Steam Condenser - Plate Type???

[novakoni]

Hi to all,

Project I am working on involves condensing steam turbine with steam extraction points.
Bleeds from turbine should be used for pre-heating of the condensate in 3 stages (3 LP pre-heaters/condensers) - by mean of steam condensation an sub-cooling of condensate.
In first stage steam is at vacuum and in second 2 at low under-pressure.

Steam condensate from upstream pre-heater shall be removed and introduced to steam space of downstream pre-heater.

Level of condensate (at steam side) in HE is controlled by control valve.

My question would be - is it possible to use fully welded plate heat exchanger for this application??? (until now I was always using horizontal shell&tube type)
considerations:
- level control
- sub-cooling
- mixing of steam and condensate at steam inlet side
- inert gasses removal
- thermal stresses and forces from connecting pipes
- etc

Thanks in advance

 

[katmar]

I have not used plate exchangers on turbines, but I have used them as regular condensers and also as reboilers (condensing steam) and I see no reason why they could not be used for your application. Speak to the suppliers like Alfa Laval. They will not offer you an exchanger if they are not confident that it will work.

Katmar Software - AioFlo Pipe Hydraulics

http://katmarsoftware.com

"An undefined problem has an infinite number of solutions"

 

[EdStainless]

One trick may be the handling of heat transfer in each zone.
Will you need separate plate HX for desuperheat, condensing, and subcooling?

= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[racookpe1978]

I have only seen shell and tube in power plants, but don't know if that due to "inertia" of
"that's what everybody else has always done"
or
" plate heat exchangers were not yet invented when this thing was designed"
or
"the shell-and-tube-hx is the cheapest and plant space is not much of a limit"
or
"because of the differential pressure possible across the hx systems during startup and shutdown ops (or during some accident mode!) would warp the vertical plates, but that maximum dp can be withstood by a shell-and-tube heat exchanger."

 

[novakoni]

Problems I see with plate type are:
- volume - because this application requires level control of condensate - volume should be sufficient and where to measure level. Maybe to install volume buffer vessel, but where to connect this vessel?
- again volume - as it is really important to protect back flow of water to steam turbine - in case of water leakage to steam space. Usually there is isolating valve at turbine bleed outlet - but closing time of this valve shall be less than the filling time of heat exchanger.
- I would like to use only one HE - both for condensation and sub-cooling.
- Steam coming from steam bleeds is not completely dry - protection shall be considered at steam inlet.
- I have to introduce condensate from downstream condenser to steam space - where this connection should be installed?

 

[SAA251]

fully welded heat exchangers can easily do condensing and sub-cooling
there will be a control valve at condensate outlet which will control the liq level in the exchanger
try to talk to any vendors, im sure they will give usefull advice
here is an example that I found in Alfa Laval site :

http://www.alfalaval.com/industries/power/downloads/Documents/PEE00213EN.pdf

 

[novakoni]

Hi Sally,

I have contacted Alfa Laval - and actually I am waiting for quotation from them.

Looks like this type I need is not completely standard - so they have contacted central in Sweden. And it already takes few weeks to make offer.

I also think it's possible to use fully welded plate type - but with some additions in order to get more volume and better condensate level control.

I could inform the result if anyone is interested.

 

[FredRosse]

There have been many applications of plate HX condensing steam. In the most simple form, a steam trap at the outlet will do OK. More sophisticated level controls would be needed if subcooling the condensate, and the high d(Level)/d(Liquid Volume) character of a plate HX can present control problems, but can be made stable with proper controls tuning.

 

[davefitz]

I would think the velocities of the steam at vaccuum would be too high with a plate type design

"Whom the gods would destroy, they first make mad "

 

[hacksaw]

appart from the piping complexity of a plate exchanger how do you plan to handle leakage?

 

[FredRosse]

"My question would be - is it possible to use fully welded plate heat exchanger for this application??? (until now I was always using horizontal shell&tube type)
considerations:"

some more detailed answers:

- level control ANS: Possible but more challenging than typical shell and tube type for level control stability. If condensing zone outlet reservoir is fitted, this problem is no worse than conventional shell and tube, and can be made even more stable than conventional shell and tube with larger reservoir.
- sub-cooling ANS: Will function with limited amount of sub-cooling within a single plate HX, however if water level maintained above the drain nozzle, then continuous venting not achieved.
- mixing of steam and condensate at steam inlet side ANS: Cascading drain can be mixed with inlet steam flow, this precludes a de-superheat zone in the heater, but probably not a requirement here since this is a lower efficiency cycle arrangement (since it only has 3 heaters). If separate drain cooling plate heat exchanger used, with intermediate condensing zone outlet reservoir, then cascaded drain can enter intermediate reservoir.
- inert gasses removal ANS: If using a plate HX for condensing without condenser zone outlet reservoir, venting cannot be achieved unless condensate level is maintained mid-way between outlet level top of pipe and bottom of pipe. In this case continuous vent can be mounted telescoped within drain pipe. Drain pipe must operate only partially full of liquid. If condensing zone outlet reservoir is fitted, this allows continuous venting from the reservoir.
- thermal stresses and forces from connecting pipes ANS this design criteria is the same for plate HX or S&T HX, must design piping to meet allowable nozzle loads in either case, and Plate heater MFG defines standard allowable forces and moments, or can provide plate HX special design to meet purchaser's specified forces and moments.
- etc ANS: While condensing plate HX has been done many times, there are applications (such as utility feedwater heaters) where it is more prudent to use conventional S&T HX. To get typical feedwater heater performance from plate HX, the intermediate reservoir, two plate HX connected in series (condensing plus drain cooling sections), plus being able to withstand feedwater pressure add up to complexities that are better served by conventional S&T HX. For small applications the welded or brazed plate HX may appear to cost less, but complexities mentioned above quickly add up to it being more prudent to use S&T HX.

I could see for a simple small plant, with elimination of drain cooling, and simple steam traps (with air vents) cascading the condensate, the simple plate HX system might be used, as long as the maximum feedwater pressure is reasonably within the pressure allowed for the plate HX. A steam cycle with main steam flow well under a few tonnes per hour, but not for steam cycles much bigger than that. However steam cycles this small are no longer made typically, although I have some pictures of plate HX in condensing service on a steamboat.