Problems with air-cooled

[muckanic]

Greetings all

I've got a similar newcomer's problem to the "can't control temp fluctions" thread, although with a different sort of pot-still. I'm using an external heating element, simmerstat-controlled, and an air-cooled condenser scrounged off the back of a refrigerator - about 20m of 6-7 mm copper pipe. The brew is a honey-fruit wine that went oxidised, for what that's worth. Theory says this condenser should be good for 800W or so, more if a combination of wet sheet and fan is employed.

My problem is that I either get foaming and run-away temperatures of 92C in the still head, or no product. All the obvious problems can be ruled out - the temperature controller is as fine as you could want it to be, it doesn't matter how full the boiler is (a 20L thin metal drum), and the wash doesn't seem to be anything exceptional. Previous experience with smaller-scale Liebig-based systems worked as advertised, ie, the temp rose to 78C and then just stayed there, so up till now I was inclined to blame the condenser. The foaming is the part that I wasn't expecting - my preconceptions were that if the condenser wasn't up to the job then it would result in vapour coming out the end, not wash.

What I am mainly wanting to do at this point is get my head around the theory of why the rig is behaving as it is. If I have to, I will go water-cooled, but before I do that I want to know why I'm having to do that.

I guess I'm a little reluctant to part with the air-cooled principle, so the next move could be to use wider pipe (13mm) of a similar length. At this point, another theoretical question presents itself: would there be any advantage in inverting the condenser with the outlet to the high side, and a take-off valve near the still head? In other words, it would be a pseudo LM design, albeit without the reflux column. My speculation at this point is that this could make the cutting out of the heads quite problematic, as the pool above the valve might not necessarily contain the most volatile substances initially. Actually, the other distinct possibility is that quite a bit of vapour might want to escape out the valve as well. On the other hand, if you take a look at Alex's Spiral Head Still at "http://groups.yahoo.com/group/Distillers/files/Air Cooled.doc", this principle is claimed to be successful for reflux stills, and you wouldn't think that would make much of a difference to the vapour escape issue. Comments?

[junkyard dawg]

I don't think I would use that condensor for drinking ethanol... unless you can be sure there is no toxic solder in there...

The 92C doesn't seem out of line, when in the run do you see that?

I like the air cooled principle, but its not going to work as well as say a coil in a bucket.

[hornedrhodent]

if you take a look at Alex's Spiral Head Still at "http://groups.yahoo.com/group/Distillers/files/Air Cooled.doc", this principle is claimed to be successful for reflux stills

This link didn't work for me - Do you have to sign up with Yahoo to view it?

[muckanic]

This link didn't work for me - Do you have to sign up with Yahoo to view it?

Don't think so. Try going to http://groups.yahoo.com/group/Distillers/files" onclick="window.open(this.href);return false;" rel="nofollow and locating the doc from there.

I don't think I would use that condensor for drinking ethanol... unless you can be sure there is no toxic solder in there...

Most fridge condensers are nothing more than a long copper pipe, painted black on the outside to increase efficiency. The only soldering occurs at the end lines to the pump, and that's about where you want to cut anyway. The pipe dimensions are slightly unusual - although the OD is 7mm, the ID is about 4mm, so the wall is thicker than standard 1/4" pipe.

A couple more data points. I scrounged a 16m/13mm condenser from a scrapyard which has the same "S" shape as a fridge condenser. Using this wider pipe helped but didn't solve the problem completely, ie, small flecks of wash were still coming through at any decent stripping rate, and the combined strength was only around 30%. I then hooked up a straight 2m length of 13mm pipe and it worked OK, although the distillate was too hot. When I drove it too hard, I got vapour and not wash, as is normally expected. Less expectedly, my still head settled at 84C during the mids, which bothers me a little as the thermometer bulb is positioned right at the start of the horizontal outlet.

Clearly, there is a bit more to condensers than just pipe width/length and ambient temperature, although most of the calculators around don't take too much more into consideration. In this case, the shorter pipe effectively beat the longer pipe. One theory I am entertaining is that maybe insufficient suction is being provided to draw out the product, through a combination of very long pipe and possibly insufficient use of gravity through predominantly horizontal lines?

Speaking of which, it now looks like I will wind up going with a Liebig and maybe pumping the water through the air-cooled condenser. Is there any reason not to employ a vertical Liebig (VM systems excepted)? I presume there could be a possibility of insufficient transit time through the condenser for the distillate?

The ultimate aim is to be able to convert readily between pot and LM refluxing. Initially, I would like to avoid an in-column condenser for reasons of simplicity, but it is starting to bother me that neither an air-cooled approach nor an inverted Liebig may do LM particularly well for the reasons I gave lost post. From what I can gather, the spiral still crowd on Yahoo seem to be doing such slow, low-power runs that they probably don't need a LM technique anyway. Comments?

[muckanic]

In a bid to keep this thread alive, here's someone else who used a narrow diameter condenser and started getting some weird temperature readings, allegedly due to pressure build-up: http://homedistiller.org/forum/viewtopi ... highlight=

Would like to get to the bottom of this...

[Harry]

It's simple fluid mechanics, my friend. A given volume of liquid, when vaporized, will occupy a space some 1,500 times the space of the liquid. So, you go putting this vapour into a very smallbore tube of considerable length, and it starts to collapse back to a liquid, the result you desire.

Now another rule of fluid mechanics takes over, that of friction. Liquid 'sticks' to the walls of the narrow tubing, and is slowed to a crawl. The vapour coming in behind it is thus blocked by the liquid, and so pressure builds up (sometimes dramatically, as in BOOM). Not the result you desire.

The vapour has to go somewhere, and if you're lucky and there's not much liquid in the tubing, it'll just 'spit & gurgle', driven out by the pressure.

There's a limit to how fast liquid can flow in narrow tubing, no matter what the pressure is driving it. If you simply MUST have air cooled, there's better ways. Ask Riku about finned tubing. If you're worried about water waste in a water-filled condenser, go for recirculation via a small pump and 50 litre bucket.

I'd also question if there's much ethanol in your wash, considering you say it was 'oxidised'. Ethanol dose evaporate, and quite rapidly, and 'oxidation' changes ethanol into other substances.

[muckanic]

I'd also question if there's much ethanol in your wash, considering you say it was 'oxidised'. Ethanol dose evaporate, and quite rapidly, and 'oxidation' changes ethanol into other substances.

The mash is oxidised in the same way that sherry is oxidised, ie, there is a small proportion of aldehydes, darkened melanoidins, rancid fatty acids, etc - all the result of storing some partially racked wine in a plastic drum for a couple of years. At this point, I am really just stripping it for new-still familiarity sake. One advantage of the over-pressuring is that it rapidly identifies all the weak seals.

[hornedrhodent]

Practice - practice - practice - you can use the same ethanol over and over again to get to know your still - just dilute it to simulate a new wash and go for it - learn.

[possum]

4mm is really small. When pipe size goes down, the friction can go up,ie; the liquid clings to the sidewalls. that may or may not be your problem, but the effects will start to show near the diamater that you are using.

[muckanic]

As I kind of indicated, the problem is a combination of thin diameter and length. Thin and short in this context works better than thin and long, which contradicts all the condenser calculators I have seen. So yes, it is probably a friction thing. People run into that problem with water pumps and condensation external to the pipe, not so much with internal condensers.