Pressure in Pipes

You left out two critical details:

What is the flow of water?

What are the outlet conditions? (Is it simply discharging from the open end of the pipe?)

Yes, air pockets can certainly increase pressure drop but since the present pump pressure is twice what you are calculating, I would look for other causes.

Greg

The nominell flow of one pump is 0.5556 l/s and the pumps are linked together in paralell, so we have a flow of 2.2224 l/s.The water just flows in to an open tank of 100m3.

Asuuming the pipe diameter 110 mm is the inner diameter a total flow rate of 2.2 l/s (133 l/min) creates a mean flow velocity of 0.23 m/s. That is not much, according my understanding. It may explain why the flow from end of pipe appears to be "weak".

However, calculating the Reynolds number I get about 40000 - means, clearly a turbulent flow pattern.

For a plastic tube I would assume a smooth surface at inner wall. For this condition a 2000 m long pipe (without disturbance from bends, bows, juctions and pipe joints) should create a flow resistance (pressure drop) in range of 0.4 bar.

Means: the lenght 2000 m of the pipe does not explain the observed (?) pressure drop of about 4 bar. Of course, bows and joints will create another part of flow resistance, but if those elements/arrangements made with due care, it should not explain the missing 3.5 bar.

By the way: if you would install check valves as recommended you can be sure to need more than 4 bar for the flow resistance: each check valve is a flow resistance and requires a minimum opening pressure. I can't agree in those recommendations. However, to protect the pumps from running reverse after they are switched off you may need a check-valve (or anti reverse lock for the pump shafts).

Air bubbles: In case the pipe is following the shape of landscape with up and down passages the pipework may enclose one or more "traps", where air is trapped. The flow velocity is too small for driving the air in flow direction, so the air remains in those places, making the flow noisy.

Local presence of air may cause higher local flow resistance, because it is acting like a throttle, causing higher local flow velocities (=higher flow resistance) to get the steady flow rate passing these sections. Again, I can't think, the flow resitances from those effects would require the same pressure than the total gravity head and normal flow resistances from pipes. Air is a soft/weak resistance: in case the throttling effect is too much the flowing water will drive the air through the pipe like a piston. And I can't see why the flow resistance for an "air piston flowing with same velocity as the water" should be greater than for the water.

No, I think it is worth the resistances from joints of pipes, bows, etc. should be investigatated. Probably there is more reason than looking after air bubbles.

Albert

If air in the pipeline and the potential "water-hammer" effect are concerns you should install "air bleed-off" valves at the High points along the pipeline. Good engineering practices would also recommend the installation of "check" valves to prevent back flow and the resulting air pockets.

Once the initial bleed-off of air is done at start-up you will only need to "Burp" the line if the pumps run dry or if maintainance is done on the line that would allow air back in. As for a "calculation" to account for the pressure drop of an air pocket at a High point in a pipe, well it's not needed if you eliminate the air.

In order not to loose the original Q, you must use check valves to avoid the kick back of pressure ,at least 10 every 1000 meters(1 per 100 linear meters)

Could you explain in more detail what you mean?

I never heard of what you seem to be describing, and it makes no sense at all.

Greg

I agree with that. I've only had a passing involvement with long-distance pipelines, but I've never heard of regular check valves being installed, and even if some engineers go for it, one every 100m seems a lot.

On the original posting, I estimate about 13.5 m³/h (velocity 0.46 m/s) to give 0.5 bar headloss in pipe assumed 102mm actual bore. Air release valves needed at high points, as others have mentioned.

Codemaster,

Other than at the pump outlet, the only conceivable place that I could see anyone wanting to put a check valve might be just upstream of a substantial rise, but he just mentions every 100 m, and "kick back of pressure", and that just sounds ridiculous. Depending on the outlet conditions, I might put a vacuum breaking (anti siphon) check valve up there but no other check valves except the pump outlet.

Air bleeds at the high points, absolutely! I thought the high (2X what they calculated) pump outlet pressure was the result of more than just trapped air, but it is possible to be that alone.

I am not at all familiar with metric pipe diameters, and don't know if they follow the "standard" method we use in the US, which is "equivalent (inner) diameter" which is not necessarily the same as either the actual I.D., O.D., or the nominal pipe size.

