Substitution of flow test fluid

[aafuni]

I have a nozzle with complex geometry that has a specified flow condition that must be met. The specification gives a flow rate (lb/hr) at a given test stand pressure (psig) using MIL C7024 type 2 fluid (commonly sold as stoddard solvent or white spirit).

I would like to be able to flow test the nozzle using water in our test stand, rather than the solvent. I was hoping to find an accurate way to correlate the flow rates of water and oil through the nozzle. Ideally I would like to be able to run the test at a different test pressure to compensate for the fluid's density and viscosity change, and be able to get the same flow rate, and therefor be able to verify the nozzle's geometry in this way.

I have experimented with similitude, but am not that familiar with this method. Using similitude it seemes that I must match both the Re and Cp values given the variables. Using the equations for both, I get an equation dP*rho/mu^2=constant between the flows. This equation makes no sense physically as it suggests that the more viscous fluid requires less total pressure drop to result in a matched flow rate.

I know some companies use different test media, so this should be possible.

 

[zdas04]

I've had good luck matching Reynolds Numbers. Calculate Reynolds Number for the solvent density and velocity. Solve the Reynolds Number for velocity with water and use that value in your tests. It is a good idea to run a couple of tests with the specified fluid to make sure you got your sums right, but this technique works.

There are a lot of representations of Reynolds Number, for a similitude evaluation I use

Re=rho*v*D/mu but you can just as easily use
Re=rho*q*(pi/4)D^3/mu

The "D" value can be selected at any point, I like to use the diameter and velocity at a point where I have a constant diameter for a ways (not just after the convergent section of a nozzle for example). For water you want to look at changes in density and viscosity with changes in pressure and temperature to see if those changes are material (they may be, you have to check).

"Cp" has at least 4 definitions that I know of (so there are probably more than 20 definitions). I'm not going to guess which one you are referring to.

David

 

[aafuni]

Thank you for your help, I will give this method a try.

 

[aafuni]

I actually havent worked much on this project but just picked back up on it...

Do you match the Reynolds number and run the new fluid with the same Reynolds number, expirementally determining what the new flow rate acceptance will be? How do I verify the accuracy of this method, other than just repeating it?

or..

Would I be calculating the exit Re for the specified conditions, and then keeping the inlet conditions the same with the new fluid and the exit Re should end up still matching? This does not seem like it would be the case.

Thanks

 

[eadwine]

I have three quick questions. Why are you deviating from the specification test requirements and has the owner of the specification approved of the deviation? You are quoting a Military Std. for the fluid required. Is the military the owner?

Better get approval first before modifying the test procedures. The owner may not accept the product due to specfication deviation.

 

[aafuni]

We are the owner of the specs. One of our vendors who makes parts for us overseas is unable to easily acquire the fluid we used when we created the specifications here.

 

[eadwine]

Next question. Why are you using a vendor that can't provide the necessary quality control and final operational testing as per specfication? Sounds like design& fabrication by bean-counters and purchasing under their control. This can lead to big problems.

 

[aafuni]

eadwine,
Our vendors are quite good, and have been able to meet our needs and specifications. In transfering products from one vendor who was unreliable to this vendor, they are having dificulties purchasing the specified test fluid due to their location in the world. They are a machine shop and not a dedicated testing facility, and they have done a good job so far with quality testing of our other parts.

eadwine, do you have any help or are you just going to point out flaws in our engineering department and our vendors?

 

[eadwine]

I've never been a big fan of rewriting specifications after the fact in order to cover a vendors lack of capability. The vendor seems to have the capability to ship product back to you. So the transportation link is open. If the vendor can't acquire the required test fluid, but can run the test. Why don't you acquire the required fluid and ship it to him? That way you you don't have to rewrite the test spec. Or another approach,, have the completed nozzle assemblies shipped to a test facility that can run the test. Problem solved and the specfication still stands.