Negative Suction Pumping

Option 2 has a lot going for it.

A non-return valve has to be provided somewhere in the system, otherwise the pipe contents will run back through the pump into the sump when the pump stops. So the friction loss of the valve still has to be accomodated. Where it goes in the pipework is determined by the available Net Positive Suction Head, with a preference at the inlet to the pump's inlet piping, so as to maintain prime.

If non-return or check valve is installed in a)discharge line after the pump outlet it will reduce the load of the pump b)suction it will acts like foot valve and help priming.

B is correct.

A is incorrect.

I believe a)will help to start the pump without creating a back pressure on the impeller

What a nice question !!

I understand exactly what you mean. Please, keep up the good work.

Your question seems to be well worded, in English, correctly spelled and "all caps" is not used in the message. For this particular goofy forum, this type of message surely is a rare treat !!!

With regard to your pump configuration question, you are missing a third option. Consider the use of a priming tank.

Some information is available on:

www.gouldspumps.com

Best regards and keep up the good work....

Or option 3: A sump pump.

Thanks for all your advice, they really help a lot.

We don't use sump pump for this case as there are chemicals involved and this may lead to corrosion.

I suppose why we always put foot valve even for self-priming pump is to "avoid time delay due to re-priming during start-up" (I don't know how long it takes to re-prime though)

No discharge check valve for self-priming pump? Would that not lead to reverse flow and cause damage to the pump (being centrifugal)?

For a previous project, we used a priming tank when the pump centreline was only slightly higher than the tank outlet (say 400-500mm). Would it work if it is 3000mm? I did see a priming tank at one of the pump manufacturer's testbed facility (similar scenario as this current one) though -- they used vacuum pump to help with sucking the air out of the tank. Now that I recall, they mentioned that it is economical with there are a lot of pumps (then can use common vacuum system), but would be expensive if it is only two pumps involved.

I would appreciate any opinion :)

"We don't use sump pump for this case as there are chemicals involved and this may lead to corrosion."

If there are chemicals/corrosion involved then you should keep it as simple as possible, avoiding unnecessary valves, priming pots, etc. A good chemical self-priming centrif (or possibly a vertical lineshaft pump?) would be best. I used to work for a company that made (arguably) the best self-priming centrifs for chemical process applications. Send me a PM if you want further info.

"I suppose why we always put foot valve even for self-priming pump is to "avoid time delay due to re-priming during start-up" (I don't know how long it takes to re-prime though)"

Priming times will depend on many things, including the volume of air to be evacuated, the pump size and speed, the priming efficiency of the pump, etc. We used to consider anything more that 2 minutes to be too long for priming.

"No discharge check valve for self-priming pump? Would that not lead to reverse flow and cause damage to the pump (being centrifugal)?"

There are two routes between suction and discharge, since internal recirculation is necessary during the priming cycle, so you would not necessarily get reverse flow through the impeller when the liquid drains back (except if you are using a standard centrif + priming pot).

Is the water clean? Depending on valves to hold prime with dirty water can be annoying. Your multiple applications raise the question, as does "foot valve usually leaks." You don't want an added flow restriction that does not hold prime.

For a pump suction head of that value, and because of the corrosive/varied nature of the fluid pumped I would look at a progressive cavity pump rather than a centrifugal. Progressive cavity pumps don't usually like running with dry stators so a "kink" usually needs to be placed in the suction so that when the water runs out the pipe the stator is still submerged. The big advanrage of PC pumps in this application is that they prime very quickly and footvalves are generally not needed unless you have a very long suction.

There is another method called priming pot. This basically a sealed vessel introduced in the suction line and piping arranged in such a way that the pot is always full of water even if there is no foot valve. The volume of water in the pot is generally kept about 50% more than volume of suction pipe. Once this sealed pot is filled through a filling line , and the pump is started, the pump pumps out the pot water creating suction in the suction pipe thereby drawing water through suction line into the pot and then to the pump.

Option 4, You could use a jet pump. They are normally used on shallow water wells. In essence the output pipe is open ended at the bottom of the well. Water is pumped out of a reservoir down the well into a jet that is pointed upward inside the well output pipe. The fast moving stream of water out of this jet, creates a venturi effect that draws water out of the well. The reservoir is refilled from the well discharge and the excess water is used for other purposes.

If you don't mind diluting the contents of your tank as you pump it out, fresh water could be used to feed the jet pump. Then your pump would never have chemicals in it at all.

Submersible pump? If there are metals that will make a good check/non-return valve, there should also be submersible pumps that will hold up in the chemicals. With non-corroding materials, several of the other pump options would work OK.