Control of Steam Heaters 2

[ixchawla]

We have been using steam to heat the well fluid and water using plate exchangers. The duty is controlled by throttling the steam pressure at inlet to exchangers. Due to gasket temperature limitations, the maximum steam side operating pressure is limited to slightly above atmospheric. During turndown, however, the steam pressure required to maintain the duty is 0.2 bara. We have pumps to route the condensate, so technically there should not be a concern. However, operationally, there is likelyhood of air ingress into the system, which cannot be vented as the pressure is sub-atmospheric. The project is still in design phase and we are reviewing for its suitability.

Recently, vendor advised that a vacuum relief device should be installed on condensate drum. This would pull air during low duty requirements. We are concerned about this design, and like to clarify the following issues:

1. Since the system is normally operating close to atmospheric, any drop in heating demand would pull the air into the system. The air being low temperature, the immediate effect will be lowering the heating media temperature and thus outlet temperature of cold side outlet will drop. The temperature indication for control is provided on the outlet of cold side outlet to maipulate the steam control valve at inlet to exchanger. Due to inherent sluggish nature of temperature controls, it is likely that the cooling side outlet temperature will remain low for some time.

2. The effect of air ingress is blocking some tubes for heat transfer. When the heating requirement increases, the air needs to be vented. During turndown flow, the steam velocity will also be low, which may not adequately vent the excess air (vent located on condensate pot). This would result in increase in steam pressure until the heat duty is met. Likely steam temperatures will be higher than without air thus affecting the gasket life. Usually in S&T exchangers, an air vent from exchangers is connected to the condensate pot, I am not sure if this can help in such scenario.

3. The air ingress and venting is an uncontrolled event, and is similar to a disturbance in the system. The temperature controls have to respond properly to stabilise the system. If the tuning and logics for temperature control are based on steam only, then operating under steam+air (with variable %), the controls may not work properly. The cold side outlet temperature is 80 deg C (other exchanger 60 deg C) with high and low temperature alarms at 90 (65) and 70 (55) deg C resp.

4. The basic control for steam throttling is the steam pressure control, which affects the LMTD and hence heat duty. However, in this scenario, the steam side will always be operating near atmospheric pressure irrespective of heat duty and cold side flow rate. So, this doesn't work as steam pressure control. A section of exchangers with higher steam concentration will be overheated, while another section with higher air concentration will be under-heated.

5. When steam pressure control fails, the condensate flooding is still possible in S&T exchangers. However, I am not sure how successfully it has been applied to plate exchangers, since the condensate level indication in exchanger cannot be direcly known and levels reaching the steam inlet nozzle can cause steam hammering. Surely, adeqate control will be required to augment heat duty to catch-up level variation, and avoiding filling the condensate side. Venting the inerts may be another concern.

 

[rzrbk]

I am not that familiar with plate exchangers; how feasible would it be to upgrade it to handle higher steam temperatures? A more stable control would be to throttle the condensate outlet to hold a condensate level in the exchanger, adjusting the area available for heat transfer. (plate exchanger configuration may not allow this)
This would mean a more expensive exchanger, but if the steam pressure could be raised high enough, it could save the need for pumps.

If the allowable temp cannot be raised you could do the same using a pressure control on the inlet steam to keep it just above atomospheric, and a level control on the pumps' discharge. (again, if the plate exchanger configuration allows level control)

If none of the above is possible, and exchanger needs to operate in a vacuum, some sort of ejector system will be needed to remove non-condensibles/air. (similar to a surface condenser on a condensing turbine)

 

[RossABQ]

I'd recommend making hot water to heat the "well fluid". Make it with the steam and pump it to the HX. You can use steam as described but it really requires lots of chemical treatment to deal with the air, air removal (deaerator), and a lot of other $$ stuff. It's real easy to control heating with hot water.

 

[rmw]

rzrbk's idea of throttling the condensate is a good one. PHE's operate just as good with liquids as vapors, so the available area would vary just enough to admit the amount of steam needed and keep the pressure high enough to prevent air ingress.

You maintain your steam heat advantage and that is the large amount of heat transfer per surface area due to latent heat transfer. The control of the condensate level controls how much surface area is available.

rmw

 

[athomas236]

Gentlemen,

You may find this link useful

http://www.bellgossett.com/Press/BG-techtk699b.asp

regards,

athomas236

 

[Montemayor]

As far as informative and authoritative information on steam heater control(s) to be found on the Internet, Walter Driedger is THE MAN. For the answer to this control problem (& a lot of other equipment control problems) go to:

http://www.driedger.ca/index.html#home

I don't believe there is a better source of plain, well-written, and well-presented control information to be found.

 

[stanlsimon]

The reason they are asking for a vacuum breaker on the heat exchanger is to prevent condensate from hanging up in the system. You need some pressure to push the condensate through the traps. In order to have a subatmospheric condition in the condensate side of the system you would have to have a vacuum type condensate pump. These used to be common on older heating systems. They use circulating water and the venturi effect to suck the condensate back and push it into the boiler. At low pressures like this a centrifugal condensate pump cavitates and is useless.

If you have a vacuum breaker on the heat exchanger that will allow the condensate to build up some static pressure and push down through the trap. Make sure you leave at least 12" drop from the bottom of the HX to the inlet of the trap. Also oversize the trap(s) x 4 to allow a sufficient flow rate at this minimal pressure differtial.