Vertical Venturi


Thread Starter


Hello all,

I was wondering if anyone has any experience for calculating the range for a d/p cell on a Venturi installed in a vertical line? The transmitter is a rosemount and is measuring the flow of lean froth which has a density of around 0.9 s.g into a vessel. we are having problems accurately setting the range and what would be the best method of zeroing the transmitter?

also does any one here have experience with measuring flow off of a vertical line, which instruments are you using?
Hi there,

To answer your last questions fist. We normally use a magflow in these vertical lines and on slurries.

To do a zero in your DP flow application, open both HP and LP lines to atmosphere and do a zero trim, just to make sure you have a good starting point. You then need to have the main line filled to past the first bend downstream so that the full height of the downstream pressure head is on the HP as well as the LP side of the transmitter. Theoretically you should now have a small difference in the zero which should be equal to the difference in height between the HP and LP tapping, times the density of the liquid, but it doesn't really matter what it is. Do another zero trim now and the transmitter will have a good process zero you can work from. Just make everyone aware that a atmospheric zero should not be done again after this.

If your line pressure is more than 10Bar you need to do another process zero as well during normal flow operations. To do this use the HP impulse line tapping point and put that pressure on both HP and LP side of the transmitter simultaneously, (use a flex hose and a T-piece)and close the equalizing valve on the manifold. If there is again a zero shift do another zero trim. This will be the most accurate zero you can do but this is only uses in high pressure lines since the high pressure causes zero shifts. If you find this easier to do leave the filling of the line method and just do this online process zero even if your process pressure is below 10Bar.

One thing I am sure about is that you can use the venturi in a vertical line, but the negative side is that there is a difference in the calculation.
I will give you a calculator to calculate the normal flow in a horizontal venturi but I can only give you a actual calculation on how to do the calculation for the vertical installation. You might be able to figure it out.

The calculator:

The vertical calculation modification that is needed:

The Bernoulli Equation
Assuming a horizontal flow (neglecting minor elevation differences between measuring points) the Bernoulli Equation can be modified to:
p1 + 1/2 &#961; v12 = p2 + 1/2 &#961; v22 (1)


p = pressure

&#961; = density

v = flow velocity
The equation can be adapted to vertical flow by adding elevation heights h1 and h2.

Hi again,

I have been in a long extended argument with myself and at the end of the day I had to say to myself, myself you gave that man incorrect advice on how to do the zero, so here is my revision of my previous post.

To do the zero open both the HP and LP of the transmitter to atmosphere and do a zero trim just to get a good starting point.
Fill the main line with your product as high as possible, the higher the better but not needed to be to the first bend. At the end of the day you just want to eliminate the differential between the HP and HP tapping points but the higher you fill the line the more accurate the zeroing will be. Once filled do another zero trim and this will give you the most accurate zero you can do on the application.

The process zero with the tubing as in my previous post will not work and can only be done on a horizontal installation.

If you cannot fill the main line, a less accurate way to do it is just to fill the HP and LP impulse lines with the product liquid and then do the zero.

If that is also difficult to do, another way to do it (accuracy will depend on how accurate you do this) is to calculate the zero, but to do this you need a very accurate sg value of your product. If you do not have that you can measure the sg by open both HP and LP to atmosphere and do a zero trim. Keep the equalizing valve close and keep the HP open to atmosphere and fill the LP line up with product liquid all the way to the point where the liquid starts running out of the tubing at the LP tapping point. Take a very accurate measurement from the middle of the diaphragm of the transmitter to the bottom of the tubing where the product starts running out of the tubing. Say this value is 1360mm
Now recalibrate the transmitter for LRV = 0mmH2O and URV = 1360mmH2O.
The value displayed by the transmitter will be a bit less since you are not measuring water, so say it display 1320mmH2O (it will be negative but ignore that), take that value and divide it by the 1360.
So 1320\1360 = 0.97058. So the accurate sg of your product is 0.97058.
Now use that sg and calculate the pressure head from the transmitter to the HP as well as to the LP tapping point.

So say the transmitter is installed exactly 1200mm below the HP tapping point and 1700mm below the LP tapping point.

Use the "sg x g x h" formula.
0.97058 x 9,81 x 1200 = 11425,764Pa = 11,326Kpa
0.97058 x 9,81 x 1700 = 16186.499Pa = 16,186Kpa
Now deduct 16,18 from 11,32 = -4,86Kpa.

So pump the HP side of the transmitter to 4,86Kpa (basically this is the pressure needed to get to the same pressure head as on the LP side tapping) and do a zero trim and this should also eliminate the difference in pressure head between the HP and LP tapping.

Finally you will now re-calibrate the transmitter with your final flow calibration. Say this is 0 to 50Kpa. Let's test the way it will work this time.

Say at the moment the main line is empty and your transmitter will read below zero, due to the zero trim you have done with 4,86Kpa on the Hp side. If you now slowly start filling the main line with the product the transmitter will start to increase the moment the level of the liquid gets to the HP tapping and keep on increasing from below zero until the product gets to the LP tapping in which case the reading on your transmitter should indicate zero. If you keep on slowly increasing the level in the downstream pipe this zero will not change and if you increase the level until you get to the first bend downstream, the transmitter will still indicate zero.

Since there is no flow a differential will only be created once the flow start to increase but this proof that any differential created by any flow will now be measured from a accurate zero point.