Excessively High Static Head Reading 5

[elcid0M6]

I am working on a pump station that has everybody scratching their heads. This was an existing pump station with a forcemain that trended downward to the discharge. Our clients decided to send the flow to a different portion of the system, so we tapped the existing 16" FM and routed a new FM to the other portion of the system. This new force main is shorter but trends upward. The discharge is to a gravity sewer manhole and is the highest point along the alignment. We have 5 combo valves at the various hills along the alignment to vent any entrained air, but again the highest point on the FM is the discharge.

The pumps should operate at 1250gpm @ 59'TDH based on calculations, but we performed several drawdown tests over the past couple of months(taking an inflow first to account for it) and keep getting readings in the 800-850gpm range. The latest drawdown was 838gpm @ 77'TDH, and this point falls directly on the pump curve. We performed a shut-off head test on each of the 4 pumps and get right at what they should be.

We then re-ran the calculations (to make sure there wasn't a bust) and came up with the 1250gpm @ 59'TDH value again assuming a C-value of 120. The discharge elevation is 42' and the gauge elevation is 20.5', thus the static head reading on the gauge should be 21.5'. The issue is that the gauge is reading approx. 42' of head. We have changed out gauges and had 2 gauges operating off the same line and they are all reading the same values. We have had our surveyors out to check the elevations 3 different times and they come back correct as shown. We have physically watched the guys manually bleed the lines to ensure there is no air in the line, and as soon as they turn the valve you get liquid.

What could possibly cause the static head to be off by approx. 20'? Why would it also carry through when the pumps are running?

Any help would be greatly appreciated.

 

[bimr]

Is it possible that the discharge manhole is surcharged?

 

[elcid0M6]

Thanks for the comment bimr. We have the operations guys going out to check that, but even if it was surcharged it theoretically wouldn't account for the full 20'. The top of the discharge manhole is approx. elevation 47.5', thus the gauge should read approx. 27' of head not 42'.

If the contractor inadvertently installed a high point between ARVs, could that cause a consistent false static head reading? The ground profile is at the highest point at the discharge, so there should be no way for an intermediate highpoint to be higher than the discharge. With that in mind it just doesn't make sense that an air pocket would cause a false static reading beacause there is a known maximum water column elevation, but I am starting to question everything at this point.

 

[SteveWag]

I have had the same thing happen earlier this year, but with a smaller mysterious false head. The problem was dumped on me, as I was not involved in the design or construction of the project. The project is exactly as yours. A long up hill force main dumping into a gravity sewer. As with yours, gauges were exchanged, elevations were rerun, dynamic measurements were right on the curve. The contractor had bid on open cut force main installation but had returned from an auction with a directional boring toy. He was allowed to install the force main with his new toy, and in doing so had unknowingly caused many “high spots” in the main. Air release valves were placed at the “design” high spots.
Now, with a water-filled main, with no air, the static head is just the difference in elevation from one end to the other, BUT, if air occupies the “other side” of a high spot some “Artificial Static Head “ can be introduced. A series of these high spots can add up fast. I will attach a sketch later. Anyway, we profiled the main using locating and depth measuring radar, and began to add air valves. With each air valve installation the static head dropped.
Steve

 

[elcid0M6]

Thanks for the insight SteveWag! I am really interested in seeing your sketch, because I think I understand but a visual would help. The re-profiling based on radar or "pot-holing" is likely where we are headed with this project.

Would we be able to narrow down where the issues might be occuring by placing pressure gauges at the existing ARVs and comparing gauge readings to calculated values? This force main is 7,500LF so a sub-surface exploration along the entire route would be very expensive.

 

[SteveWag]

If the pipe is full of liquid the pressure at “A” and “B” would be the same (excluding the situation where the rise is more the 30-33 feet). But when trapped air is present, and if unequally distributed as would be the case if flow was traveling from left to right, an “artificial head”, and equal to the difference of the levels appears. The pressure at “B” is transmitted to the left-hand water surface and the pressure at “A” is a little higher because of the vertical distance from “A” to the left-hand water surface.
Steve

 

[cvg]

stevewag
your website gswagner dot com was blocked by our corporate firewall as being weapons related. can you re-post the image using engineering.com (the link below)

 

[SteveWag]

Done. Yes,lots of firearms n'stuff.
Steve

 

[elcid0M6]

Thanks for all of the insight SteveWag. Looking back through the plans it seems that there are several places that the FM should have been laid to a grade and not follow the ground profile. My guess is they followed the ground profile more than the grade and put several high points without ARVs causing the phenomenon you illustrated.

Contractors....can't live with them and can't live without them.

 

[bimr]

I am not following the idea behind this concept. If the pipeline is at rest, I don't believe the static head would be different.

Have the surveyors traversed this line from end to end? Are the refernce points correct?

 

[semo]

Like Bimr, I don't understand this concept. It is against Bernoulli's Law. That isn't to say that the line won't surge in pressure up to 40 psi; however, it should come back down to static (high point minus low point) at rest.

Air is not like water, it is a compressible fluid. In general terms, given pressure it will compress at pump start; but, once compressed, it can violently expand and cause high pressure readings as well as break things.

The air in a pipeline will also cause an artificial restriction in the pipe diameter and without enough velocity cause the pump to operate at a lower point on the curve (giving lower flow and higher head). This explains your duty point problem; but, without flow (at equilibrium) I don't see how your pressure will increase. Is there another pump connected to this line somewhere between your pump and discharge point?

