Rupture Disc Application? 7

[rcontractor]

I am looking into using a rupture disc in a simple system (1/8" Stainless Steel tubing) to protect a very expensive pressure transducer. For expample, the guage is used upto 300psi, but it will break at 330, and so i need the rupture disc to go off if the pressure gets to about 320psi but not before 300psi.
Am I barking up the wrong tree, or is this a feasible application?
PS. It gets easier, because the 1000psi transducer breaks at 1100psi, and so the disc shoudl blow between 1030 and about 1090...

Any ideas?

Thanks,
rcontractor

 

[TD2K]

It's possible but you'll need to talk to your rupture disk vendor.

If the disk is set to go at 320 psig and you are operating at 300 psig, that is nearly 94% of set point, close for a rupture disk I believe (fatigue due to pressure cycling is another question to raise with them).

If you want the disk in service, set at a nominal 320 psig, to not fail above 330 psig, you need an accuracy of 3%. Again, that is doeable but I believe tighter than the standard deviation they will quote.

There's a couple of people here who are much better in this area than I am, hopefully they will add their opinions.

You've lost me on the second part with the transducer, is this another rupture disk on an associated system?

 

[rcontractor]

Thanks TD2K!
Thats what I thought in terms of the fatigue life of the disc. I will be in contact with a vendor later today and see what they have to say as well.
The second part comes out of nowhere, I apologize. I need to size several rupture discs for several different transducers. Each transducer is used separately in the system and then replaced with a higher one to achieve more accurate calibration. I hope this makes sense.

 

[ilan7]

Try with the vendors, asking for strick tolerance disc at tight tolerance range of -0% +5% or better.
Recommanded: Tyco (Marston), Bs&B, Continental Disc, OSECO, FIKE Co.

 

[saxon]

rcontractor, This question may be out in left field, but why are you using pressure transducers in applications that are so close to their failure limit? Wouldn't it be better to get properly spec'd transducers that don't require a pressure relief? They have got to be cheaper in the long run and you don't have to constantly shut down the process to change the discs.

Remember, the accuracy of the guage or transducer is best at the midrange of its scale. Operating so close to its failure point is going to give inaccurate readings.

saxon

 

[rcontractor]

saxon,
A very valid observation, and one that I had myself once they started me on this mini-project. However, this process has been running long before I got here and the people here are very qualified. I will ask this question to them, as it puzzles me too, but I think it's already been dealt with....

thank you

 

[JAlton]

Here's what you can do to protect transducers. Use a pressure limiting valve (Chemiquip by Dresser) upstream of the transducer. For a 300 psi transducer, Set the pressure limiting valve to close at 275 psi. The down stream of the pressue limiting valve, but up stream of the transducer, place a relief valve (kunkle, parker, swagelok)which will open at 275 psi should the pressure limiting valve ever leak. As saxon says, never use a transducer so close to its max range. The advantage of the pressure limiting valve is that it can be rated at up to 5000 psi, but set to close at 300 psi. So you could leave the 300 psi transducer in line and install a 1000 psi transducer with its own pressure limiting valve and pressure relief valve set for 950 psi. This would eliminate shutting down to change tranducers or rupture discs. We use this technique on our Test Equipment.

J. Alton Cox

http://www.delucatest.com

 

[CHD01]

Yes, rupture disks are out of the question.

For an 1/8" Union Type safety Head you cannot get a burst pressure as low as 330 psig. You might get a 1000 psi disk; but then your problem will be the limitation that operating presssure not exceed 90% or 95% of the stamped (marked) burst pressure.

If you go to other type rupture disks even with a 0% Manufacturing Range, the 5% Burst Disk Tolerance the code allows will cause you to exceed desired range.

For you first example assuming a 315 psig stamped burst pressure:

MAWP = 330 Psig, with a stamped burst pressure of 315psig with a 0% manufacturing range = Stamped Burst Pressure of 315. Accounting for +/-5% TOLERANCE (DIFFERENT FROM MFG RANGE) = 315 +/- 15.75 psi or an allowable actual burst pressure range (versus being stamped at 315 psig) for the disk of 299.25 psig to 330.75 psig. This is unaccepatble and any other selected stamped burst pressure is even further outside your spec on either the l;ow or the high side!



Seek another way.

The more you learn, the less you are certain of.

 

[CHD01]

rcontractor:

By the way, your spec seem very ODD! I interpret your memo to say that the MAX NORMAL operating pressure is 300 psig and that the transducer actually fails at 330 psig.

Normally we design vessels and other equipment to have an allowable overpressure above its design pressure (or MAWP) for relieving purposes.

