#### Welcome!

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

# Cooling Surface Area?

edited September 2014

Anyone have calculated cooling surface areas for the various dephlegmators/condensers? Something incorporating the number of tubes by the circumference of the tubes by the length of the tubes?

Tagged:
«1

StillDragon Australia & New Zealand - Your StillDragon® Distributor for Australia & New Zealand

• It is easy enough to do. However, perhaps more usefull is practical examples of achieved cooling capacity from actual runs.

The reason I say that is because variations in coolant temperature and flow rates can significantly change the performance of the condenser. All the condenser calculators are just guidelines.

I tend to use the condenser calculator from the HD home site with default settings, just to provide a starting point for condenser sizing.

So for 3kW you might need to consider 2m of 12mm dia tube. So a 5 tube condenser 400mm might be called for. Now with a decent coolant flow rate that condenser could easily cope with more than 3 kW.

• Everyone's practical example isn't so practical. There are always differences.

Coolant temp can vary of course, but surface area can't. I'd like to be able to calculate the numbers, but the length and diameter of the pipes isn't published (as far as I know).

For example, people report take-off rates, but at what proof gallon?

What I want to do is to construct a grain in mash cooler, but without knowing surface areas its difficult to know whether to get another 3" (long) shotgun or something else.

• Ah I see. Perhaps the SD team could put up a table of tube dimensions for the condensers. I have the numbers for copper tubes but not for stainless.

• edited September 2014

The math for this makes my head hurt, and usually I'm ok with math. This is like an entire engineering discipline in and of itself.

If you are so inclined, here's about the easiest tutorial I ever found on how to do the calculations. It doesn't look insurmountable, but I'm really not looking forward to spending an entire weekend trying to teach myself this:

Shell and Tube Heat Exchangers Basic Calculations @ PDH Online (PDF)

I think the problem is that even if we were to come up with a set of numbers, real world operating conditions would play more of an impact than condenser sizing and geometry. For example, if you only had warm and slow water, you'd need a massively larger exchanger than someone who had an unlimited supply of 50 degree F water at mains supply pressure.

I did calculate the total surface area of the 12" dephlegmator, it's got nearly 12 square feet of surface area including the top and bottom faces. Something absurd like 38 feet of tubing.

• True, it would be hard to exactly know what effect a condenser would have on a particular liquid, but at least a rough calculation of surface area would help. For example, if a 8" dephleg has 4 times the surface area for twice the price of the 3" long condenser, that makes the dephleg a better choice.

I don't think this should be so hard. A rough calculation would just require the number, diameter, and length of the tubes.

• edited September 2014

I can't sleep.

It's all about keeping as many things as possible constant. @grim is absolutely correct. One of my main jobs is programming PLC's for cooling of pharmaceutical lines, making drugs. This process doesn't start at the condenser, it ends at the condenser. The process starts at the cooling tower. Keeping a constant water temperature at a constant flow. The second place is the temperature of the process variable, in this case the fluid. In between there is a shit load of variables but these are all minor in the equation. Constant coolant flow at a constant temperature with a constant process variable makes for a constant product. I will start a thread on here soon about cooling tower design, this to me is the most important part of keeping the whole process constant. It makes it easier on the temperature controller. Cooling towers are cheap to make and provide the best possible solution to the start of the problem, they consume very little water, evaporation is the best way to cool a liquid. So this is only my opinion and feel free to tell me to go jump if you don't agree, the starting point for anyone should always be the cooling supply. How am I going to supply this rig with a constant and controlled supply of heat rejection? Once I have that supply how do I transfer the heat to it efficiently? What do I use to control that thermal transfer accurately?

