Using venturi principle to drive flow through sidestream HX

[MinnesotaSlinger]

I am looking to run a small fraction of a flow (say, 10% of 200 gpm) through a heat exchanger. The HX would be selected with calculated venturi D/P in mind. The HX would be for cooling condensate from a steam-to-water HX, which will operate at about 50 psig and therefore have condensate coming out well above the boiling point when going to a tank at atmospheric pressure. Ahead of the steam-to-water HX, I was thinking of creating a sidestream that could be powered by a circulator pump, but it would seem almost simpler to just drive flow using the venturi that connects the inlet to 3" piping and the outlet to a 2" section downstream, which should give me about 2 psid to drive flow through the HX tubes. (Doing this would also keep flow proportional through both HXs.) If actual flow through the HX varies somewhat from calculated, I don't think it would be a problem given the lack of need for precision. The only thing I'd be concerned about is erosion in the narrow section, ensuring D/P is strong enough to drive require flow to cool condensate as desired (i.e., below 212 F), and ensuring D/P is not so strong as to drive too much flow through the condensate cooling HX as to cause tube erosion. Is this a bad idea? Should I just go with a circulator pump for the sidestream?

 

[Latexman]

A sketch would sure help!

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[MinnesotaSlinger]

I don't have a place to save a sketch online. Basically though, I'd have a 3" schedule 40 pipe with 200 gpm flowing through it. It's downstream of a strainer that would serve both HXs. This line has a tee in it from which 20 gpm will flow out. The other 180 gpm continues on and reduces from 3" to 2". In the 2" section, there's another tee from which the 20 gpm from the small condensate cooling HX returns to the main line flow. This 200 gpm 2" line then expands back to 3". There should be about 1.5-2 psid of difference between the two tees that would drive flow through the small HX.

 

[Latexman]

Read the fine print under the Attachment line, especially "Need file storage? Try SugarSync Free! …or upload your file to ENGINEERING.com".

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.