Testing and setting safety valves

[doubletop]

I have two safety valves, one I made and a commercial version and they both appear to leak even under a low pressure. As they both have 5/16 x 32 threads I have been connecting them to a mountain bike suspension pump, which has a schrader connector (also 5/16 x 32)

This pump will go to 300psi. It conveniently has a release button so the pressure can be dropped easily.

With one of the safety valve installed and pumping to say 50 psi the pressure will leak away to near zero in seconds. Putting a blanking plug into the pump I can pressurise the pump to any pressure and it will hold, so the the pump isn't leaking. Screwing the safety valve right down so the spring is full compressed achieves the same thing. (no doubt not recommended practice though)

I am assuming that both these valves aren't working properly and any pressure should hold until it is pumped up to exceed the preset value and the valves should dump their pressure. Or do I have this wrong? On a working valve should the pressure drop be gradual, say back to the set pressure, or a step function to drop the pressure to near zero?

Short of sticking the valve on a boiler with a suitable pressure gauge and firing it up what is the prescribed method of testing and setting these things?

 

[d.bick]

With very little volume of air in the pump any leak will be more noticeable. On a boiler in steam the leak may be no more than a whisp of steam.
The safety valve should however be steam tight and release at the working pressure and no more than 10% above with a good fire and blower full on. They often do not close until 10 or 20 P.S.I. under working pressure, unless are of the POP type.
You could try lapping the valve seat with a ball the same size (not the ball out of the safety valve) stuck on a rod with a little very fine cutting paste with finger pressure and a revolving motion.
Dave Bick

 

[Tin Falcon]

what happens if you fill the boiler with water then pressurize with water . will the valves hold water . steam is water not air if you are using saturated steam not super heat i would think this would be a more realistic test. The steam guys can correct me if i am wrong here.
Tin

 

[firebird]

Hi DT

I tested the valves on both my boilers with the hydraulic test set up. Both valves were water tight until the blow off pressure was reached when they started to leak water then sealed again as the pressure dropped. They work slightly differently under steam, i.e. they blow off at a slightly lower pressure but that's easily adjusted. I can only assume that the ball isn't seating well enough on yours. I seated my balls (no silly comments please) by using a spare ball and giving it a tap with a punch onto its seat then refitting the working ball. Once steam has been reached it is then maintained with the safety valves just fizzing away nicely.

Cheers

Rich

 

[doubletop]

Some good clues in here thanks guys

Both valves were water tight until the blow off pressure was reached when they started to leak water then sealed again as the pressure dropped. They work slightly differently under steam, i.e. they blow off at a slightly lower pressure but that's easily adjusted. I can only assume that the ball isn't seating well enough on yours. I seated my balls (no silly comments please) by using a spare ball and giving it a tap with a punch onto its seat then refitting the working ball. Once steam has been reached it is then maintained with the safety valves just fizzing away nicely.

I've been tapping balls, to the extent of giving them a seriously good thump. I've made 'D' bits to re-form the seat and purchased Torlon balls but was never convinced I had it right. In fact I think my earlier version worked better than my current version. I do now recall that there was something on this subject in your small boiler thread. This has useful input from the experts. http://www.homemodelenginemachinist.com/index.php?topic=2092.195

what happens if you fill the boiler with water then pressurize with water . will the valves hold water . steam is water not air if you are using saturated steam not super heat i would think this would be a more realistic test. The steam guys can correct me if i am wrong here.

This a good point so I went out and filled the pump with water and used the commercial valve after giving the ball a good tap to re-seat it, as I hadn't done it with this one. Maintaining a constant pressure on the plunger had the valve dripping water slightly but pressure was maintained. If I dropped the pressure on the plunger it would leak away. However, when I increased the pressure to 100psi the valve dumped all the water in a gush and the pressure dropped back to zero. This one looks like it behaving itself now I've given it a beating.

With very little volume of air in the pump any leak will be more noticeable. On a boiler in steam the leak may be no more than a wisp of steam.

The safety valve should however be steam tight and release at the working pressure and no more than 10% above with a good fire and blower full on. They often do not close until 10 or 20 P.S.I. under working pressure, unless are of the POP type.
You could try lapping the valve seat with a ball the same size (not the ball out of the safety valve) stuck on a rod with a little very fine cutting paste with finger pressure and a revolving motion.

This is useful so I now understand what should happen. I don't have any cutting paste available I can't try that right now

I assume that as its a spring holding the ball down it can't keep the ball completely seated whatever the pressure so there must always be a leak of some sort once the gap gets bigger than the size of water (steam) molecules. I do have copy of Tubal Cains Model Engineers Handbook and there is a section in there on spring design and a safety valve spring is used as an example. I'll do some theoretical tests on the dimensions of my two valves and see if I can reverse engineer the expected pressure ranges.

Off to do more reading, calculating and testing............

 

[Tin Falcon]

I don't have any cutting paste available I can't try that right now

tooth paste can be used for lapping
Think you may have some on hand
Tin

 

[doubletop]

tooth paste can be used for lapping
Think you may have some on hand
Tin

That's why were here; lateral thinking, and the best ideas not being the so obvious.

Thanks for that.

(does the fluoride help with de-zincification as well?)

 

[agr]

Another thing that might be worth considering is that air and hydraulic testing will also differ from "real life" operation due to the fact that in use they will obviously be hot, so differential expansion of material, etc may have an effect.

I have had this with a full size injector check valve which was professionally reconditioned, tested and certified. It obviously worked fine in a hydraulic/Nitrogen test tank but when in use on the boiler, will constantly dripple water once its hot.

However in my opinion, at the end of the day, a 'dribble' of water, or wisp of steam, as long as it is not significant, adds the the character of a living, breathing steam plant - after all, we all have our own dribbles and wisps of gas from time to time ;D

Tony.