 

[elcid0M6]

To bimr:

Our surveyors have run re-run this route 3 different times and they come back as correct.

To semo:

This force main is dedicated solely to this pump station, so the static this station should see would just be the elevation difference.

I agree this concept seems illogical, I just can't come up with any other explanation. If there is pump operating every 5 - 10 minutes would that truly give the 7,500LF of forcemain time to come to an equilibrium if there was air in the line? When the pumps cut off the gauge bounces around a good bit, but it always settles back on 42' of head between pump operations.

 

[SteveWag]

elcid0M6, you can try running as many pumps as possible and see what the gauge indicates after shutdown. The higher velocity may push along some trapped air. If the gauge indicated something less than 41-feet it may indicate something.
Steve

 

[coloeng]

I've had a difficult time wrapping my arms around this concept as well. However, the more I thought about it the more possible it seems. Let me see if I can explain what may be occurring.

If there is a pocket of air in the pipeline that is trapped by a high point it will be compressed when the pump begin pumping and increases the pressure at that particular reach of pipe. When the pump stops the air pocket will expand and push some of the water out. Because there is probably a check valve by the pump, any water that is displaced by the air pocket will be pushed towards the discharge side of the pipeline. This would mean that the water on the downstream side would maintain its level at the peak of the flowline. The water column from the discharge point is now placing a hydrostatic head on the air pocket and if the air pocket occupies an area that encompasses the entire area of that section of pipe, the pressure of the air pocket will match the hydrostatic head. That air pocket will then place that same hydrostatic head on the column of water on the pump side of the pipeline. Because there is a portion of the water column on the pump side of the air pocket that is higher than the bottom of the water column on the discharge side of the air pocket, there may be an increase in hydrostatic head.

Not sure if I explained this well or not, and I'm not sure I'm totally convinced in my explanation.

Let me know what you think.

 

[SteveWag]

That’s as good an explanation as any. The key is the unequal water surfaces. If as shown on my sketch, the air actually causes the elevations of the water surfaces to become equal, raising the right hand column of water UP, making the pressure in the left hand side higher. Now, the right hand side does not physically move up, but makes the gauge at the pump read higher. Forget about dynamic conditions and the compressibility of water, we are talking static.
Steve

 

[bimr]

If the pipe goes over a hill and then down to the discharge, you would have a higher static level. The static level is going to be the difference between the high point of the force main and the low point, not the difference between the discharge point and the low point

Howver, you are stating:

"The discharge is to a gravity sewer manhole and is the highest point along the alignment."

 

[SteveWag]

I am seeing the system as something like the one shown on the attached PDF (sorry it needs rotated). Static head discussed here is from discharge point, the high point, to gauge centerline, not water surface. Assuming the F. M. to be 16-inch and the best flow rate to be 838 GPM, the velocity would be about 1.3 FPS, not enough to remove the air from the line. There are three automatic air release valves and some un-released high points. If two feet of air is trapped at each un-released high point, 12-feet of mysterious head would appear, with a gauge reading of 33.5-feet.
Steve

 

[elcid0M6]

The more I think about this the more it makes sense to me.

It seems like if there is an unequal water surface around a high point with pressurized air between them, you "stack" the water columns. Using your intial sketch, at point B the air is getting the pressure from the water column on the down stream side (from point B to the discharge). The air then transfers that pressure/head to the upstream water column, causing an additive effect.

For example, assume the pump elevation was 0', the intermediate hight point was 5', the low point was 2' and the discharge was 10'. If the pipeline is full of water the static head would be 10', but if the air pocket was contained in the pipeline between the intermediate high point(elev 5') and the next low point (elev 2') then the water columns would add up, 5' - 0' = 5' and 10' - 2' = 8' thus the static head at the pump would read as 13' not 10'. That is why it only is an issue if there are unequal water surfaces.

Does that sound correct?

 

[SteveWag]

Stacking is the word I was looking for! Yes, the individual columns of water are “stacked one on the other and the sum is greater than the whole. Is the FM 16-inches in diameter? If so, at 800 GPM, the fluid would be pressurized up to the top of a high spot and then turn into gravity flow in the pressurized pipe, pressure would push it to the top of the next high spot and then down as if on a sled. Each downward turn occurs in a pressurized section of pipe full of air. The pressure of each section of air grows less as the main rises. As coloeng says, the pocket is compressed by the dynamic loss when the pump runs and expands when the pump stops. This is greatest at the first air pocket and lessens as we go along the FM.
Steve

 

[bimr]

I can understand the effect while the pump is operating. But once the pump stops, the head due to the compressed bubbles will dissipate.

Then you would be left with the static head being the difference between the high point and the low point.

In addition, as the bubble travels moves to the high point, the static head should decrease.

After checking the hydraulics of this application, one would think that the flow rate is too small causing the forcemain to be fouled with solids.

I think what you have is the forcemain is full of solids and this fouling is causing the backpressure to be held when the pump is shut off.

When you operate, you should only see 10 Ft of friction @ 1250 gpm or 4.5 Ft of friction loss @850 gpm. You are saying that you have 77 feet which indicates that the forcemain is fouled.

Note that it takes 3.5 ft/sec velocity to resuspend the solids once the forcemain stops and the solids settle out.

You may have to have the forcemain cleaned to remove the solids.

Some additional questions:

How long has the forcemain been operating at this rate?

What size sewer does it discharge into?

Are there any valves in the forcemain besides the check valve?

What is the brand name of the check valve supplier?

Have you inspected the "combo valves" along the route and are sure that the vaults are not flooded?