Greg

Hello Greg

Metric pipes are specified by nominal diameter ND which is the OD. A few examples either side of this one are 63, 75, 90, 110, 125, 140, 160, 200 mm. There is also reference to the NW which stands for something in German I believe, and is roughly the nominal bore.

The actual bore of course depends on wall thickness hence pressure rating, but the 4mm I used would be typical for 110 mm ND.

Obviously the OD is the controlled dimension, as it is in inch systems, so one range of fittings covers the various pressure classes.

Hi Codemaster,

Thanks for the info! I looked up some metric pipe info after my post, but it didn't specify PVC pipe so I wasn't sure.

Regards,

Greg

So you are saying that in this 2000 m run of pipe, you will need 20 check valves???

Are you a check valve salesman?

That sounds ridiculous.

Air pockets not only increases the pressure drop , but also decreases the rate of flow Q. Therefore you have to vent the pipeline at the higher points.

Abdel Halim Galala

Expert Design Gen. Mgr.

Egypt

So you agree with me on that the problem might be airpockets. The waterflow into the buffertank is also very weak.

Firstly general advice regarding solar power for watter supply.

A reservoir of adequate size to store water for a few days is essential.

The water can then be discharged at 0 pressure. And obtain a constant and maximum efficiency.

A solar power booster pump can then be used to deliver additional pressure whenever light is available.

If enough flow is available below consideration should be given to a ram pump, water wheel or turbine driven pump.

The pipe size must be designed to be economically and financially viable.

A general rule of thumb is to design at ± 1 m /100m friction.

The total head would then be 30 + 20 = 50 m

Note air release and check valves.

There is written a lot about flow resistances from pipes - for nearly each kind of bow and straight pipe segments. I am not experienced in those calculation, but as far my knowledge one practise is, to express the flow resistance from each pipe segment and valve (devices) in a "required gravity height" to compensate this individual resistance. Adding the required gravity height for all segments results in a gravity height which has to be overwhelmed in addition to the geometrical height between pump and pipe outlet.

For Minimising flow resistances it is recommended to have a small flow velocity in the pipes, because the flow resistance typically increases in square with the mean flow velocity. Further more the flow conditions are of influence: laminar or turbulent flow patterns.

You have said: we know the pump pressure is 7 bars ...

Does it mean, this is an experienced value? If true there is reason for assumption your have a pressure drop due to flow resistances of about 4 bar (if outlet of pipe is against ambient air pressure = open pipe end).

On the other hand you have considered a pressure drop of just 0.5 bar. What is the base for this value? What is the flow rate?

I guess there are experienced technicians in the CR4 commununity, which can tell you best procedures for calculations from view of practise.

good luck for finding the mismatch

To answer your questions specifically, it seems that air is trapped at the high points in the pipeline, and bends and turns (elbows, etc) are adding to the pressure drop. If my conversion numbers are anywhere correct, you should be getting 4 or 5 times the flow that you indicate. I assume from your post, the system is operating but not functioning properly. There are modifications that can be made to improve the flow. A "wet tap" system can be used to install a small vent valve at each high point without shutting down the system. An automatic air vent can then be installed on the valve.

If perchance the pipeline is exposed on top of the ground, it could possibly be repositioned to straighten out some of the bends, etc. Even if it is buried, it might be worth the effort to reroute the pipe to straighten it, and at the same time improve the elevation gradient by removing high points.

You state that the system has four pumps connected in parallel. This being the case, it is necessary to have check valves on the discharge side of each pump, and each pump should have a valve (butterfly or gate) on the suction and discharge side in order to isolate the pump. The check valves prevent water from being recirculated through the inoperative pumps. These are the only check valves you need. I would also suggest installing a pressure gauge manifold across each pump to measure the dynamic head of the pump in order to determine its discharge. Using one pressure gauge to measure the suction and discharge pressures is more accurate than using two gauges. You should generate a series of pump curves showing the discharge of 1, 2, 3, and 4 pumps. Knowing the discharge of the pumps, you can construct a "system" curve which will provide you with the capacity of the pipeline at the different flows.

Future projects of this nature could include some preplanning in routing the pipeline such that a known high point could be incorporated into the system. I'm curious to know if this is a "pilot" project, possibly to be repeated.

My best wishes for a successful project, and I commend you for this type of work.

g scott

"""The nominell flow of one pump is 0.5556 l/s and the pumps are linked together in paralell, so we have a flow of 2.2224 l/s.The water just flows in to an open tank of 100m3."""

