sewer forced main problems 10

[kentgolding]

A forced main sewer was installed on my last job and is having some issues.
The line is about 1.5 miles long and goes from pump station over a small rise and then drops lower than pump station into a draw and then goes up over another hill higher than pump station and then drops into a gravity sewer on the top of the hill, a built in air bleed here. We have one vacumn/air valve at the top of the first hill and a blow out at the lowest point but for some reason no clean outs were installed; deleted through a higher authority so we have what we got.
We have been losing GPM and a slight increase of pressure has occured since start up, about two years ago. A lot of different ideas have come up and a lot of checking of system has been done and all things point to a block in the line.
We do have valves in the system and one engineers idea was to shut one of the valves down in known degrees (25 turns to go from open to close so 8 turns would be 1/3, 16 turns 2/3 and 25 turns full shut) and check pressure and GPM. If the pipe is 1/3 clogged then we will see no change when valve is closed 1/3, and so on. From the other side of the room we are hearing that this won't work due to variables in design(?). Seems like it would work to me, and if so our test idicates that our line is about 2/3 clogged. We have dropped from a specified 522 GPM to 330 with a pressure increase of almost 15 PSI.
We have a sump pump in the collection well and a centrifigual pump in the pump house to push over the hill. At start up we were at 522 GPM (although documentation is rather weak).
The big question is should the valve test idea work? Seems pretty nice and simple to me, but I am a "simple is good" type person. Any ideas on this would be great, thanks to everyone. Also if you wish some more info let me know.

 

[semo]

Have you checked your air release valves to make sure that they are truly operating? Air buildup in the peaks can cause "air-lock" which will reduce your flow and cause an increase in pump pressure. That is where I would go first. If unsure, take the things off and manually bleed any air. It is not uncommon for air release valves to become clogged in a sewage system. Some manufacturers even provide backwash kits for their valves.

As to your other idea about closing a valve. You can shut a valve a considerable amount (50% or more depending on the type of valve and the system conditions) without seeing decrease in flow and notable increase in head.

You also can't assume that a closed valve will give you an indication of any percentage of blockage. A partially closed valve will act like an orifice and cause an isolated increase in headloss. Clogged piping can act the same if it is a clog in a small area; but, a 100 foot long clog will act more like a section of smaller pipe. (ie you might get 10 feet of headloss with small constriction of 75% pipe diameter; but, you could also get the same amount of headloss over a long constriction of 10% pipe diameter.

The only way to check if it is clogged is to cut a section out and see.

 

[kentgolding]

Thanks Semo good info on why the valve testing may or may not work, probably not.
Actually we did open our air release valve yesterday after some other work and found quite a bit of air. It was brought up that since the location of the air valve almost always has pressure on it (beacuase the air valve is located at the lower elevation hill and has the water column pressure on it from the next higher hill) that our release valve may not work right even after it was torn down and cleaned. The facility staff is aware of this and is keyed in to do a manual drain at least for a while to see if release valve is working or not. We did see a slight increase in GPM (about 20 GPM) after bleeding but still missing about 140 GPM. Since we opened up the line at low point the same day we bled air valve we are not sure which action actually made the difference (I hate it when that happens).
Thanks again for the comments, appreciate everything we can get.

 

[RWF7437]

http://www.westernfilterco.com/pigging/pigging_applications.html

 

[kentgolding]

We were hoping to go here (pigging) last but thank you very much for the link.
The pigging may very well be our final shot. Would have been far nicer if cleanouts had been installed. Be intersting to see how this all turns out.
A last tid bit for anyone reading this, we have two different design firms and two different responsible parties (city and state) invloved, one for the state from pump station to the low spot and then one for the city from the low spot to the gravity sewer. Of course the problem is in the "other guys" part

 

[bimr]

The valve idea is not going to work because the pressure drop vs. % valve opening is not a linear relationship. The valve pressure drop is going to increase exponentially only when the valve is almost closed.

You do not mention the velocity in the force main. Velocity criteria for force mains are derived from observations that solids do not settle out at a velocity of 2 ft/sec or greater. Solids do settle out at lower velocities or when the pump is stopped. A velocity of 3.5 ft/sec or greater is required to resuspend the deposited solids.

Try pumping periodically at a higher velocity to clear the line. Get a flow meter to document the flow.

 

[semo]

Bimr has a another good thought there. I know you can't make the existing pumps increase flow (ie velocity) with the existing forcemain; but, do you have emergency connections for bypass pumping?

If so, bring in a larger temporary pump to connect and increase the scour velocity to see if that helps.

