Heat Exchanger

Tony Conner

...is sub-cooled condensate. In the case of grossly oversized shell & tube HXs, they wind up running mostly (or totally) flooded with condensate. The ones I saw had gauge glasses mounted on them so you could see the condensate level in the HXs. In the year and half I was there, I NEVER saw a level because the HXs always ran totally flooded.
The shell & coil HXs were mounted vertically, with the steam in the coils, and the loop water in the shells. The installation was specifically designed to sub-cool the condensate before it was sewered, so every last BTU that was worth getting could be wrung out of it before it hit the drain. There was a signal from the condensate line that would over-ride the other control points and shut down the show before shooting 125 PSIG steam into the drains. I'm not aware that it's ever had to come into play.

The only problem encountered was that they installed ONE control valve for THREE HXs in parallel. This is the same as one trap on multiple points of use. The valve was oversized most of the time, hunted wildly causing no end of operational problems, and wore out long before it's time. All this in addition to being very expensive up front. Once we got rid of the single 4" monster and installed a 3/4" valve on each HX, everything worked great.

z

Heat Exchanger Help

I have inherited a heat exchanger (steam, to make hot water) as part of a laboratory experiment. In addition to poor piping practices (no swing joints) the heat exchanger (technically, a condenser) is oversized.

The water it produces is too hot, and the steam pressure (maintained by a regulator) drops with very low water flow rates - the regulator cannot keep up.

Rather than install a new regulator, I want to reduce the heat exchanger area by flooding half of the condenser. The proper way to do this is to hang a tank next to the exchanger, pipe the condensate from the exchanger to the bottom of the tank, pipe an equalizer line to the top of the tank (the steam pressure in the exchanger is supposed to be 10 psi), then pipe a condensate from the bottom of the tank through a pneumatic valve. Then, use a level controller to modulate the valve to keep the exchanger half flooded. (Is this understandable?).

Anyway, being low on budget and not needing to be able to vary the level in tank, I could (possibly) do without the controller and the pneumatic valve, which are sort of expensive. My thought is to simply pipe the drain line from the tank from the side (half way up - sort of like the way a hartford loop is done) and run this drain line through a trap (which I already have). Under normal operation, condensate would come half way up, run through the opening in the side of the tank, down the pipe and out through the trap.

Will this work? Will it hammer like mad or do you expect that it would be OK.

Thanks for any input and advice.

z

Jerry Boulanger

I don't think your regulator is too small

I think it is too big. At low flow/load, it is doing what it is supposed to do - reducing the steam flow and pressure. It is important to correctly size control devices for steam. I suggest that you get someone who sells steam control valves to size one for the load you have.

If you want to reduce the heat exchanger capacity, plug some tubes. It's easier and cheaper than flooding the shell, and won't create any other problems.

Tony Conner

Actually...

...the correct way to run a flooded HX is to remove the steam trap, and install a control valve. This valve in effect becomes a variable orifice trap. You don't need a steam control valve with this arrangement. When the controls sense a rise in water temp, they throttle the "trap" closed. This floods more of the HX, effectively reducing it's surface area - like you were thinking of. On a drop in water temp, the "trap" opens up, allowing more condensate to flow out of the HX, increasing the surface area exposed to steam. I've seen this on traditional horizontally mounted shell & tube units as a retro-fit for the HX being grossly oversized, and on vertical shell & coil units as part of the initial installation. Both worked fine.

An added benefit of this type of control, is that in the many installations where there is a condensate lift after the steam trap. This typically causes no end of problems at low loads as the temp control valve throttles-in, dropping the steam pressure to the point where it can't push the condensate out of the HX. With the steam just ON, full bore, all the time, there's always enough pressure to force the condensate out of the HX.

JimGPE_3

So what happens

when the load increases to the point where the valve stays open and the HX is no longer flooded. Doesn't the condensate velocity increase to the point that water hammer insues? Similarly, won't this be a problem if the control valve were to hang up (open)?

Not being critical, just trying to learn something.

z

Thanks to all who replied.

Jerry, I realize that I wasn't clear when I stated the problem. The regulator is too small (or the exchanger is too large, depending on your politics). The steam pressure is OK at very low water flow rates, then drops to zero as I increase the water flow rate. But the flow rate where it drops is too low for our application. Having looked at the different options, I think I am going to plug a few tubes and not mess around with flooding the exchanger.

Jim, though I realize I asked the question in the first place, I can (I think) answer your question about hammer. Although using a control valve in place of a trap can work, you do run the risk of having the condensate valve full open, thus leading to hammer and discharge of live steam when the shell is no longer flooded. This can be avoided by proper tuning of the controller parameters and proper sizing of the condensate control valve. Using the condensate storage tank, as I described (with the level controller) eliminates that problem completely, but requires additional components. The condensate tank is taller than the diameter of the exchanger shell, so when the level is set so that the shell is not flooded, there is still liquid in the condensate tank and you aren't blowing live steam out the condensate line. There was an article in Chemical Engineerging (magazine), Sept 3, 1973 by Jimmy Mathur that gives a pretty good description of the various options for controlling steam heat exchangers.

Overall, the advantage of flooding versus plugging is that if you by using a control valve to flood the exchanger, you have a lot more range available for your process because the flooding level can be changed. Once you plug some tubes, you have limited the range of the exchanger. Since I don't need a huge range, plugging is OK.

Again, Thanks to all.