Reduced port globe valve - Max K value?

[burtonwr]

Quick question regarding the Crane 410 manual and reduced port globe valves. Is there a limit to the resistance coefficient (K factor)? Using A-26, 27 I'm getting some very large K values and can find no examples anywhere of K values over 100. Also, if my K value is over 100, is my expansion factor (Y) limited to .718? Is my Dp/P1 also limited to .920?

Dealing with a choked flow conditions, determining max gas blowby rates in the event of an upset through a globe-style LCV. The globe valve is a 2" body with a 3/4" port...the numbers make sense as i'd expect a large pressure drop through the reduced port of the valve, but looking for more assurance.

thanks for any help!

 

[vpl]

Did you remember to multiply the K₂ value by the friction factor on the top of page A-26*.

I did a quick calc, following Example 4.20 (specifically step 4). I assumed your K₁ was 340 ft (the largest value for globe or angle vales), that d₁ was 0.75" and d₂ was 2". This gave me a K₂ of 319.8 ft, which I then multiplied by 0.019 from the "K" factor table, resulting in final K value of 6.06 ft for the valve.

*If your pipe isn't clean steel, you would need to go back to the graph on page A-24 and figure out the friction factor





Patricia Lougheed

Please see FAQ731-376 for tips on how to make the best use of the Eng-Tips Forums.

 

[burtonwr]

Patricia - Thanks for the reply. Are you sure that I need to multiply formula 7 for K2 by the friction factor? Referencing pgs A-26,27, I use the friction factor (Ft) to find K1, which is then used in the formula to find K2, my overall resistance coefficient for the valve. As a quick check, the value of 6.06 for K is less than the K value for a full port globe valve (beta = 1), which doesn't make sense as I'd expect the K value to be much higher for the reduced port. I may be missing something though so I appreciate the help.

 

[pleckner]

burtonwr:

Assuming I read the equation correctly (it is late for me here in New Jersey, USA), and assuming the full port valve with K=340ft, then I get a final K value (call it K2 if you wish) of 348.8. The reason you don't see high K values in any examples is simply Crane doesn't use any reduced port valves.

You would be better off obtaining the full open Cv of your valve and use the ISA control valve sizing equation to determine the actual pressure drop.

Be that it may, your expansion factor maxes out at 0.718 at K=15 (Cp/Cv = 1.3). As far as the pressure ratio goes, you can't do any better than 0.92; you're getting close to 1 any ways.

The better way to do this calcuation is to us the ISA control valve formula (as I state above) for the valve and use the Isothermal gas equation overall. But following CRANE TP40 Example 4-21 with my comments above should be OK.

 

[katmar]

burtonwr, one way to check the K value you are getting is to get the Cv for the reduced bore valve from the manufacturer and then use Crane 410 Eq 3-16 to convert it to a K value. In fact, I would be much more comfortable using this approach than the Crane A-26 Formula 7 approach.

Concerning your post of 23 Mar 07 16:29, I agree that you cannot get a K value for your reduced bore valve that is less than that for a full bore valve. How does your K₁ value using the equation K₁ = 340[&fnof;]_T compare with the data for your full bore valve? Using the second form of Equation 7 (i.e. not the one involving Formula 2 and Formula 4) it is impossible to have K₂ < K₁.

I am sure that Patricia made a simply typo when she reminded you to multiply K₂ by the friction factor. The only place where the friction factor comes into play is in the equation K₁ = 340[&fnof;]_T.

But Patricia's comment brings up an important point. You should NOT use the graph on page A-24 to get the friction factor. The value 340 in this equation is the length of clean steel pipe that would give a pressure drop equivalent to the valve, ie it is what is called the equivalent length. The pressure drop through the valve is determined by the shape and size only, and not by its material. Irrespective of the material of your valve, or your piping for that matter, the pressure drop through the valve is given by Crane in terms of the length of clean steel pipe that would give an equivalent pressure drop and so you must always use the friction factors in the table at the top of page A-26 to calculate K and never try to substitute a friction factor for some other pipe roughness. If you have lots of time, and are feeling very strong, you can read a never-ending debate on this subject in "Crane 410 fittings" thread378-173164

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[pleckner]

Harvey, there is no doubt that Patricia is in error but you are thinking in terms of liquid and this post is about a choked flow gas/vapor system.

