The Electric Dephlegmator Acid Test

Ok - purely on the fun side - I've been playing around with the idea of a fully electric dephlegmator/reflux-condenser using TEC/Peltier Modules and air-cooling.

I've used Peltier modules in other hobby projects a few times, and they are generally easy to use, easy to control, fairly responsive (depending on the thermal mass). They have also gotten to be incredibly inexpensive. Even what was historically a very large element, and hugely expensive, can be had for a couple bucks - though larger 250w+ modules will still run around $25usd a piece.

Using multiple lower wattage elements - you could get close to 1kw of cooling capacity for less than $50usd - elements only - power supply and controls is another factor.

Some have gone this route, but still used water as the cooling medium. My thought is to eliminate the intermediary, as water poses additional complexities, and an all-electric system would be a much more simplified approach.

Control would be relatively simple, and actually very similar to controlling the heating elements - you would simply manage total power input to the TEC/Peltier elements to control knockdown. You could monitor vapor temp, but from a stability and repeatable perspective, managing wattage input would be very easy.

Why I think this has merit? Theoretically - very fast, and very accurate temperature control, based on the thermal mass, could be very quick to respond. I am wondering if this would provide a mechanism to PID control with Vapor Temperature as the PV (not dephelgmator water temperature).

The real challenge here is the metalwork required, as TEC modules need good parallel surfaces for heat transfer, large heatsinks (with fans) - AND - the cold side would need to have a similar cold-sink, which protrudes through the column into the vapor stream. If you have a machine shop, or know someone who does - shouldn't be terribly complex - but trying to do this in a garage workshop - going to be tough.

I've got a few sketches and some ideas for mounting multiple modules around the TEC "dephlegmator". The most realistic design uses multiple modules mounted in square around a central circular column - as there would be no realistic way to make a top-mount unit - simply not enough real estate to add enough TEC modules.



  • So is 250W the max power in or out? ie the cooling capacity?
    Sounds like a cool (he,he) idea but my understanding of these things is they are terribly inefficient.

  • The other thing is what do you do with the waste heat? These will help pull it out of the vapour but you need to put it somewhere. Have I got that wrong?
    I know you said air cooling but there could be quite a lot.

  • I'm struggling to picture this - are you saying creating a setup similar to a cars Radiator?

    Sorry if I've oversimplified it but that's what it sounds like to me...

  • Peltier plates are what is in electronic solid state fridges... but they usually can only generate what 40 degrees difference?

    most ingenious idea is to have that heat go to the boiler... the logistics of that, however, are tough since there is a big separation distance between them....

  • I've been having a similar thought for my next setup. Using an air cooled heat exchanger with a fan on a vfd to cool the dephleg. The waste heat could be pumped into a barrel room to keep it nice and toasty... or maybe just outside.

  • edited May 2016

    @EZiTasting said: I'm struggling to picture this

    No radiator, no fluid. TECs mounted to the "dephlegmator" with fins protruding inward and fins protruding outward. The modules would be sandwiched in between the heat sinks. The exterior heat sinks would need fans. The inward fins would act as the condensation points - like a cold finger in a VM or LM - or like the tubes in a standard liquid tube in shell. The unit would be self contained - other than the electronics - no tubes, no hoses, no valves, no solenoids, no water supply.

    TEC is really just a heat pump - moves heat from Point A to Point B - where it can be dissipated over a larger surface area.

    Just some rough calcs - thinking about 720 watts sufficient for a 2", about 1440 watts sufficient for a 4".

    The problem with the larger units isn't the TEC modules, it's the power supplies. 720w would use 12v TEC, 60 amp power supply, and 60 amp PWM control. 1440w would use 24v TEC, 60 amp power supply, and 60 amp PWM control. Both of these are still realistic - since you can find this stuff coming out of the Orient pretty cheap.

    Going to a 6" column would probably need 120 amps at 24v - this is some serious current. 8" column something like 240 amps - 5.8kw.

    Suspect maintaining full reflux at higher power levels may be tough.

    DeltaT Max for many TECs is about 40c, although at higher current level

  • Not sure how that got cut of - that's usable delta T about 40c, max can push higher, but by 60-70-80c, you risk destroying the TEC.

    Max Delta T is at minimal power consumption, so mostly theoretical, at max power delta T drops to zero.

  • Damn! I was expecting Ken Kesey, The Grateful Dead and LSD...

    I'm more like I am now than I was before.

  • Why do you want a large deltaT? You're only after a phase change with minimum cooling usually.

    I think the temp difference between the hot side of the cell and the ambient air (which is what you need to shed the heat) will be high regardless??

    If you did have a high deltaT you could pump the heat into the boiler using a transfer medium like oil. That's getting very elaborate though. How would that work for controlling the RR? Putting the heat back in that you've just taken out?

  • but if you are going to pump anything, why add the peltier to begin with?

    I thought about this in a continuous setup, the peltier providing reflux and the hot side being cooled by incoming beer right next to it......

    This seems best where cold air is free and heating power is nearly free...

    or is a super tiny system where even a pump is too large...

  • edited May 2016

    Ultimately - this is still a variation on the "airstill" approach.

    What the peltier is providing is a way to trade increased electrical consumption for reduced surface area - both inside and out. Imagine what a dephlegmator would need to look like using only finned copper tubes and fans - it would be an enormous contraption.

    No free lunch.

    Being able to transfer the wasted energy - both the watts that are "pumped" plus the waste energy associated with the peltier operation - back to the wash would be a nice way to waste less energy. Agree with Cotherman - the obvious approach is a wash pre-heater in a continuous design.

    No idea if this would even work effectively - but it wouldn't at all be efficient, that's for sure.

  • As I said, if you want to put the heat into the wash it needs to be hot enough. The cell would do that for you.
    If your just preheating wash in a continuous setup then what's wrong with a plain Jane heat exchanger?
    They're cheaper, easier, don't need electrickery and are a hell of a lot more efficient at the job.

  • edited May 2016

    Talked to a tech at a peltier manufacturer today.

    He said that the efficiency should be slightly better than normal in this application, as we are pumping heat from an area significantly hotter than ambient, having a high COP (Coefficient of performance, in this case potentially COP>1) - unlike trying to cool something below ambient, where the COP moves fractional. However, he mentioned that high temperature modules would absolutely be required, as the hot side would be well in excess of typical operating range.

    What attracted me to this idea is having a very fast system, with minimal thermal mass in the dephlegmator - meaning the system could potentially work fast enough to be able to control Vapor Temperature as the PV, not simply maintaining a fixed dephlegmator temp in hopes that vapor temperature is stable. You could also very closely manage sub cooling. Even potentially be able to manage reflux ratio directly (as you would know the numbers in real time). And to be able to do this in a single package - no pumps - no valves - there is an appeal to that.

    He thought it was crazy due to the power requirements as well as the overall cost of heat sinks. He knows of semiconductor process chillers in this size range - but also said that temperature accuracy significantly outweighed the low efficiency and sheer system cost.

    He shipped me out some high temp modules for cost, just because I was the most entertaining case he had heard.

  • @Grim, any progress this project? This is very interesting to me. My only concern is the lifespan of the Peltier modules, as they seem to be short-lived for the Coleman electric coolers (which seem to last longer after being rebuilt). This may not be the case in this environment, since they will only be ran hard enough for a phase change.

  • A bunch of parts sitting in the corner of my office.

Sign In or Register to comment.