Entrapped air in horizontal pipe 1

[orlandobill]

Does anyone have experience with this? Here's my situation:

Vertical pump sends 15500 gpm of water into an empty 30" diameter 120' long horizontal pipe that turns down 10' vertically, turns and runs another 60' horizontally and then turns up another 20' vertically. The 60' long "trap" holds water every night when the pump is turned off, but the 120' long horizontal pipe drains completely.

Is the pump flow (~7 ft/sec) enought to push all of the air out of the horizontal pipe? Is there any information out there that might help? Should I see huge slugs of air coming out of the discharge? I have flow issues and I'm thinking that a huge air bubble in the horizontal line is restricting the flow. Thanks for any information.

 

[katmar]

I did not mean anything too specific or technical by "designed for". It was the 15500 gpm at 21' head that I took as the design point.

At what elevation is the pump? The 19' of static head could have a different effect if your pumps are mounted on the level of the 120' horizontal section rather than if they are mounted below the sump liquid level. The impact will be on the NPSHa.

You are correct that you need to consider all alternatives. But from my calcs I am convinced that trapped air will be a problem, and if other things are causing the flow to be lower than the 15500 gpm (you mentioned 12000 gpm) then it will make it even more difficult to flush out the air.

I strongly believe that your first action should be to get pressure tapping points on the pump discharge. This, together with the pump amps and the flowrate, will tell you if your pump is operating on its curve. If possible also get pressure points at the far end of the 120' section and somewhere near the middle of the 60' horizontal section. In addition to telling you the pressure, the tapping point on the 120' section will tell you if air is trapped there before you go out and buy an automatic vent.

It seems to me that you will have to do a "first principles" study before you can pinpoint where the problem is. It could be anything from a pump running in reverse to a welding bottle left in the pipeline. We have all experienced these things!

Good luck

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[orlandobill]

The vertical pump intake bell is at 72' and the centerline of the discharge pipe is 93'. The sump water level is 82'. I know my NPSHa (~40') is marginal to my NPSHr (~34'). That's material for another post.

katmar- What calcs are you refering to? I assume it has to do with the self-venting vs. siphoning, but I'd like some more information on the numbers you mentioned. I can calculate the Froude number, but am unfamiliar with the 0.31 and 1.0 numbers. Do you have links to information? Thanks.

 

[katmar]

Bill, yes I was referring to the self-venting and siphoning calc. I don't have any online links, but then I have not Googled it. The best reference that I have found is an old one from Larry Simpson, Chemical Engineering, June 17, 1968, pgs 192-214. In many ways this article is more useful than the Crane TP410 manual. A more recent article, with a slightly different slant and not quite as relevant to your situation is P.D. Hills, Chemical Engineering, Sept 5, 1983, pgs 111-114.

I'm at home now and I only have a 5th edition of Perry (Chem Engrs Handbook) here. Under "Drain Pipe" Perry gives almost nothing other than the reference to Simpson. The newer editions may have more info.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[Artisi]

seems we are back to where we were sometime back - a static discharge head of around 12ft ( 93' pipe centreline + half a pipe dia. minus 82' (sump level))
15500 USGPM thru 30" pipeline length 120' plus bends velocity etc = approx 2' loss.

Total head is therefore approx 13'.

Back to my earlier posts- I would be looking at entry problems as the pump is probably way out on the curve resulting in NPSH problems or vortexing or etc. Still interesting in what the "flow issues' are?

 

[katmar]

Artisi, The discharge is at 100' with the sump level at 81' giving a static head of 19'. The pump has to supply all this differential head, even though 11' of the 19' is on the pump suction side. I agree that the pump is probably way off on the right hand side of the curve, and operating at a head of approximately 1/3 of the shut-off head. It could well be an NPSH problem. Bill, do you hear any cavitation?

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[Artisi]

Take back what I said, seems I was confused on the static head - I assumed the 93' was the discharge point whereas it seems to be the pump discharge from the sump.

 

[lilliput1]

What is the elevation of the pump with regards the sump water level?

 

[orlandobill]

It's a vertical pump.

Inlet bell elevation = 72'
Sump water level = 82'
Pump discharge elevation = 93'

 

[katmar]

Bill, do you have a single pump supplying the whole 15500 gpm? If there is more than one pump (duty or installed standby) what does the valving look like. Can you give us a pump make and model so that we can view the curve on the web somewhere?

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[Gator]

You guys really need webspace to post sketches so you are not all running around in circles guessing about actual installed configurations/proposed solutions.

The fact that comments are not threaded just adds to the general confusion, IMO.

 

[Artisi]

Gator is correct - a total confusion at the moment as to what the static head is.


Can we clear up what the static head is - please advise the vertical distance from the sump level to the dischrage point - NOT the pump discharge level - the level at which the pipeline discharges.

The following is a free image posting site -

http://www.freeimagehosting.net/

 

[Artisi]

orlandobill -- If you think that the air in the pipeline is a problem a quick fix to prove or disprove this point is an extension of BigInch earlier advice - (hole/s in the pipeline) - fit a 4 or 6 vavle to the pipeline which you can manually open to allow any trapped air to escape - see if this overcomes the problem - if it goes why not fit a standpipe to the pipeline with the highest point above your discharge point by a foot or so.

