Water vapor condenser 1

[Lukezo]

Hi,

I have to heat 250L/min water from 20degC to 50degC by using superheated steam (150°C, 3Bar), mass flow rate by demand.

The question is, which heat exchanger is the best choice.

Thanks!

 

[Latexman]

Can you sparge the steam into the water? No heat exchanger needed!

Good Luck,
Latexman

 

[Lukezo]

Condensate is needed in a next process.

 

[TiCl4]

I assume you mean 3 barg, which means you are within 12F or so of saturated temp. So the superheat won't be much of an issue - you'll quickly become saturated inside a heat exchanger.

Shell and tube heat exchangers are often used for steam heating, with steam on the shell side to allow for good disengagement of the condensate from the heat transfer area. Condensate is collected via steam traps and can be pumped with condensate pumps to wherever you need it.

 

[Caloooomi]

I agree with shell and tube. Would be able to get it in a fairly small unit to be honest. I sized it as a BEU style unit, plain tubes, ended up with 3.5 m2 SA (lots of overdesign) and allowing for the outer tube rows to be removed under / above shellside nozzles.

Actual type would depend on if you want cleaning lanes etc. Look at surface condensers for similar, larger, units.

While compact units are probably more attractive, it would depend on how many thermal cycles you are considering for the process too.

 

[Lukezo]

Is it ok just to assume LMTD with Tcond on vapor side

 

[TiCl4]

Lukezo,

Only if the condensate outlet temp is close to the saturated steam temp. If you have any significant cooling of the condensate, the analysis of the heat exchanger is typically split into two sections: the condensing section with constant hot side temp and the condensate section with variable temperature.

Edit: Unless the heat exchanger is operated partially flooded, I assume that disengagement of the condensate is usually good enough that assuming a constant hot side temperature is okay. In reality the condensate will cool a bit even in a steam-only HX as it slides along the heat transfer area. You will use more steam than running partially flooded, but I have not seen any steam heat exchangers operate partially flooded. All the ones I’ve seen have all operated with no condensate retention Perhaps others could explain why.

 

[Lukezo]

Which is best method for accounting phase change in shell and tube heat exchanger?

 

[TiCl4]

Maybe I didn’t explain clearly above. If the phase change does not include a significant cooling of the condensate, the HX may be treated as a constant temp on the condensing side.

If the condensate (or vapor, if designing a vaporizer), changes temp significantly, then you need to analyze the the HX in two sections - the one with a condensing vapor and the one with a cooling condensate. You treat each section as an individual heat exchanger, with the outlet conditions of the condensing section as the inlet conditions of the normal HX section. So the first section has a constant hot side temp for condensing service, and the second section has a normal LMTD curve.

 

[Lukezo]

Can someone please ellaborate on sparging process?

 

[EdStainless]

Companies like Pick make direct steam injection systems, these are combination diffusers and control devices.
A lot of commercial food is cooked this way.
If you build a power plant to run at one power level (or a very narrow range) you can use these for feedwater re-heating in place of heat exchangers.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[goutam_freelance]

Assuming you are using potable water quality to be heated by steam. Boiler quality water is of costly high purity and should be recycled to boiler for economy. So shell and tube type HEX is recommended where condensate is collected separately. In addition since the saturation temperature (133.5 deg C ) at 3 bar is quite different from the water temperature range of 20-50 deg C, there will be significant thermal expansion difference between tube side and shell side. To take care of this, U-tube type HEX is recommended as is normally done for power plant feed water heaters.

 

[Lukezo]

For heat flux 2,5MW I tried to calculate kA and received enorm numbers for shell and tube (length of a tube 20m)


I predicted 1 tube pass with V_water=2m/s (ht=4600W/m2K)

Resulting length is 18m per tube with 25 tubes

water MFR=15L/s
Tout=60degC
Tin=20degC

Can I lower tube ID? I never have dealt with U-tube, can you please give me a hint?

 

[goutam_freelance]

I am not sure about your design method. I give below some data from a thermal power plant feed water heater.

Heat Load: 41 MW
Steam Press: 5.65 bar
Steam temp: 290.5 deg C (Sat temp-155 deg C)
U Tube OD: 19.05 mm
U Tube length: 10500 mm (avg half length)
No of U tubes: 966

You may have a rough comparison.

 

[EdStainless]

feedwater heaters often end up over 40' long with utubes in them.
Because of the high pressures they are easier to build small diameter and longer.
These have 3 distinct zones, de-superheating, saturated steam, flooded condensate subcooling.


= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[goutam_freelance]

All feedwater heaters are not high pressure. There are LP feedwater heaters also which operate under vacuum in steam side. LP feed water heaters mostly do not have desuperheating zones. The last heater under condenser neck have neither desuperheating zone nor drain cooling zone.

 

[TBP]

Lukezo - If you're looking to fully understand the method for calculating a heat exchanger for your own interest, that's commendable. But unless you're intending to do this for a living, it's a lot of work for limited return. Also, I'm pretty sure that you'll wind-up with a custom unit that will be hideously expensive, along with a delivery date that will make you wince.

If you're just looking for a unit that will do the job, I'd give the water flow rate, inlet & outlet temps, along with the steam pressure available at the HX inlet to a manufacturer or their vendor, and let them do the heavy lifting. They'll have software (that they won't share with you) that will spit out a range of suitable standard units they offer in a few minutes, along with the pressure drops across their units.

 

[georgeverghese]

Agreed, read up DQ Kern on sizing shell and tube HX in condensing service if you want to learn how this is done manually. You can also send out an RFQ to HX manufacturers with your filled TEMA datasheet stating process and other mechanical requirements.
Presume this query is related to your other recent post on computing mean tube wall temp. This computation for Tw is also in DQ Kern.

 

[Lukezo]

Hi,

we chose to inject vapor directly in the tube. Which mixing valve do you suggest, since the needed mixing temperature deviation is +-5°C.

feedwater
MFR=1200kg/min
T_in=20...40°C (range)

heating vapor MFR=100kg/min
2,5BarG
T=145°C

T_out=65 (+-5) °C

MFR's are correct, I just need a cheaper sollution.

 

[MJCronin]

Luke ....

You have come to your senses.... "Condensate was not needed in a next process"

Get a pick heater ....simple and easy to control

https://www.pickheaters.com/

(Be careful though .... they are very loud)

MJCronin
Sr. Process Engineer