Condenser sizing, buying a pre-built unit, isn't really that difficult as someone else has already done the work for us, condenser design is hard. Designing with copper isn't that hard, it has a great thermal conductivity at 385 W/m K. You can be off in your calcs by 10% and the condenser will still work, no such luxury with stainless steel. You can get a 5mm copper rod 100 mm long and heat it and it will burn your fingers in about 5 seconds. Stainless on the other hand is around 14 W/m K. It's total shit at conducting heat. Get the same 5mm X 100mm length and heat it and the heat wont reach your fingers. With those figures we look at stainless as an insulator and regularly use it for that purpose in corrosive environments. But here we are talking about the total opposite. We want something to conduct as quickly and efficiently as we can. When the engineers I work with design with SS they use insulating tables and NOT conductivity tables in their calculations. Its the nature of the beast. Stainless for us serves one purpose and that is to not contaminate the solution or corrode in the process. Beyond that its nothing but a pain in the arse when we need to maintain a constant temperature in a non-contact process. This is a science, we have thermal engineers that design our condensers for us. They tell me what I need to supply in terms of coolant quantity and temperature and I do it. This is what I know, a 2 metre single tube SS 1mm wall thickness condenser is more efficient and controllable than 10 X 200mm tubes. Its more about how long the vapour is in contact with the wall than how many walls you have.

The thickness of the tube used in the condenser plays a massive role in the condensers' effectiveness. If you use a 200mm X 0.2mm tube in the condenser instead of a 200mm X 1mm tube thickness, the condenser will be 45% more efficient using stainless. If you go to 1.5mm and a short tube length you need really, really cold water to be effective otherwise the tube will simply act as a pass through vapour pressure drop. On some of our condensers, if they could be made of copper they would be 85% SMALLER. The difference here is massive. In copper its only about 1% difference between 1mm and 0.5mm in thermal conductivity. If you have noticed all QUALITY stainless steel frying pans and pots have a copper loaded base for this reason.

So this brings another question. There is a common consensus that copper in the vapour/liquid path is a good thing to reduce sulphur/sulphites. Maybe the outer shell should remain stainless and the components that do the work, the tubes, should be copper. We cant use copper in my industry as many drugs are extremely corrosive and are usually acid based and boiled off, acid loves copper. For anyone that uses a common cholesterol drug, its base is 99.9% acid which is why you have to take an antacid drug with it. But as far as alcohol is concerned copper seems like a wanted thing. So maybe @Lloyd can make a single condenser with copper tubes and do some tests or make a dephleg with really thin 304 walls, down to say 0.2mm. Its not as if the tubes are under any sort of stress or pressure as they are enclosed, they are highly unlikely to be damaged, their job is simply to conduct heat, but copper does that better than anything we know.

PID control on the condenser, PID control on a cooling tower to maintain a constant coolant flow/temperature and a material that provides maximum thermal conductivity over the shortest period of time for the condenser. Its not that hard, but its not that easy.

• If you just need a cost vs surface area for the purposes of comparing condensers, then you can ignore all the other variables. The surface area calculation is not that bad.

• edited September 2014

I still cant sleep.

Sorry Myles my post had nothing to do with cost, I was talking from a purely engineering/scientific point of view. Thermal conductivity rules. The shorter a condenser or the shorter the vapour contact time is with a given material is the more the rule applies. I love physics and in this case it's very rarely wrong. You will be suprised how much difference 95% in thermal conductivity can make between copper and 304. I simply wanted to point out the differences between the different materials and that the end product is more dependant on the start of the heat rejection process and the material that transfers this thermal energy. The difference between 10 X 200mm tubes in parallel as opposed to 1 single 2000mm tube is about 4 times in cooling efficiency, not 10 times as you would think, vapour speed increases through that single tube but none the less it is still more efficient at cooling due to the longer contact time with the walls and the pressure drop this creates as it approaches the top of the tube, Bernoulli effect also creates cooling of that vapour. But the the surface area is still the same, the vapour contact time with that surface area is immensely different. When you compare SS to copper the numbers are really shit. And then factor in this time lag between how long it takes copper to transfer this energy and how long it takes SS to do the same thing 95% is BIG. If someone would/could come up with a material that had the thermal properties of copper with the corrosive properties of SS they would be a billion billion aire over night. No one has so we are left with a trade off. And so we are having this discussion.

• edited September 2014

Still can't sleep

@grim Great link too mate. It's not that hard to understand. Well not for me :))

• edited September 2014

Stainless vs Copper debate rages on! Thunderdome!

Unless there are specific size constraints, you can design in such a way that the differences in thermal conductivity are accommodated and the condensers will provide similar performance. I agree, copper rules, but copper foil is going to be significantly better than copper pipe, but good luck getting that to work! Cost and complexity of manufacture are important too. And lifespan - the brazing of dissimilar metals in this instance can sometimes cause premature failure (thermal conductivity being your enemy in this case).

