Fluid Film "thickness" on a gravity return line 1

[Henrique Orlandini]

Hello.

I work with drive-pump trains that usually have API 614 lube oil systems (LOS) installed. Those LOS always have an oil return pipeline that works only with gravity, and this return line have quite a few temperature gauges + thermowells to check how hot the oil is leaving each equipment bearing. One question that I receive a lot is that, since the return line is not pressurized, how can I be sure that the fluid film thickness of the oil in the piping is in touch with the thermowell tip?

I use a standard 3 mm gap from the thermowell tip to the bottom of the pipe, the good old and dangerous "we always did like that, never heard a complaint before", but I want to be able to actually calc this thickness to have another argument for this question.

According to API 614 6th edition, annex I, item I.3, I can use the Darcy-Weisbach method to calculate this, but I have a doubt in which they call "pipe length". I know it sounds silly, but my oil return line doesn't have just one "source" of oil, it has several branches that connect to the header, thus the oil flow throughout the oil pipe varies with the length.

I'm sorry if this is way too specific, but it would be of great help if someone could shed a light on how to properly use this method (or some other method) to calculate the fluid film thickness of a gravity return line.

I'm attaching some snippets of the standard, in case it could be of any help.

 

[hydtools]

Look at open channel flow calculations.

Ted

 

[hacksaw]

"I use a standard 3 mm gap from the thermowell tip to the bottom of the pipe, the good old and dangerous "we always did like that, never heard a complaint before", but I want to be able to actually calc this thickness to have another argument for this question."

That practice is based on old measurement guidelines, by measurement tests and CFD simulations as long as the well is immersed at least a few tip diameters, you'll have a viable measurement.

 

[1503-44]

Gravity flow in pipes is typically found in storm water and sanitary sewer design.
The Manning Formula is most commonly used for gravity flow, although DW or other formulas can be used by accounting for the hydraulic radius, as most of those assume pipe flowing full, and by making certain adjustments for viscosity, Manning could be used for fluids other than water.

With gravity flow, you are looking more for "depth of flow", not film thickness.
Film thickness is more associated with a pressurised flow of gas and liquid, which under certain conditions produces an annular flow of liquid next to the pipe wall while the gas flows down the center of the pipe, leaving a certain (film) thickness of a fluid ring moving down the pipe at a different flow rate than that of the gas.

The following pdf explains the hydraulics of pipes not flowing full, in which case the pressure at the liquid surface is constant along length of pipe. At a constant flow rate driven by gravity, the depth of the flow rate will be such that the total friction loss of the fluid stream at the contact with pipe wall will equal the energy gained by the drop in elevation of that fluid stream. At any given flow rate there will be one slope of the pipe at which the depth of flow remains constant for the length of that pipe. A greater slope will cause the velocity to increase and the depth of flow to decrease along the length. A decrease of slope causes the opposite effects, velocity slows and flow depth increases.

From PDF of CHAPTER 2 HYDRAULICS OF SEWERS - New Jersey Institute of Technology

https://web.njit.edu/~washd/cim305/download/notes/CHAPTERS/CH 2.pdf

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

So are you measuring each branch before it joins the header or the header downstream of each connection?

Either way you can't calculate anything unless the branch or the header is on a constant slope.

You need to calculate velocity to work out depth.


Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[1503-44]

Surely the pipe slope is constant. But you generally do need to know the inflow rate. I hope you have a bucket and stop watch. Then you can calculate depths and corresponding velocities at all points along the length of the pipe segments using its (their) slope. Unless it is the slope that balances friction loss, both depth and velocity will not be constant, as I have said above. It becomes an iterative solution at each point along the pipe.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

We don't know if it's level/ constant slope or not. It's a LO header so could wander all over the place...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[1503-44]

Its gravity flow, so there is probably at least one slope.
The fewer the slopes, the less segments you have to make to analyze.
If the elevation drops are steep, then it quickly becomes a pressured flow situation, unless flow is very small. Small pipes tend to fill completely. Slopes for partially full pipes are relatively flat, where 1:10 is getting pretty steep. In fact, gravity flow in this context may only mean it does not have a pump and the pipe is pretty much flowing full almost everywhere. "Gravity powered pressurized flow"

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[Henrique Orlandini]

Thanks for all the replies and sorry for the delay.

The slope is indeed constant in the main header and on the branches. We try to keep at least the minimum as per API (2.3° iirc) + the ship pitch/roll values.

And about the measuring, mostly I'm trying to look at the worst case scenario to find that even at the branch with the shallowest depth, I still have a good reading.

Our equipment are custom made, so it's tough to actually measure at the equipment, because if the piping is all done, it means the equipment is almost ready to be shipped.

 

[hacksaw]

Since the degree of pipe roughness has not been referred too, if your current installation works and is safe, don't "fix it."

 

[1503-44]

That is a 7 in 100 ft fall in elevation, or a 0.07 slope, Reasonably steep.
Estimate the roughness by choosing the "C" coefficient (I think it is C, ) for the pipe material.
Calculate depth of flow using 0.07 slope at various flow rates to get an idea of the possibilities you have there. And pick a flow rate that looks reasonable, then you will have a depth estimate.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[LittleInch]

Can't you strap surface temp measurement to the bottom of the pipe instead and then insulate that bit of pipe? Then it's guaranteed.

Or strap them to the branch lines coming from each bit of equipment?

Something like this

https://thermometer.co.uk/thermocou...e-5024368334602.html#/320-clamp_diameter-50mm

or this

https://www.thermosensedirect.com/a...sCiWXI3Sj7u88TeXMgCTpCWg0UsZ_wUxoCa_wQAvD_BwE

or this

https://uk.rs-online.com/web/p/ther...jkJY-fyWKY4rvJh0RsxoClHMQAvD_BwE&gclsrc=aw.ds

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.