For example, if the MAX NORMAL operating pressure was to be 300 psig and you wanted to design a vessel for this you would proceed as follows:

300 / 0.9 = 333.3 psig = min design pressure (normally I would round this up but we'll not do this here). The 0.9 allows a min margin between relief device setpoint and desing pressure of vessel.

333.3 x 1.1 = 366.67 = allowable overpressure at required relieving rate (this can be as high as 1.21 normally except for occasional overpressures in piping. The relief device cannot be set higher than 333.3 psig in this example = Design Pressure or MAWP.

The Vessel is then hydro'd, older vessels were hydro'd at 1.5 the design pressure (it can now be reduced to I believe 1.3); plus the ASME Code has a built in safety factor of about 4, so the actual FAILURE pressure for the tank would approximate (lets use 1.5 ratio):

Hydro tested at 333.3 x 1.5 = 499.95 psig

Failure pressure = 333.3 x 4 = 1333.2 psig; we take no credit for this at all, it just gives you an idea where the actual catastrophic failure pressure may be without a finite elelment analysis.

The more you learn, the less you are certain of.

 

[rcontractor]

I just posted this to get some feedback on the quality of Rupture Discs for such a close tolerance of pressure....
The system I am dealing with is a calibration system with MAWP way above what is being dealt with. The Quartz pressure transducers are most accurate at their maximum range and that is why they are used that high. The only concern I have is that there is only a 30-40psi window within which the rupture disc should burst.

My inclination at this point is to source out a system similar to that suggested by JAlton and get it as close to the operating pressure we require as possible.

I apologize for not being more descriptive in my first thread, and thank you for all your suggestions and advice...

 

[zdas04]

rcontractor,
Have you considered a solenoid-operated slam-valve that is driven off your quartz transducer? Two valves (one to block the process and one to vent the transducer) would be reasonably inexpensive in 1/8" service and can be set to be very accurate. Neither rupture-disks nor PSV's will give you the precision that it looks like this application requires. I'd put in a spec-break.

David

 

[rcontractor]

David,
That sounds like a really good idea. However, I dont' quite know what you mean by "put in a spec-break." Also, if you have any vendors you would recommend, I would much appreciate that.
Thank you for your advice.
Ras

 

[JAlton]

The Pressure limiting system I suggested is similar to what zdas04 suggests. The limiting valve closes, but if it should fail, the PSV still protects the transducer. zdas04 is correct though about the precision. His suggestion of quick acting solenoids to close off source and vent transducer simultaneously is similar to how we protect against hydraulic failure of our Test Equipment Clamping System. If we have a PSV on a Test Bench held in place by hydraulics, their are solenoids set to close the Test Connection Isolation Valve and Vent the Test Plate should the hydraulic equipment fail. It is fast and accurate. I think zdas04 has a good plan.

J. Alton Cox

http://www.delucatest.com

 

[zdas04]

Ras,
A "spec break" is simply a break between one specification (generally either temperature or pressure rating) and another. If you put a slam valve between a pressure source and a lower-rated device it is generally called a "spec break". Sorry about lapsing into short-speak.

I'm just a flangehead. When I need to pump something as small as an electron, I always call the I&E guys and they speak very slowly to me and we get what we need. Can't tell you what brand of solenoid valves they're using this month.

David

 

[zdas04]

Ras,
I was talking to a vendor on Friday about one of my projects and we got to talking about solenoid valves. Your problem came up and he suggested a 3-way solenoid valve (to be able to slam and vent with one valve). He said Peter-Paul or Ashcroft both have excellent 3-way valves for around $US90.

David

 

[FLUIDM]

RCONTRACTOR, Before you decide about the rupture disc you are going to use, why not give a thought to Buckle Pin (Rupture Pin)? See the following link. A lot of problems, already discussed here in the forum & which may come up while you are using rupture disc, are addressed by Buckle Pin.

I won't write much because the website explains a lot itself

http://www.burnspressuresystems.com/Products/Rupturepin/rupturepin.htm

Your opinion shall be appreciated.

Regards,

FLUIDM

 

[JAlton]

I agree with the FLUIDM, the Rupture Pin Technology is a good idea. I have been involved from the Standards Committee aspect with this technology for years. It is a very reliable method of overpressure protection. I would recommend it.

J. Alton Cox

http://www.delucatest.com

 

[CHD01]

While the buckling pin device is better than the rupture disk in terms of reliability in my opinion, you will pay for this advantage. Without detailed testing the pin is still limited to the same standards as the rupture disk in this case(that being limiting operating pressure to 95% of buckling load and the additional limitation that the actual failure load be within +/- 5% of set load).

The more you learn, the less you are certain of.

 

[209larry]

rcontractor,

Maybe you should change your transducer. Check what is the signal output and inquire from other pressure transducer vendor. You might find a specification that matches your requirement. Maybe a pressure transmitter can do the job.

rgds,

209larry