 

[katmar]

Phil, I picked up that it is a choked vapor flow situation, but it seemed to me that your previous post had fully answered the expansion factor question and that the determination of the K value for the valve was all that was still unresolved. That is the only reason why I discussed the K value only. If there is any other error or omission in my post lets discuss it.

regards
Harvey

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[pleckner]

No real issues here, just I never considered using Crane 3-16 for gas/vapor, only for liquids since the derivations are all based on Q (gpm).

 

[burtonwr]

Thanks to all for the comments.

Pleckner - I had the same for the K value, and used the values you described for the expansion and pressure ratios. I guess i just found it curious that Crane didn't have any examples with higher K values/reduced porting, as they seem to be pretty prevalent in the energy industry.

I understand what your saying about using the Cv values, but the particular valve in question is from a smaller outfit and published Cv values are no where to be found.

Again thanks to all for the info...

 

[burtonwr]

So, after all this i've now found the published Cv values for the valves, which has me all confused.

The published Cv value for a 3/4" reduced port 2" body valve is 13.11 at 100% open. Using equation 3-16 from the Crane manual equates to a K value of only 82.9 (assuming i use 2" for d opposed to .750, which would make it even smaller). In comparison with the calculated K value (shown above for a reduced port globe) of 348. I realize they're not going to be an exact match, but am i missing something, is Crane really that far off? There's nothing fancy about this valve, is just a regular globe body valve w/ a reduced port.

Any advice or guidance is much appreciated!!

 

[katmar]

It all depends on what you are actually trying to calculate. Do the calculation using the Cv of 13.11, and then again assuming K=348. Does this really make any difference to your decision or design?

As I said before, I would be happier using a published Cv value than a generic correlation. Checking in my ancient Fisher Catalog 10 it seems that for a 2" valve with a 3/4" trim a Cv of 10 to 11 might be more realistic, but 13.11 isn't that far different.

To try to get a feel for the difference, I calculated what orifice plate would give a similar effect. Obviously there are all sorts of things that affect the calculation, but under some reasonable assumptions I get a CV of 11.0 being equivalent to a K of 135, which is what you would allow for the permanent pressure drop for a 19 mm orifice in a 2" line. (sorry for mixed units - I think in SI)

Similarly, a Cv of 13.11 is equivalent to a K of 95, which corresponds to the value for a 20.5 mm orifice. So there isn't really a great deal of difference.

To get to a K of 348 you would need an orifice of about 15.3 mm. You have to start to wonder whether your other data allows you to split hairs like this. Do a sensitivity analysis by varying the K or Cv and see whether it affects your decision.


Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[burtonwr]

Katmar-

I'm trying to find the maximum gas flow through a hung-open control valve to size needed venting capacity on a downstream tank. The valve in question is a LCV, so in the event of gas blowby, i want the tank relief adequate. Currently the tank has a relief capacity of approx 8mmscfd, so the difference in the calculated K (348) and Cv (13.11) is important to me.

As such, using Crane eq 3-20 w/ choked flow and the calculated K value of 348, i get a maximum flow through the valve of approx 4mmscfd. This is with an upstream pressure of 600psig, downstream pressure of 0psig. I used table A-22 to find maximum Y and Dp/P1 factors, in the same way as example problem 4-21. SG of the gas is .65.

Using the vendor supplied Cv through the valve, i get a maximum flowrate of nearly 9mmscfd. All other losses through piping and valves has been omitted in both cases, as they are small in comparison.

So, assuming i've done the calculations right, you can see how the difference w/ the Cv value is very important to me as it influences the allowable port size in the LCV or causes the need for a retrofit on our storage tanks.

 

[Ashereng]

I go with Cv values if they are available. This is a measured value by the manufacturer, specific to that valve. It is a real value.

A K value is an estimate at best, and a generalization most of the time. Each valve internal is different, and Crane can not account for all the possible manufacturing designs out there.

For your application, I would used the Cv value of your valve provided by the valve manufacturer - they measured it. This is probably as good as it gets, unless you do your own measurement.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[katmar]

burtonwr, yes I can see that you have an application where the true value is important. It is rare to use a control valve in the full open position, but when it has to be that way then the rated Cv is very important.

In your particular case I would be wary of using the stated max Cv. I don't know who the valve manufacturer is, but I have found instances where I believe the manufacturer has been a bit "optimistic" in his rating. If you are designing a safety circuit then you want to be sure. Can you test the valve - even just for a relatively short while and monitor the pressure decrease? This would allow you to back-calculate to an effective Cv or K value.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com