We have been assessing some Spray Desuperheater valves from an HRSG that are having unreliability problems. The operator is experiencing problems with leakage which are caused from damage to the plug and nozzle seats. The units are SD2K and supplied from Copes Vulcan. The steam temperature where the DSP valves spray into is close to 500°C. The feed water temperature pressure around 170 bar and 150°C. We have assessed a number of these DSP valves that have been in service for approximately 12 months. The failure are caused by metal fragments from the spray nozzles disintegrating inside the cage and then fragments getting trapped between the plug and cage causing damaging the seats. The construction of the spray nozzles is very odd. It consists of ring with the centre spray nozzle piece attachment by interference fit with no locking attachment. This appears to work loose in operation and eventually fragments get ejected which damage the internal seats resulting in the valves leaking. Another odd feature of the design of these superheater valves is that the stem and plug are manufactured from 17-4PH alloy and I don't think this alloy is suitable for the working temperature of 500 deg C. The material is heavily degraded even after 12 months in service. At this stage, I don't know why the spray nozzles are working loose in service and what is the root cause. Has anyone experienced a similar failure?
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Desuperheater Spray Valves 1
- Thread starter Metaljon
- Start date
- Thread starter
- #2
I have added a photograph showing the nozzles in various states of failure.
- Thread starter
- #3
Photograph shows damage to plug and seal caused by debris.
-
1
- #4
The damage looks like cavitation damage to me. I once had trouble with a CV desuperheater of a different style where the water flashed to steam in the body of the DSH due to the high temperature of the metal parts it was exposed to prior to exiting the distrubution nozzles into the flow stream. This was a large in-line DSH and it literally sounded like a worker had been left in the pipe and was trying to alert someone with a sledge hammer. The noise was horrific.
As the water entered the chamber that distributed it to the nozzles, it heated up and boiled before it could get to the nozzles. Then because the steam created couldn't flow through the nozzles fast enough, the downstream temp increased causing the control valve to pass more water to the DSH. When the additional (and colder) water pressured up the distribution chamber, the steam in there collapsed causing the cavitation hammering. Once things settled down, the process would repeat itself.
The problem was rectified when the DSH was modified so that the distribution nozzle cross sectional flow area was closed down (went from one large nozzle to multiple smaller nozzles at a greatly overall cross sectional area) and a higher pressure drop was taken across the nozzles (instead of taking a large pressure drop in the upstream water flow control valve). This kept the water in the "chamber" waiting to pass to and through the nozzles into the steam flow stream at a higher pressure and a press/temp above its saturation pressure/temp.
Your pictures look to me like something similar is going on here.
Check the press/temp relationships of the water with respect to the metal temperatures of the nozzles that have to contain it.
rmw
As the water entered the chamber that distributed it to the nozzles, it heated up and boiled before it could get to the nozzles. Then because the steam created couldn't flow through the nozzles fast enough, the downstream temp increased causing the control valve to pass more water to the DSH. When the additional (and colder) water pressured up the distribution chamber, the steam in there collapsed causing the cavitation hammering. Once things settled down, the process would repeat itself.
The problem was rectified when the DSH was modified so that the distribution nozzle cross sectional flow area was closed down (went from one large nozzle to multiple smaller nozzles at a greatly overall cross sectional area) and a higher pressure drop was taken across the nozzles (instead of taking a large pressure drop in the upstream water flow control valve). This kept the water in the "chamber" waiting to pass to and through the nozzles into the steam flow stream at a higher pressure and a press/temp above its saturation pressure/temp.
Your pictures look to me like something similar is going on here.
Check the press/temp relationships of the water with respect to the metal temperatures of the nozzles that have to contain it.
rmw
- Status
Similar threads
- Locked
- Question