The following is from a draft I last updated in March 2015. I was very inspired at the time but it took me forever to let the rubber hit the road.
"I want to build a nice looking, powerful condenser.
Things I do not want:
I don't want it to be 3+ feet long.
I do not want it to have to be run vertical. I would love the ability to run it at close to horizontal angles too.
I want to eliminate round hollow tubes for the vapor to pass through. Round hollow tubes allow vapor to center itself and extend down the tube and then collapse causing the huffing effect we have all seen.
I do not want to run copper or SS packing inside of it either as I believe they contribute to smearing when run at shallow angles.
I'm pretty sure I have eliminated just about every commonly used condenser with my pickiness.. Time to start fresh.
Things I would like to incorporate into this design:
Low(er) vapor speed than 1/2 inch tube. 1/2 inch type L has an inside hollow area of .23 of a square inch. It's not much when you think about it. On a stripping run the vapor can be moving very fast through that small of an area. Instead of calculating vapor flow speeds from the boiler I will just calculate the difference from the standard 1/2 inch tube.
18-20 inch total length to take off. I want it to get far enough from the boiler but not be long and cumbersome.
Eliminate the chance of pooling. I don't want scrubbers in it, or any stepped down fittings that could collect liquid when run at a near horizontal angle.
I want my water connections to be near the boiler so they are not pulling down on the end of the unit. I want this for balance, hose routing, and to keep water from going in my collection jar in the event of a leak.
It wouldn't kill me if it looked kinda cool too.
The only design that physically comes close to what I want is the double/triple jacketed "liebig" condenser. There seems to be a few variations of this design. Most seem to require somewhat tough soldering and leave no way to visually inspect the inside. Cleaning could be a problem and pooling could be as well. I decided to make a variation on a double wall condenser and make it to fit my needs.
To start with I ran some numbers on a standard 36" x 1/2" liebig. The area of the inside of the 1/2" tube is .23 sq in. Not very big. The vapor contact area inside a 36" long, 1/2" copper pipe is 61.56 sq in. The inside diameter of 1/2" type L is 0.545". The inside circumference is 1.71". 1.71 x 36 = 61.56 sq in. It's volume is 8.4 cu in. So the vapor has an opportunity to come into contact with 61.56 square inches of copper as it passes through. When you push your condenser hard enough to let vapor escape it is because the velocity is too high for the amount of cooling contact area available. This can be improved by adding more contact area, slowing the vapor speed, and can also be affected by turbulence in the vapor path.skow69 wrote:Ya know...
Thermal conductivity is proportional to contact surface area according to this paper that still_stirrin' posted last week.
The area of the inside of a 1" pipe is 0.83 sq in(.870 if you get type M.) This is 3.6 (3.78/typeM) times the area of the 1/2". If a given volume of steam or vapor was pushed through this larger area it would pass through the tube at approximately one quarter of the speed. Using an empty 1" tube for a condenser would not be efficient though because of the large open area in the middle with no cooling. So what if we put a 1/2" copper tube up the middle of it? and ran cooling water through it? The area taken up by the 1/2" copper is .31" due to it's .625" outer diameter. So if we had a 1/2 inch copper tube running up the middle of the 1" copper tube the area between them would be .52 sq in(.560/typeM), a little over double that of the inside of a 1/2" copper pipe. Not only that but there would only be a 0.2" (.215"/typeM) gap between them if they are perfectly centered. That gap will be between two cooled surfaces assuming a jacket on the outside and cooling water going through the inner tube. That will cut the vapor speed by half at any level, and eliminate the cylindrical vapor path. The same amount of vapor traveling through this path will be spread out instead of bunched together. Couple that with the fact that it will move at half the speed and the vapor has no chance to slide through. Using a length of 20 inches the inside contact area of the 1" tube is c = 3.22" x (l=20") = 64.4 sq in (66.2"/typeM.) The outside surface area of the 1/2" inner tube is c = 1.96 x (l=20) = 39.2 sq in. So this design will have 103.6 (105.4/typeM) square inches total cooling surface area, around 40% more than a 36" liebig, with half the vapor speed.
The volume of the area in this design is equivalent to having a hollow 3/4" type M (.032 wall .81 id) copper vapor path. Which would be the next step up, but because of the not-hollow vapor path this will be much more efficient.
This thread cannot exist without pictures.
I guess I'll just bury this as a draft for now. Without pictures it's useless. VVV
Re: Cooling water vs Distillate temp
Postby still_stirrin » Mon Apr 06, 2015 1:34 pm
I agree with Rad....your vapor is moving too fast, i.e. - 5.5kW power into a 1/2" vapor tube (inside your Liebig). That's out of proportion. If you had a couple or 3 of those vapor tubes in the condenser (think "shotgun"), you'd have slower vapor velocity and greater heat transfer to your cooling liquid.
For grins...did you do a calculation on the parent site to see how big the condenser should be for your input power? (see example below).
To (approximately) size the length of a shotgun, use the vapor to liquid contact surface area.
For example: the calculator calls for a 132" Liebig (with 13mm, 1/2" diameter vapor tube), with 15 deg C inlet and 50 deg C outlet (water) and 5.5kW power.
That equals pi x diameter x length; 3.14157 x 1/2" x 132 = 207.3 sq.in.
2 vapor tubes (1/2") requires 66" length for 5.5kW
3 vapor tubes (1/2") requires 44" length
4 vapor tubes (1/2") requires 33" length
5 vapor tubes (1/2") requires 26" length
6 vapor tubes (1/2") requires 22" length
7 vapor tubes (1/2") requires 19" length
You can improve efficiency of heat transfer by adding turbulation in the water side as well as the vapor side. Also, contact time with the heat exchanger (think "vapor velocity") will improve the conduction of heat through the vapor tube's material (copper, I assume).
Bottomline here is that you're overpowering your 3/4" over 1/2' by 36" long Liebig. Hence, the high water flow rate to cool the liquid product. And if you turn the water flow down till the water outlet temperature is where you want it, the product comes out hot...liquid, but hot.