Flow rates are not additive from pumps in parrallel. I know the math would say it is but unless your using positive displacement pumps the math has to change. I'm going to assume that the pumps being used are the common "centrifigal" style. With this style two pumps running in parallel will yield only a few GPM more flowrate than a single pump running.

Another thing to look at is the spec sheet for the pumps paying very close attention to the "Pump curve". If you are above or below your optimum pump curve for your application then you can be chasing your tail. If you are not familiar with reading a pump curve diaghram I would stronlgy recommend going to a local rep for more infomation.

Here is were I will advertise my company a little.... Contact your local 'Siemens Water Technologies' Rep, he should be able to review your system for you and make the best recomendations.

RIchard,

Excellent point!

The pumps in parallel thing flew right by me. Its amazing how many engineers don't pay enough attention to pressure/flow diagrams unless they are used to working with various kinds of pumps. But then, some of the pump manufacturers don't make things easy by trying to crowd a wide range of pump sizes and motor hp on a single chart.

When I converted my steam heat to hydronic in my home, I wanted to minimize the temperature drop through the zoned loops and maximize the heat exchange in the boiler by increasing the water flow. The larger "commercial" circulator was $300, while the smaller normal one was $60, I compared the pump curves and found I could get almost the same flow out of 2 small ones, and not only that, for my system, it mattered very little whether the small ones were in series or parallel! So, I put them in series because it was much simpler to plumb.

Regards, Greg

HI Greg..............Thinking about using 2 pumps in series vs parallel. What type impeller is in the pumps you are using? I don't understand how the discharge would be the same. Pumps (centrifugal) in series add pressure, and pumps in parallel add quantity, same as batteries. I don't doubt what you're saying, would you mind explaining it to me..........Thanks,

regards

g scott

OLD F**T,

I dug out my folder on my heating system and looked over the graph I had made.

I was initially pleasantly surprised too, but the 2 circulator pumps I used were TACO brand, models 007, and the flow vs head pump curves dropped off very quickly for a single pump (max flow at about 5' head, 0 flow at about 12' head): In parallel, the flow was as expected much greater at low heads, whereas in series the flow was much higher at high heads. They crossed in the general region where my piping system calculated out to with 2 of the 3 zone valves open. With only one zone in operation, the series did better, and with all three, the parallel did better. In an ideal world, I would always take some empirical measurements to reference the calculated pressure drops of my piping system, but the curves of the commercial pump, a single TACO 007, and the curves for two 007s all plotted in such a way that two 007s in series made the most sense, all things considered. The piping was 3/4" copper, with 3/4" gate valves for isolation and Honeywell zone valves (1/2" internal openings in the valve bodies).

Regards, Greg

I would just like to add to my previous post a few extra details:

I have a small Slant-Fin cast iron boiler, containing a relatively small amount of water, that is very efficient for an atmospheric combustion unit, but of course, less efficient than the forced air combustion units and considerably less efficient than the latent heat recovery types of gas boilers.

I wanted to minimize the boiler temperature I run at (I have a separate hot water heater, so the boiler is only used to heat the house), and maximize the water flow through it for better heat transfer and to minimize "overshoot". By overshoot I mean the temperature increase of the water in the boiler caused by the transfer of heat from the cast iron that occurs especially if both flame and the circulators cut off at the same time. (My controls fire the boiler when the circulators initially come on if the water is at any temperature below the max setting). Therefore, any temperature rise in the boiler after the thermostat has been satisfied and the circulators turn off (along of course with the burner if it happened to be on at the time) ends up mostly just increasing heat loss out the flue. I was a certified installer for a brand of flue dampers many years back, long before I bought this house, but didn't install one here because the payback would have been unfavorable for my type of boiler and the lower temperatures I run it at.

With the circulators my goal was two-fold: maximize flow through the baseboard and the boiler, and have as a high a flow through the boiler in a "only one zone on" (highest pressure head) condition. Otherwise I would have simply used one TACO 007 circulator, as is the norm for 2 or 3 zones.

For numerous reasons, heating contractors very rarely optimize these things, and service personnel tend to set the boiler aquastats at much higher temperatures than required simply to be on the safe side (for them) so they don't get service calls of insufficient heat in the cold weather. On new installations, higher boiler temperatures allow the use of less baseboard footage, so that incentive is there also.

