that wouldn't result in some pooling. Running mixed products would be significantly more complex, especially if you are going deep into the tails or running a gin or other infused product. The shell is going to get fouled.

What do you estimate the potential efficiency benefits are of running reverse? 5%? 10%? 50%? What size reduction does that translate to? There isn't any difference in heat exchange at the tube, since the distance between the coolant and the vapor is still the same in either case (the width of the tube wall). Yes you gain some benefits of the outer wall for exchange, but unforced room temperature air is generally a poor cooling medium.

Heat exchanger is a heat exchanger is a heat exchanger.

Link to comment
Share on other sites

Vendome has been building their condensers using this type of design (coolant inside the tubes vapor outside the tubes) for years. I think Artisan is building some of their condensers this way as well, or at least they appear to be built this way. It is simply a better and more efficient design.

I am sorry but I will not get into specific detail about my calculations or designs. Cleaning is not an issue. Our current design (which is the design used on our 300 gallon Pro Series Whiskey Still, which is the one in my member picture) has the vapor going through the tubes and as you can see the condenser is oversized. I designed the condensers for my pro series stills to handle a 1 hr stripping run producing 25% low wines. That way they can handle anything that is thrown at them. No one is going to do a 1 hr stripping run but some of my customers do a 2 hr stripping run. Anyway, when we tested our new design with the same number of tubes and same diameter and length of the outer shell used in our current design, the maximum amount of vapor that the condenser was able to condense went up by more than 30%. My original calcs showed a greater increase, mainly due to the baffles. I think that I found the problem and so my employees are going to tear apart the prototype in a couple of weeks and sometime before the end of the month we will have the new baffles installed. We have several projects in the works right now so it's hard to do as much R&D as I would like to do.

And I agree, a heat exchanger is a heat exchanger but that little pool heat exchanger will not condense the vapor from a 700 gallon still, not even close.

Link to comment
Share on other sites

Coming form the refrigeration industry, old refrigerant condensers were all baffled, straight tubes, water inside the tubes.

As industry progressed there came "pin fins" on the outside of the tubes, eliminating baffles.

Things change because of manufacturing innovations, but these are copper tubes, so for SS you are back to baffles.

From looking at some of the equations used, I seem to recall that surface tension of the gas and it being suddenly made into liquid with its own surface tension, I have wondered how the difference gravity plays pulling the condensed fluid downward on a horizontally oriented condenser and a vertically oriented condenser.

It just intrigues me, but how much that makes a difference, who knows except by massive experimentation.

Link to comment
Share on other sites

the baffles are there mostly to fight gravity and force more interaction between the vapor and cooling tubes, yes they do offer surface area for heat transfer, but its minimal.

the Scotts have been using this method for their condensers for a long time and it is far more efficient on cooling water usage. water usage is 1/5 to 1/10 as much with this arrangement, with much higher knock down power.

only way to get a multi tube system with vapor flowing through the ubes to approach this level of efficiency is to baffle the top of the condenser to force the vapor through all the tubes and to tip the condenser at an angle of 45 degrees or better, to force further cooling of the falling liquid.

Link to comment
Share on other sites

Yea, we learned about baffling the tops of our larger tube and shell condensers the hard way. The first large tube and shell condenser that we built was 12" in diameter with a 2.5" diameter line arm. The dam thing was barely working at 20% of what it should have. We figured out pretty quickly that all of the vapor was going through just the center tubes. We tried a couple of different baffle configurations in the top until we hit upon the correct one. I thought about changing the angle of the condenser from 90° to lesser angle to slow down the vapor but I decided against it because of the way it would have looked.

My conclusions about the baffles, in the new design, were that they would slow down the vapor velocity enough that the interaction between the vapor and tubes would be increased dramatically. My figures showed at least a 4 fold increase in efficiency over my current desighn. But I did not get that. Now I think that I know why. We originaly cut the baffles so that each baffle would extend to the center of the column from the outer shell. Then after they were cut I decided to angle them down a little so that no condensate would pool on them. Of course we did not cut the holes for the tubes until after we decided to angle them down towards the center. Anyway, due to the angle of the baffles, they did not extend to the center anymore, so I think that allot of the vapor is coming off the first baffle and falling strait to the bottom instead of flowing over the other baffles. My new baffles will extend well past the center and we will keep the baffles level. I think that will do the trick but we will see.

Link to comment
Share on other sites

Stumpy,

In my opinion the 8" diameter 5' tall condenser that you are planning to build is to small the way it is designed. Considering that your still is 700 gallons with a 16" 4 plate column you will always be slowed down because of your condenser. If it were me and I was going to have the vapor go through the tubes I would build it at least 12" in diameter by 6' ft tall with as many tubes as I could fit into it and I would baffle the top. This will allow you to do a 5 hr whiskey run at 135 proof. My figures actually show that a 12" x 4' condenser in that configuration would do the job but I like to make sure. If you put the vapor on the shell side I would probably go with an 10"x 6' condenser and that would be oversized as well. These sizes are for stainless.

Link to comment
Share on other sites

I always thought the Vendome approach of coolant in the tubes effectively was meant to allow the tubes themselves to act as baffling for the vapor surrounding it? Any vapor not condensed in the shell will stay near the top (since bottom is cooler) until it has a chance to impinge on tubes.

Link to comment
Share on other sites

In the picture below, there is an example of a small, well built, well designed shell side condenser. Without the baffle plates this condenser would be no were near as efficient as it is with them.

Wow Paul, your craftsmanship looks great!

Link to comment
Share on other sites

I didn't build the condenser in the picture, I was just using it as an example. We usually build stainless condensers for our larger stills and columns. Below is a picture of some of our copper work. I tried to add more pictures but it would not let me. I must be over my limit, I guess. We have sold equipment to over 40 legal distilleries. You can check us out at http://distillery-equipment.com

post-3459-0-09340200-1420416093_thumb.jp

Link to comment
Share on other sites

I have been planning to build a new/larger condenser for my still and after following this thread I am thinking of changing my plans. I would appreciate any feedback you guys have.

Current setup is a 150 gallon, 4 plate, steam jacketed still powered by my 320,000 BTU output / 400,000 BTU input gas fired boiler.

Current condenser is a 5 foot long x 6 inch all copper shotgun condenser (mounted vertically) with twenty three 1/2 inner tubes. Vapor goes down the tubes and city water goes up the shell.

Currently we can not run with the steam full on for most of a distillation run as the condenser can not keep up; hence the desire to upgrade and cut down our run times.

My plan was to upgrade to a 8 foot long x 8 inch tube with thirty some 3/4" tubes and go all stainless and stick with the vapor on the tube / water on the shell side.

Now after reading this thread it sounds like I might get more cooling bang by sticking with a 8 foot x 6" tube but changing the vapor to the shell side and adding baffles. Thoughts?

Also any opinions on 304 vs 316 stainless for this type of application?

Here are photos of my current setup; can you see that I have room to extend the height of the condenser from five to eight feet without changing the layout/placement of things..

1488288_556614361099794_253084162_n.jpg?

547830_512853082142589_122220622_n.jpg?o

Any and all feedback appreciated!

Link to comment
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
×
×
  • Create New...