Also, I don't know how old the system is or what type of pumps you have; but, have you checked the pump to see that it is still operating on it's curve? If it has one and the wear plate is adjustable, any excess gap between it and the impeller will also cause the pump to lose efficiency (ie flow).

 

[kentgolding]

I guess I left out some of the information in the start, sorry and thanks for the input.
Our engineer on the state side of the project (we were responsible for the pumps, commissioned on 3-27-07) noted that we should be getting 3.7 FPS at design flow of 522 and even with the slow down at 380 GPM we should be getting 2.6 FPS,(we have a Siemens meter in the system which only shows GPM and total pumped, upgrading to a better model but hasn't happened yet, we should have a third party come in and verify FPS rate) which if I read above correctly should keep solids in suspension (we run the material through a grinder that reduces solids to 2mm). The design built in a redundant system with two of the same sump and house pump, and pump set A and pump set B cylce trade off run times. With this design we do not have hook-up for an emergency pump but that would be a good idea.
We have a 4x4x11 submersible 40 hp sump pump with a 103' tdh at 1750 and a 4x4x14 torque flow with 170' tdh at 1750 for the house pump. One other item that makes the whole system interesting is the wet well (collection tank) is small enough that it takes approximately 6 run cycles before the water that has left the tank reaches the gravity sewer. The system is designed for a prison facility that we finished a year ago, and the flow during the day is fairly constant and the pumps cycle quite a bit, but at night time there is not much activity. The pumps are not that old and have not been run a lot. throw another log on the fire, the drop in flow began while project was still under the GMGC one year warranty so needless to say there is some third party finger pointing in here, trying to figure out whether a design issue (I don't think so) or a construction issue)
Our engineer for the project was laid off several months ago from the firm that did our civil design, he was nice enough to help us out a bit after lay off but no one here (state) wants to hire him back on to help with this problem. I think it looks closer and closer to going back to the "pigging" web page.
One of the group left thinks that the sump pumps might be running in reverse, so they are going to have a submersible pulled and run just to see if that might be the case (at this point everyone feels we should do all we can to make sure the problem does not involve pigging), luckily the sumps are set up with a fast release mount, just pick it up along guide rails and set it back in the same way.
Again thanks to everyone for the input and ideas, I will pass them onto the rest of the group.
(I mainly do QC work during the construction and warranty period but I enjoy reading this site and was hoping we could get some ideas here, thanks, a lot of good stuff. This is my first big job and I hope they all don't go this way, every other warranty item I have fixed, this is driving me crazy)

 

[rconner]

Do you know what size and material/description of piping is involved (the flow & velocity numbers look just a little bit odd, at least if this is some common inch-sized piping).

 

[kentgolding]

we have 6" pipe through the pumphouse and where it 90's down to go into the floor and underground outside it goes to 8". I believe we have a 4" to 6" expansion off of the pump. Ductile iron is used through out the system, from the wet well to the gravity sewer (I just tried to shoot a photo of the pump house, not sure if it worked. Sorry my photo program is not working so it is a rather large photo.)

 

[RWF7437]

You've probably already done this but, I would suggest you draw an accurate ( to scale) profile of the force main. Using that, plot the hydraulic grade line based on the original design, i.e. how it was supposed to work. Then plot the HGL of how it actually is working based on your measurements. The difference may give you important clues as to where and what the problem(s) may be.

good luck

 

[civilperson]

The pumps are too big as evidenced by your statement of cycles. A primary pump that starts in November and quits twice a decade is the correct size, (use a second stage high level pump for flows greater than the first pump can handle and a large volume equalization basin to smooth out the diurnal peaks.)

 

[bimr]

The suction piping looks to be undersized. You should have 8" dia. seuction piping for 500 gpm.

It looks like you have silent checks on the pump discharg. These will plug. You should have a free flow discharge type of check valve.

Is the pump suction plugged? If so, the pumping rate will be reduced.

How long is the suction piping?

Are both pumps pumping at the same pumping rate?

Is it possilbe that the pump check valves are leaking backward?

Obtain the pump curve. The pump should be operating on the pump curve. If not, there is a problem with the pump.

