Pumping dirty water using a normal centrifugal pump 4

[Muud]

Hello guys I would like to ask a question regarding dirty water & centrifugal pumps. I have recently started working in filtration company and we test the filtration efficiency of our filters using water mixed with test dust. The test dust has particle size range from 1 to 100 micron. The centrifugal pump takes the clean water from a reservoir and pumps it into the system. After that a dosing pump injects slurry (test dust in water) into the water pumped by the main pump and it then passes through the test filter. It is a closed loop and and in normal circumstances the dirty water is filtered again before it goes back into the main reservoir. In other words the centrifugal pump doesn't come in contact with dirty water. However for another experiment I need to recirculate the dirty water and present it again to the test filter. I wonder if feeding the main centrifugal pump with dirty water will cause any damage? Could someone give some information about that?

Thanks

 

[ashtree]

You don't mention specifically anything about the test dust other than size and nothing about the pump.
However unless the dust is particularly abrasive or you are putting in a lot making the water very "dirty" it should not give you any unusual problems.
There are lots of standard centrifugal pumps working well in raw river water, industrial processes etc that would be similar or worse than your application.

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[LittleInch]

Depends a lot on the model of pump and more importantly it's shaft sealing system and bearings. Depending on exactly what this "dust" is made of it could erode the impellor or casing if it's a sharp edged particle.

Pumps designed for dirty water usually have pressure fed seals with clean water such that the seal flushes dirt away from the seal edge. If you have any bearings in the fluid then they need to be suitable for the dirt level or located outside of the fluid.

Find the pump details and then call the vendor. It might just need a different seal type, but you really need to know what this "dust" is.

If you don't know or just run it, you could severely damage the seal faces and internals of the pump.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[moltenmetal]

There are centrifugal pumps intended for pumping sewage, which is a great deal dirtier than your proposed slurry.

There are centrifugal pumps designed to pump full-on, high solids slurries as well.

Just stay away from magdrive pumps in this application, as the small clearances between the driven magnet and its housing tends to get clogged with solids, interfering with pump function and eventually wearing out parts. Many magdrives specify a maximum solids size of 0.006"

 

[Artisi]

What's the dust,how much is there. Also is this an existing application or something new?

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Valvecrazy]

A sewage or effluent pump can handle abrasives, but may not build the pressure you need. In sandy water wells I use a submersible with open impellers. Maybe you can drop a sub over in the supply tank instead of using an end suction centrifugal? The sub doesn't have a seal to worry about and the open impellers will last a long time with dust, grit, sand, etc. Wolf Pumps in Abernathy Texas makes the old HPC open impeller design pumps. There are thousands of those in really sandy wells used for irrigation in the high plains.

 

[Muud]

Dear all. Many thanks for your valuable feedback and I am sorry for the late reply. I tried to get as much details before I replied. Perhaps it makes things clear for you now but if any other specific information is need I will look it up as well. I look forward to your suggestions

@ ashtree

Regarding the type of dust, it is ISO test dust. I have attached the chemical composition of the test as well as the chemical reactivity copied from the safety data sheet.

http://files.engineering.com/getfile.aspx?folder=402c86ef-8a23-4ccf-8ddd-5e073202076f&file=1.JPG

http://files.engineering.com/getfile.aspx?folder=156f490b-20ab-4c34-825c-c5ce9da01a73&file=2.JPG

Looking at that information it doesn't seem that bad for the pump. Normally this dust is filtered prior to water return to the main reservoir so the main centrifugal pump doesn't come in contact with this dust. The amount of dust usually added is around 3g/l so if not filtered the concentration can be as high as 12 -15 g/l water. However, for filter made up of very open materials it takes ages to clog them so I was told to put normal sand instead of the standard test dust. This type of sand/dust contains considerably large particle and also we add it large quantities (upto 5-6 kilograms per 700 liters water).

Regarding the details about the pump, there is no documentation available and I have looked up the pump type plate.
Here are the details
Q = 70 m3/h H = 40m
Motor type LS 112M P= 4 kw
I = 8.3 /4.8 A
U = 380-DR Vn = 2900 min-1
f = 50 HZ , IP 55 , ISO = F
Cos ϕ = 0.9

It's 12-15 years old pump and couldn't find much information specifically about that model. What I have learnt is that it is a single stage end suction pump with mechanical shaft seal. Also I don't know about the impeller type unless I open the pump.

@LittleInch,

I will try to get in touch with the manufacturer but it seems like the original company has been acquired by another company. So don't know if they can provide me information whether there are bearings immersed in water. But as mentioned above I know what this dust is made up of. What do you think about this dust.

@Artisi

I am not sure what you mean by a new application. Normally the test dust that we mix in water doesn't come in contact with the pump because it is filtered and only clean water is fed to the pump. However, I would like to try another type of test in which the water with dust is not filtered and recirculated through the dust. In that case the concentration could reach as high as 15 -20 g/l.

@Valvecrazy
Good suggestion but changing the pump at the moment doesn't seem likely due to cost cuts. I am sure I will be told to stick to the standard test and don't try anything.

 

[LittleInch]

Doesn't really tell us much about the type of particle though.

However if this is silica dust it's likely to be quite sharp and hard, so I would think your mechanical seal and the shaft seal surfaces will get worn away quite quickly and wreck not only the seal, but the shaft surface also.

The bearings will be outside the impellor, but it's the seal you need to get some details on.

These give you some idea of the issue.

http://www.waterworld.com/articles/...cal-seals-for-water-and-wastewater-pumps.html

http://www.mcnallyinstitute.com/03-html/3-6.html

If you're going to do this you really need some clean flushing water to lubricate the seal.
Or buy a progressive cavity pump. Those things pump small rocks!



Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[moltenmetal]

An air diaphragm pump will last a lot longer than a progressive cavity pump if its feedstock includes small rocks- and has the added benefit of not having any rotating or travelling seals between the fluid being pumped and the outside world.

 

[ashtree]

The other question to ask is how long are your tests, how much water do you intend to circulate and how often do you plan to do this?

Regards
Ashtree
"Any water can be made potable if you filter it through enough money"

 

[Artisi]

Ashtree's 1st post summed up the problem, or lack of problem. The solid / liquor ratio is minute and should just be treated as dirty water.
What pump brand / model / material info would be worth knowing.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)