Since I installed everything but the boiler (I just converted it from steam to hot water, since it was made for either), I put in as much baseboard as I could, to allow me to run as low a boiler temperature as possible, which increases heat transfer, and decreases heat loss up the flu, thereby increasing overall efficiency. If on the coldest days, I see that my circulators are running almost constantly, I will bump the boiler temperature up a few degrees and then lower it again as the weather warms. I installed an external adjustable temperature switch in series with the boiler controls (which has a higher cut off temp set), so I don't have to fiddle with taking any covers off or even using a screwdriver. The temp switch has an exposed fingertip adjustment and is mounted in the hot water outlet just above the boiler. The boiler burner control is energized only when a thermostat opens a zone valve and turns the circulators on. If the boiler temp is below the external setpoint I have set, the burner fires at that time, and will remain firing until the setpoint is reached unless the thermostat has been satisfied first (depending on which zone or zones are on, the boiler temp will overshoot about 10° F after the burner cuts off. Otherwise, for the rest of that heating cycle, the burner won't fire again until the boiler temp drops 10° F or so below setpoint.

I did permanently install a pressure gauge just downstream of the circulators (which are on the output side of the boiler), and the pressure head was in good agreement with my calculations.

Greg

GREG.......Thanks for the input That little pump must have a real steep curve. I don't recall ever using hot water pumps in series. I have used pumps for boosting pressures in high rise buildings on domestic water systems.

Intersting you mentioned the Slant Fin boiler. I prefer cast iron boilers but have had to use the new high efficiency, condensing boilers in several projects, although they are not my preference. I was "raised" on cast iron boilers. You can't beat an old sectional cast iron boiler. If it is not abused and even half maintained, it will last forever. Just replace the gaskets and cement and reset the sections, and they're good as new.

Have you considered using an outdoor air reset remote bulb thermostat to automatically reset the boiler water temperature. This would minimize the temperature over-run off your boiler. I used that control quite frequently in the past on hydronic systems. Now, with the electronic systems, the reset just becomes part of the overall energy management strategy.

Thanks again for the input..........

g scott

OLD F**T,

My pleasure. I have enjoyed reading your knowledgeable, informative posts.

That TACO is a neat little pump. When I replied earlier, the original data sheet was not in my folder, only the graph where I plotted the commercial pump against a single "007" and two 007s both in series, and parallel. Their presently stated specs are: 0-20 GPM, at 0-11 feet of head. Since too little head can overload a centrifugal pump motor, and/or the published graph was slightly different 17 years ago, I plotted only down to about 4' head, but otherwise the curve I generated from the data points is the same. You can look at the TACO 007 specs here:

http://www.taco-hvac.com/uploads/FileLibrary/101-029.pdf

I certainly considered an outdoor referenced control setup, but honestly, the boiler is only about 20' from where I am sitting, and I couldn't justify the added expense at that time, although it would have paid for itself long ago. (You have me re-visiting that option.) For most of the heating season, I run at about 130°-140° F, and ranging 120°-155° at the extremes.

Slant Fin happens to be located here on Long Island, and when I bought the house, it had not been installed too much earlier, only as I mentioned, it was originally configured for steam, but was designed for hydronic use also. The only problem I have had was a leaking plate gasket where the optional tankless hot water coil would install. I've had to replace the thermocouple pilot light sensor a few times, and disassemble/reassemble the millivolt gas valve once, due to it being reluctant to open unless the thermocouple was brand new, but other than that its been entirely maintenance free for over 20 years. I keep my automatic boiler water feed shut off with a valve, because it never requires additional water unless there would be a leak, or a few years pass. The pressure gauge down stream of the pumps allows me to monitor pressure. I keep the automatic feed off for several reasons: one is that overtime, they can fail to seat properly and since my line water pressure is about 60 psi, I didn't want have that pressure end up in my system and blow the relief valve, flooding my basement. Also, I have heat pipes in the basement, that are below the level of the boiler, and I didn't want to risk a problem there so I installed a pressure switch that prevents the boiler from firing unless there is at least 15' head. Having seen boilers destroyed (rusted through) after years of tiny pipe joint or gasket leaks that can take place under the sheet metal boiler housing, and evaporate before they ever actually produce a visible drip or puddle, that was another incentive to keep the system sealed so any leak would show up as a loss in pressure relatively quickly.