 

[kentgolding]

Thanks for the input. We have been playing with this issue for several months now and everyone just wants it to get over. We may be calling in an outside engineering firm to come in and do the calcs as you have mentioned.
Civilperson, good point about the pumps cycling so often, unfortunately that was a design item that I will look for next time around.
bimr, I am hoping you are wrong about the pipe size, hopefully if we get an outside firm in they will find that pipes are correct or I guess we will have to figure a way to change them. Both pump sets pump the same, and we have checked the valves in the pump house. Yes the checks are spring and we have seen debris already, but not enough to cut us back 140 GPM. Someone said that they thought a regular flapper type valve would beat it's self to death rapidly because it was so close to the pump? We do have flapper checks in the piping from the sump.
The piping from the sump pumps in the wet well is about 40 feet, I guess it is suction line although it runs about 35 PSI when the sump pumps are supplying water to the house pumps which also are running. I do need to have the piping from the sump to the house pumps checked before we do a lot more work (like digging up pipe or running a pig) and I hope we can do that soon.
Everyone who has answered here has been a great help, thanks to all of you

 

[alexcmmi]

I am not following everything you are saying with sump pumps (?) but just looking at that suction piping, I am wondering if you might have NPSHa issues. Suction pressure is 35 psi? How does that compare to your pumps NPSHr? How do the pumps SOUND when they are running?

Our those pumps just hanging out in the middle of a room - the discharge header seems to be in the middle of nowhere w. plenty so space on either side...just seems very strange use of space.

 

[kentgolding]

Our facility sewer system dumps into a 25' deep 10' in diameter wet well that has two submersible pumps located in the bottom. They pump out of the tank (one or the other runs each time, cycle and trade off same as the two pumps in the house) and into the pump house you see in the photo to the house pumps and then into the header pipe which travels a bit farther to the right than shown in the picture and then the pipe turns 90 degrees and goes through the floor of the pump house and then over the hills to the gravity sewer.
The buildings were all pre-cast panels and the pump house is about the smallest design on the whole site, probably something to do with minimum size of pre-cast and design. We also have a huge bioxide tank in the building and there is room for storage and good access for working on equipment. Being a sewer system it is nice to have the room to avoid drips and what not.
Not sure what you mean by NPSHa or NPSHr. As far as sound of the equipment running we don't here grinding noises or cavitation type sounds, and both sets of pumps sound the same with the same output.
One of the ideas with the design is supposed to be ease of maintenance. Being a 1800 person correctional facility the sewer system has to work correctly or there is a lot of trouble. Piping and pumps were placed to be easily serviced, unfortunately finding this issue so far has not been easy. We have a heavy grinder that reduces solids to 2mm, and from what I have heard inmates flush all kinds of items into the system. The area we are having trouble with is after the grinder, so not an inmate flushing issue.
Side note: one of the "tricks" that inmates like to pull is flush all toilets in faciltiy at the same time, haven't ever seen it but hear a lot about it

 

[redbridge]

I would calculate the operating point on the curve based on existing profile. Several startups that I have seen had flows higher than what it should be because they did a drawdown test without a full line.

This may seem like a stupid question but are you sure all the valves are open. We just finished investigating a similar situation for one of our clients and the were 100% sure all the valves were open. After looking at the as-builts they noticed they had a buried valve and it was about 1/2 open.

Good Luck!

 

[alexcmmi]

NPSH = Net Positive Suction Head
a = available
r = required.

NPSHa > NPSHr for you pumps to work.

I am still not following you, but it sounds like you have an inline situation where the sumbersible pumps in the wet well feed those horizontal close coupled pumps in your picture? I am guessing you did this b.c of the high heads? If this is the case, you get into a tricky operating situation and your problem may very well be that your submersibles are not feeding the horizontal pumps w. enough volume and they are essentially sucking the suction header dry...

I may be way off here, but I have a hard time following what you are saying. Diagram of the pumping situation might help.

 

[bimr]

I think you are correct, alexcmmi. He has the pumps operating in series. Don't know why he would have just installed a single chopper pump.

One additional point to check is the clearance below the submersible intake. It should be installed 7-8 inches off the wetwell floor.

 

[kentgolding]

To everyone who has helped so far thanks a lot. Sorry I was not in office the last day or so but from what I have read the supply to the house pumps is sort of the direction my group is headed. I do agree that every valve needs to be (and already I've been told it has been done once but I want to see it)checked and opened and closed the full distance of travel. all the gate valves in the ground are the same so they should have the same turns ot open and close. I am including two drawings form the project that show entire lay-out from grinder pit to house pumps and a picture of the sump pumps in the wet well. The grinder located with the auger reduces material to 2mm and the auger takes out anyting that is crushed but not ground (i.e. toilet paper, strips of cloth and anything else an inmate might decide to flush)before the material ever gets to the wet well and the first pumps.
Right now we are planning on doing a check on the piping from the wet well to teh pump house, easiest place to look (and the last place) that we don't have to spend a lot of money or use the pig.