Tell me why this design wont work

[xahvres]

Hello everyone. My first real post here! (and I hope its in the right place...)

I've been thinking about a simple LM setup and I've come up with something that seems too easy, so there must be something I'm missing, but since most the LM talk here is about Bokas and this is different, I havent really found an answer. It is very similar to the large fractioning columns they use in oil refining, as far as I know.

I did a quick sketch of what I'm thinking about:

IMG_20220609_135800.jpg

How it works: 100% of the vapour coming off the column is condensed after a U bend with a vertical condenser. A "T" and gravity makes sure the condensate hits the needle valve first. The needle valve sets the amount of liquid to go back into the still. The reflux pipe has a U-trap shape (marked "A" on the diagram), to insulate against the slight pressure from the still. If the reflux is less than 100%, then the small space above the valve fills up, and the overflow comes out as the offtake.

Seems very simple, too simple. The fact that we regulate the reflux (and it being constant if we dont move the valve) and not the takeoff eliminates one of the big problems with LM systems, the transition from hearts to tails. It only needs one condenser, and you dont even need to really regulate it.


The only pitfall I could see is if the pressure in the still is larger than I thought, then the U-trap would need to be very large and hold a lot of liquid.

So, anyone has tried something similar? Is there something basic I'm forgetting? What are your thoughts?