Negative Pressure at Orifice Flange Exit 1

[KimWonGun]

I am reading a negative pressure at the exit side of the orifice flange assembly even though the main reservoir (shown on the left in the attached image) and scavenge sump (not visible to the right)are vented.

Creating a larger vent in the scavenge sump did not affect the magnitude of negative pressure.

As shown there is a 1" T-valve connecting all three.

This phenomenon occurs only at higher temperatures (10F - 200F).

The speed is 3,600 RPM.

Does anyone have a suggestion on the source of the vacuum?

 

[hacksaw]

is the down stream tap on the pump suction?

 

[KimWonGun]

No.

The transducer is tapped into the orifice flange, which leads to the T-valve and then to the main reservoir and scavenge sump.

Initially I thought the scavenge sump was inadequately vented, creating the vacuum caused by the suction line from the scavenge. But after observing the magnitude of the negative pressure unchanged despite doubling the vented area, I am stumped.

 

[hacksaw]

your piping configuration is less than clear

you have orifice flanges that is clear but you are trying to mesasure gage pressure from the downstream flange tap.

are they weld neck flanges?

Flange size?

is a plate installed?

is the orifce assembly discharging into a sump or into return vessel? If so is it a submerged discharge.

I would start with a sketch of your piping arrangement

 

[KimWonGun]

I hope the attached sketch clarifies the configuration.

I have a pressure transducer at the upstream and downstream flange taps.

The flanges are screwed boss type and 1" size.

An orifice plate is installed.

As I attempted to explain, the orifice assembly discharges into a main reservoir AND a scavenge sump using a T-valve to control the flow ratio. In neither case is it a submerged discharge.

 

[Pumpsonly]

What is the up stream pressure?
Flow rate , Beta ratio? Liquid property?
Type of orifice?

For orifice, if the flow is too high
you get cavitation.

 

[KimWonGun]

The upstream pressure is 74 psi.

The flow rate is 30.9 in^3/s.

The beta ratio is 0.24.

The specific gravity is 1, and the kinematic viscosity is 6 cSt.

The orifice is square-edged rather than sharp-edged through a 1/8" thick plate.

 

[Latexman]

This one is a long shot, but . . . have you calibrated your pressure transducer lately? What's it read when open to atmosphere?

Good luck,
Latexman

 

[KimWonGun]

All transducers were calibrated simultaneously immediately before the tests.

Keep in mind a key detail mentioned earlier: The magnitude of the negative pressure decreases as the operating temperature decreases (and viscosity increases). So at about 0F the exit orifice pressure is what I had expected from the start: zero.

Someone in my office suggested that at higher speeds the fluid flowing through the T-valve may be lowering the pressure enough to create a vacuum (Venturi effect). Does this seem plausible?

 

[KimWonGun]

The "key detail mentioned earlier" described in my last post actually only stated that this phenomenon only occurred at higher temperatures. So what I wrote in my last post supplements rather than reiterates the earlier information provided.

I should add one more detail that might help. At temperatures below 0F the exit orifice pressure reads slightly positive values. Does that strongly suggest a constriction (the T-valve) is creating back pressure in this case (lower temperature, lower speed, higher pressure, higher viscosity)?

 

[25362]

What about the effect of viscosity on the vena contracta?
Could it explain the downstream pressure variation on cooling?

 

[hacksaw]

on the basis of your data the meter d/p is 1601" wc with a meter loss is 54 psi at normal flow! There is little doubt that the absolute pressure is dropping below the vapor pressure of your fluid at higher temps.

while 1" piping is always a concern for orifice metering, but your meter d/p is soo large as to invite problems.

if you increase the bore size to 0.500 or so, the meter d/p drops to 64"wc at normal flow, and the net meter loss to 1.6 psi

 

[hacksaw]

you have a secondary issue here apart from having to recalcualte your meter for small bore piping (requires use of the special meter coefficients below 2" in your case the d/p at normal flow increases to 77.5 "wc to give you some idea)

and more importantly the reynolds number is below 10,000 so you really cannot get a good measurement with a traditional square-edged plate

 

[hacksaw]

rather I should say, below the Re limitations of sq. edged orifice plates, not laminar flow

why are yu taking so much drop across the plate?

 

[KimWonGun]

I was given a test specification to follow which stipulates the high pressure drop. That may need to be re-visited.

 

[hacksaw]

Kimwongun,

the orifice may have been intended as a flow restriction to avoid use of a valve, if you do use a larger orifice, make sure your valve can handle the throttling service