 

[katmar]

Artisi, thanks for the reference to the image hosting site. That would be a great help. If orlandobill lives in Orlando the chances are that he is still asleep right now.

If you scroll back to orlandobill's posts of 22 Feb 07 10:11 and 22:07 he made the relative elevations pretty clear. Where I went wrong was to assume that he had a horizontal centrifugal pump mounted somewhere between the sump level and the 120' horizontal section, i.e. somewhere between the 81' and 93' levels. In fact "Vertical Pump" was almost his first words in his first post so it was my mistake.

It seems the pump suction is 10' below the sump surface, which agrees with the HPSHa of ~40' (i.e. 32' atm plus 10' depth less a foot or two of vapour pressure and friction). On this basis I can't see anything fundamentally wrong with the setup. I suspect the air would be a real problem at start-up, but being realistic I would expect that if you ran the installation a whole day the air should eventually be flushed out at 12000 USgpm. I don't think we can make any more deductions until we have some basic data like discharge pressures and motor amps. I asked Bill for a reference to the pump curve, and that would also be a help.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[Artisi]

Like myself I didn't read some of the info all that well, however, I agree that the head static head is probably as advised - although I would like clarification to make sure for my own peace of mind. Also, orlandobill hasn't really addressed the "flow issues" so we don't really know what the issue is, - making it difficult to advise with any certainty.
Is it air in the pipeline, poor intake design, free vorticies or sub-surface vorticies etc etc.

Coupled with this a curve, discharge pressure, motor detail and current draw would also go a long way to clarifing a few points.

 

[orlandobill]

Thanks for the image hosting site. I hope these help. While I certainly appreciate all of the free advice, my focus at this point is the possibility of entrapped air as shown. I also should have pointed out that all of the pipe is either buried under 10'-20' of dirt or encased in concrete. Adding an air vent at the supposed collection point will be very difficult, but not impossible, which is why I am looking for information to predict whether the air could collect or not.

My "flow issue" is that while I expect to get 15500 gpm from the setup I am only getting 12000 gpm or so.

For clarification:




and:
[url=http://img2.freeimagehosting.net/image.php?f387f1c74f.jpg]

 

[katmar]

orlandobill, there are defintely other things that you can do before you go and dig down through 20' to install an air vent! While I am still sure the air will be an issue at startup, the more I think about it the more I believe that if you run this pump all day the air should all come out eventually.

The sketch makes it look like the discharge point is visible above the water level. If this is so, can someone get close to it and observe it during start-up? If they see massive bubbles of air being ejected for a few minutes, and then it dies down and becomes a nice steady flow of water then you can be fairly sure you are getting the air out. On the other hand, if they only see an ongoing stream of tiny air bubbles you will know that the air is not being removed rapidly at start-up. Have you tried this observation?

Once you have the pressure gauges installed next week that will give you more pointers. Was the pump shop-tested before installation? Is there more than one pump that is behaving this way? I assume you have a "-1" impeller size. If it is, you should easily get >13,000 gpm with the calculated hydraulics but running at 15,500 gpm will be very close to the edge of the curve.

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com

 

[orlandobill]

katmar- Do you ever sleep? I'm kidding!
I will try observing the startup for bubbles. We have not been able to shop test this pump. Our rebuilder does not have the facility to do this. We thought about setting something up to try that, but haven't gotten there yet. We have a few pumps acting this way. Look at my forums and you will see my other question about our sump. I will post more info (pressures, amps, flows, etc.) Thanks for the help everybody.

 

[Artisi]

The curve makes a big difference, at first look I would say you could have a hydraulic problem. Being so far out on the curve you could have a pre-rotation problem which will reduce flow or intake design problems.

Is this a new installation or an existing one?
If an old installation, did the pump operate ok before going to the rebuilder?
When you get some discharge pressure and power information we might start to see what is happening.

I keep asking questions but it's the only way to get a feel for what is going on.

 

[lilliput1]

What size is the pump inlet & discharge?

 

[katmar]

orlandobill, if these are the same pumps as you previously discussed in your sump problem ( thread378-173908 ) then we are all barking up the wrong tree here.

If you want to get sensible answers I suggest you start again and put the whole story on the table. There is simply too much confusion, contradiction and missing information.

In the sump posting you said that you run 2 pumps at once, each at around 14400 gpm. Because your static head greatly exceeds the friction head, running 2 pumps at once will give you close to twice the flow rate in the pipeline. This would certainly flush out the air, or are you now running only one pump at a time?

In the sump posting you said this problem has arisen every year for 10 years. It seems that at the beginning of the year, after the pumps have been rebuilt, they run for a while and then gradually deteriorate. After rebuilding they are back to spec, but then the deterioration cycle starts again. This is a very different picture from what you have painted here.

In the sump posting you said that you would be inspecting the sump for build-up and the pipeline for scale and blockages. Was this done? What was found?

At what point did you decide that it was not a sump problem and that it was a trapped air problem? Why?

Katmar Software
Engineering & Risk Analysis Software

http://katmarsoftware.com