Regards, Greg

GREG.....In my previous post, I referenced outdoor reset contrrol for the boiler, then commented that electronic controls provide for energy management systems, etc. I was speaking of the new high efficiency "condensing" boilers.

By an large these boilers use a high rate heat exchange using copper tubing or a derivative. The volume of water contained in the tubes is very low, such that on light off, the combustion chamber comes up to temperature in a matter of seconds, and begins to heat the water in a few more seconds. If the water is not circulating, the tubes will overheat and fail. It is my understanding this has happened. Also, the water temperature is critical. I have designed several systems using these boilers that have been successful in their operation.

This is an outline of the sequence of operation that I have used to assure proper operation.

1) Call for heat, any one zone, energizes water circulator. 2) After flow is established through flow switch, boiler is energized. 3) Boiler water temperature maintained according to outdoor temperature. At outside design temperature, water temperature maintained at 160 dF; at 55dF outside temperature, water temperature reset to 80dF. 4) When all zones are satisfied, boiler is de-energized, and pump continues to run for 5 minutes, then is de-energized. Settings may be adjusted as required.

These boilers are very efficient, running 96% - 98% according to manufacturers. I still prefer the 85% cast iron.

g scott

Hi Gents,

what is going on with the conversation/discussion - would you shift the focus to a private discussion about heating systems, boilers and circulation pumps. This discussion will help nothing useful to the first question of the whole discussion.

Back to the theme!!!

Heaters have closed circuits and pumps require pressure capebility only to overcome the flow resistances in the circuit, no pressure is required to lift the water from ground to top of bulding (gravity height is compensated between upstream to downstream lines). Our friend with the strange pressure demand has a fully different constellation: he has an open system and has to elevate the water by 30 meter - and his equipment obviously needs to create 7 bar, to get the flow running. Now tell me the circulation pump which is able to create 7 bar pressure? Unlucky person trying this! For sure he has 4 volumetric discharging pumps, which are able to produce pressure, and which will (!) result in an additive flow rate if connected in parallel (but for reduced energy efficiency due to quadratic with flow rate rising flow resistances).

The problem of our fried with the pipe possibly is not the absolute pressure demand, but he has to satisfy the unexpected high power consumption (7 instead of expected 3.5, maybe 4 bar). And this additional power requirement makes requires too much investments for using renewables instead of petroleum.

Up to know I havn't heard any practical explanation for the pressure drop along the pipe from a person with profund knowledge in pipeline systems. I am not experiences in these things and therefore have pointed on the more theoretical souces for calculating hydrodynmic flows and their resistances (pressure drop). It would be nice if someone from practise could explain WHAT (most probably) is causing the large pressure drop along the pipe. The discussion about enclosed air does not convince me - the arguments do not match with physical properties of air and water. So I guess, these arguments are more opinions/ideas than based on experience.

best regards to all - but keep in the context of the initial question

Albert

Albert,

"We are to design a pumping system for water in a remote place in Africa. .......... The case is that we have very little information from the system today,because we are doing the project from Norway."

Ample effort was initially made to address the OPs question as to possible causes of extra pumping effort being required in the existing installation, for which he is attempting to design a new pumping system.

However as he indicated in the quote above, he himself has no knowledge of the actual pipeline conditions at the site other than the few details he mentions. Without more specifics of the installation, it's all guess work about the source of the discrepancy.

As to the subject of our replies drifting away from the original topic, that is one of the features of this type of forum, especially when the original question has been exhausted. We share information, learn from each other, and have fun in the process.

You are not happy with the trapped air scenario as the only cause, and I am not either, although it might be. To render any specific answer requires more specific knowledge of the pipeline than we have and will apparently ever get.

In the meantime, feel free to join in and talk about heating systems!

Regards, Greg

OLD F**T,

Yes that is the most efficient approach for sure.

At the time I did the conversion, I had my hands full but was aware of the savings from an outdoor temperature sensor to regulate boiler temp. It was one of those "would like to have, but later on when I get a chance ..." additions that was pushed over the edge of consciousness by time and the everyday priorities of family life. I also became aware of European (an Italian model) gas boilers that recaptured latent heat around that time, but my boiler was almost new, and I didn't want to spend a lot of extra money on a foreign boiler using what was then "exotic technology" for a home boiler here.