I can throw a monkey wrench into the debate (well maybe not so big, but a wrench nonetheless). Are you sure that smaller, more efficient condensers are better? Was an interesting thread over at ADI about the difference between a dephlegmator and a reflux condenser, and some material posted about separation taking place within the condenser. So, what about the additional mass transfer and rectification that takes place within the reflux condenser tubes? As vapor moves upwards and condensed distillate rides down the walls, you'll have additional separation taking place. By using a less efficient condenser, with more surface area, you may yield higher levels of separation. The effect is probably minor in reality, but points to a benefit of a larger surface.

• A pure silver condenser....expensive? *-:)

• @grim We do stainless to copper/bronze/brass/alloy bonding all the time in different instruments, EutecRod 157 is a purposely designed alloy for this. That's not really the problem. Our instruments soldered with 157 last years without failure. For anyone interested in making their own parts but cant afford the TIG welding path look at 157 as an alternative. Shit easy to use and massive tensile strength. All you need is a soldering iron.

And I'm not pointing to a smaller condenser, I wouldn't change the condenser length from what it is, I would simply like to see what changing the wall thickness of the condensing tubes or the material itself would do to decrease the thermal transfer time. Separation will take place inside the tubes, I understand this. The pressure at the top of the tube will be lower than the bottom of the tube, Bernoulli effect. The more efficient the condenser the greater this pressure drop will be. This separation would also act as a mini reflux inside the dephleg, a la reflux column. You would have multiple liquid condensation points inside a really efficient condenser. The condenser would ultimately end up more efficient with the same surface area, dew point for any given fluid would occur at a lower point inside the condenser.

• @captainshooch said: A pure silver condenser....expensive? *-:)

Build one and get back to me, I want to know now you have brought it up. :))

• @Mickiboi thanks for the tip on the EutecRod 157, I may have to try and scour eBay for a bargain. I have used Harris silver solder and liquid flux for some ss work before, it this stuff more or less the same, but with a flux core?

• @Mickiboi that silver cond. is out of my price range so I just bought a second one from @Smaug last week to connect in line for the summer months. :)

• Would be interesting if everyone could report on their 100% knockdown capacity in KW. Might be an interesting way to compare capacity. Suspect coolant temp and flow rates are going to make a big difference though.

• edited September 2014

@grim could not agree more. 7x12mm is fairly standard in a 2" shell. It would be nice to have a range of performance statistics based on coolant flow rates. Just as guidance. Coolant temp is massively significant also.

• @brewsmith said: Mickiboi thanks for the tip on the EutecRod 157, I may have to try and scour eBay for a bargain. I have used Harris silver solder and liquid flux for some ss work before, it this stuff more or less the same, but with a flux core?

No its not flux cored, you need to use liquid flux with it, the same stuff you use with silver solder will do.

• thinking out loud..... for efficiency, the peak would be when you have coolant jat or just below vapor temp, at an infinite flow speed... you never cool the vapor more than needed... if you have too cold of water at a decent enough flow, you are chilling your distillate and sending good alcohol downwards further, not refluxing it to separate the water and good alcohol, correct?

• @CothermanDistilling Agreed if we are talking dephlegmators. I think it is correct that you do not need to do any more than condense the vapour. For a product condenser though I do like to cool the product slightly. Although there is a good argument for not doing so - as we are going to air it out anyway.

• It's a good point, there is a risk of subcooling the condensate when talking about reflux condensers. If the refluxed condensate is too cool, it will result in decreased column efficiency.

However in practice, I think you might need to go out of your way to create this problem, under typical conditions you'd be hard pressed. In a typical plated arrangement, I'd imagine you'd likely hit 100% reflux before you began to subcool.

This is really the theory behind running reflux condenser coolant temps on the warmer side (for example, by using the product condenser output coolant), since the coolant temp itself will represent the lowest minimum bound. You simply can't cool lower than that. Compared with, for example, a very small reflux condenser that required very low temperature coolant.

• @CothermanDistilling said: thinking out loud..... for efficiency, the peak would be when you have coolant at or just below vapor temp, at an infinite flow speed... you never cool the vapor more than needed... if you have too cold of water at a decent enough flow, you are chilling your distillate and sending good alcohol downwards further, not refluxing it to separate the water and good alcohol, correct?

Absolutely right @CothermanDistilling. Going back to the earlier thread on here somewhere about bypass and mixing valves, the mixing valve set up had a pump in the circuit pumping a shit load of water through the condenser circuit. The formula for heat rejection is DeltaT(degC) X Flow(L/Sec) X 4.186. There are 2 things you can vary here, the difference in the entering and leaving temperatures of the coolant water or the flow of that water. I opted to try the bypass valve set up for UZGin as it was impractical to setup a small system like a 4" with the mixing valve and an in line pump, but in any case the bypass worked perfectly, we ended up with 4 degC difference between the in and out. I would still like to try the mixing valve setup though, maybe when he gets back from Europe we do some mucking around. A massive flow of water keeping the whole condenser at a constant temperature with virtually zero difference between the entering and leaving water temps.

• So if I am running two separate ground water lines (no recirc for now) with gate valves controlling flow of each separately, what temp/range should I be shooting for when measuring water output on the reflux and product condensers?

• Probably makes more sense to determine your own set points experimentally, since the output coolant temperature is going to be dependent on the heat load, flow rate, and coolant input temperature.

Relatively easy to for the dephlegmator, to measure this start full open, and then adjust flow downwards until you get a touch of condensate, than reverse until you have none. that's your bottom end, label this 100% reflux.

Then keep reducing the flow to the deplegmator until you get zero reflux being produced. That's your top end, 0% active reflux.

If you want to get fancy, find the temperature that gives you the flow rate between the two temps, and make a note.

Lay it out on an X/Y chart, plot the points, draw a line or two, and now you've got your own personalized reflux curve. Nobody else's is going to be similar. Frame it next to your still. Realize though that if you change the operating parameters and configuration, this chart will change. If you are using ground water that is warmer in summer, colder in winter, the chart will change with the season. Change your reservoir pump and it'll change, etc.

Your dephlegmator temperature is controlling the reflux ratio, which is making your physical plates act differently. Higher reflux and you get more separation, perhaps your 4 plates act like 6. Lower reflux, maybe your 4 plates act like 2. The this is giving you the ability to adjust the ABV output, within reason (at the expense of flow rate and power consumption).

Operationally, if you are starting out in 100% reflux, typically you will adjust the dephlegmator coolant flow lower until your target take off ABV is reached.

Seems much easier if you really just focus on understanding the temperatures that correspond to 0 and 100% reflux and then working within that range to get what you want.

• I've found, like.many have said, every still, situation is different. If you have a still type you want to use mostly, and a set product you want to produce, you need to build it to suit them. I've built many condensers, started with over engineering and slowly tapered down to suit my still and what I want to drink. All calcs I've seen or tried are not correct, they give you a ball park figure which is good.

• edited September 2014

Just a quick thought that could make a copper core condenser a bit more attainable... both for SD and us hobby guys.

whatever you want to call it, a product/reflux condenser or dephlegmator, we're talking about various form of heat exchangers. That got me thinking about my uncle's Stanley Steam car where he built a custom boiler for it that essentially looks like a massive dephlegmator (picture around 30" diameter x 24" height with 200+ copper tuber running through it. All pressure tested to 1400psi. Similar pic attached.

The main housing is steel and there are zero copper to steel welds used. He hand flared and rolled each tube to fit using a tool like this:

Condenser Tube Expanders

If we were to get a standard SD dephleg or condenser without any tubes in it we'd get a shell like the mockup picture shown. Making an assuming the holes are near the size of common copper tubing, we could then cut and install the copper tube with an expander.

No welding dissimilar materials, high quality mechanical fit, stainless look, copper performance, and part DIY to have more connection to your still.

Here's video that shows part of how the boiler is assemble with a nice tool customized for it if you're doing 100s of tubes:

• edited September 2014

wow..... thanks! here is the video of the seal being made with a roller

• edited September 2014

wow... this roll beading expander makes a NICE bead.... but only in larger sizes...... you could make some NICE dephlegs with one of these...