There was a housing development that had slab houses with the hot water pipes in the slab, like the Levittown development, (the slabs took a long time to heat up, and continued giving off heat long after the hot water flow stopped, which wasted fuel and caused temperature swings in the house some times) only they added an outside temperature sensor to change the duty cycle and timing of the water circulation in the slab, rather than the boiler temperature (they needed a high boiler temp for the tankless hot water supply). These were built in the early 1950s. Unfortunately, the burner servicemen of the day were not familiar with those types of controls and ended up disconnecting them over time.

As I said, this conversation has prompted me to revisit this topic again. At the very beginning and end of the heating season, I never go below about 110° F. boiler temp so I would save throughout the season.

BTW: In windows, and most application software running in it, a degree symbol can be created by <ALT>248. Hold down the ALT key while typing in the number 248.

Regards, Greg

Hi Greg and Old F**T - I installed a condensing boiler about 6 years ago, 15 kW output (~ 50000 BTU/hr) if I remember right (data at home but I'm at work) perhaps a bit small by US standards? Incidentally condensing boilers mandatory in UK now for new and replacement. Honeywell system W controls i.e. 2-way diverter valve to HW cylinder or rads, with HW preference, but not mid-position. Extra controls suggested by Old F**T sound like a good idea but mine has worked OK without. On start, boiler is powered same time as pump, but it takes 10 - 20 secs to do its checks and ignite, giving water time to get going. I did think about an off-delay timer for the pump but never got round to it. The claimed efficiency is ~ 103% based on lower CV.

With ref to control on outdoor temp, Danfoss sell a unit called BEM4000. That was a few years ago when I was thinking of buying one, they may have a later model now. It comprises a strap-on temp switch which opens at variable temp, reducing as outdoor temp rises. When temp switch opens boiler is de-powered but pump continues. Early version had the temp switch on the boiler flow pipe but later it was on the return, they must have found it worked better. Boiler control-stat left at usual ~80°C. If hot water calls temp switch is disabled, reverting to control-stat. Or at least that is what I'd expect but if I remember right it works a bit different and I thought illogically, which is one reason I didn't go for it. Other reason was it wasn't cheap and I didn't think the benefits justified it. Also if the system has a mid-position valve the temp switch needs to be disabled when HW or HW + heating called, which cancels some of the benefit.

On subject of mid-position valves, I've never thought them worth the extra complication, I've had 2-position types for years and not noticed drop in room temp due to HW preference.

Cheers....Codey

PS I agree with your post to Albert, it's often the digressions that are interesting and hopefully we all learn from. Albert's welcome to join in if he wants.

Codey,

You hit on one of the problems with the outdoor sensing equipment: its expensive, co even though it pays for itself, it takes time to do so, and during that time, other investments my yield greater return.

In your case, I would suggest looking into a separate hot water heater. It may not pay, but then again, they typically do because then the boiler can run at a much lower temp, and not at all in the warm weather.

Regards, Greg

Hi Greg - have you considered a run-on timer for the pump? stop the boiler immediately the room-stat is satisfied, run the pump another 5 - 10 minutes to dissipate the heat.

Codemaster,

It's a good thought. I certainly have plenty of delay-on-break time delay relays around. The only problem would be then overshooting the thermostat because one of the weakness in my manually adjusting boiler temperature is that it is always at least a bit higher than it has to be to allow it a margin for the colder nighttimes. The replacement thermostat I recently installed doesn't have much of a "heat anticipator" adjustment, although I could add one in the form of an adjustable resistor located below the temperature sensor and switched on and off with the heat. This is one technique commonly employed in thermostats: the resistor (activated at the ame time as the heat) slightly raises the ambient temperature the sensor reads, thereby causing to to shut off sooner than it would have to take into account the heat remaining in the baseboard or radiators.

Regards, Greg

maybe you can buy a oil burner, it will supply more heat. just have a try....

Centrigual pumps have a limit to how much pressure they can create at a given motor speed.

So if one pump can create 100 psi and a second pump can create 50 psi, water will never flow through the second one because it is being over powered by the first when in parallel.

In the typical parallel pump skid setup the pumps are connected to a common header within 2-3 feet of each other so each pump has the same system pressure applied to it. All pumps are normaly matched and run at the same speed during normal operation and of course check valves are installed at the discharge of each pump. Parallel pumps configurations are typicaly for system reliability not flow capacity. Flow capacity is determined by the pump and motor sizing.