Sciencemadness Discussion Board - Increasing water pump pressure using 2 pumps in series?

Increasing water pump pressure using 2 pumps in series?

RogueRose - 25-7-2018 at 16:14

Now I know in electronics running 2 identical power supplies in series doubles the voltage. If you take 2 air compressors and do the same, you get a higher PSI (up to a point where the seals/cylinders or motor can't take it). But when dealing with water, i'm wondering if this is different because water isn't compressible.

I have been looking at those "high pressure" chinese water pumps (12v) that claim anywhere from 100PSI to 160PSI and from 3.5 to 8L/min (the 8L is the 160PSI, so that is hard to explain) and all of them look identical for the most part. They claim 70-90 watts and 6A, some list 14.4v so that may be why some list 90 watts. These things weigh the same (off by a couple grams or so) and priced from ~$11 - 25 for what all look like identical or real damn close.

Someone on here has used one as a circulation pump and maybe for vacuum (aspirator suction). Some said that there just isnt' enough pressure from these to give great vacuum, so I was wondering if putting 2 in series would increase the pressure while maintaining the same/similar flow rate.

I need one for the same thing, but I also need something like this that does a TRUE 180-250PSI (would be happy with 2-4L/min) and any pump near this is crazy expensive in comparison.

I've considered old electric pressure washers. I have 2, one is a 1100PSI Coleman Powermate which has a pretty nice/beefy motor (15A breaker built in) and looks kind of solid but I haven't tried it yet. I'm wondering if I ran full bridge rectified AC (with smoothing caps) through this with PWM to change the flow rate (and keep high tourque & pressure hopefully) instead of a resistance potentiomenter/rheostat to slow the motor (loss of lots of tourque this way). I could get high pressure and decresed flow with the "DC" and PWM with this. Just a thought.
(note, I don't like idea of 120vdc running around water though)

I also have a 1800PSI electric which I haven't bothered taking apart, told it works, and have been wondering if either of these might be fun "not making a flame thrower" (ignition source is already completed and 100% working nicely, lol).

On a final note, I've seen these same cheap chinese pumps mentioned above (70-90 watt) in 110v version and they are about 10-12x more in cost, about $90-120! I can't imagine any reason making it run on 120 would be so much more expensive.

JJay - 25-7-2018 at 16:19

Pumps have velocity limits. It might work well for pumping water upwards a long distance.

MultiplePersonality - 25-7-2018 at 16:28

I don't know all the details, but my instict tells me that's impossible. You can't just put anything in series, thinking you put it in series! What means in series? How do you know it is really set up in series? I could burn 2 fires, and this will only give me more surface area of same temperature as one fire. Ask yourself is it really in series like you think it is? Common sense, instict, obvious. I mean even theoretically it is unimaginable.

JJay - 25-7-2018 at 16:52

Quote: Originally posted by MultiplePersonality

I don't know all the details, but my instict tells me that's impossible. You can't just put anything in series, thinking you put it in series! What means in series? How do you know it is really set up in series? I could burn 2 fires, and this will only give me more surface area of same temperature as one fire. Ask yourself is it really in series like you think it is? Common sense, instict, obvious. I mean even theoretically it is unimaginable.

Oh, that's not common sense at all. Do you have a PhD?

MultiplePersonality - 25-7-2018 at 16:58

JJay, try to imagine, and you will see it's impossible. Pressure is relative, it happens only between inside and outside of something like wall. Either reduce outside or increase inside, or both. Yeah, I have D. P., N. D., D. C., D. N. T., D. N. Sc., D. N. Ph., D. N. Litt, Ph.D., D. Orthp., among many other entitlements I forgot already.

And compare that with fire. You can burn whole forest in normal furnace and never melt even a bit of furnace, or you can melt a bit of furnace using small piece of wood for 5 minutes. That would be series in fire.

Most things are parallel and diluted in life, because as you increase pressure, you increase area too. Like 2/2 = 4/4, same shit. There are many ways to increase temperature or pressure:
1.) Stop increasing area! For example make stronger pressure but on same previous area or volume.
2.) Increase rate of reaction or combustion or current or motor speed...
3.) There are many ways...

[Edited on 26-7-2018 by MultiplePersonality]

RogueRose - 25-7-2018 at 17:11

"How do I know it is in series" well b/c it is connected as such.

Water flows into the infput of pump A increases PSI/flow and out the output of the pump where it is being pushed by more water coming from the pump. This water reaches pump B's input at an elevated pressure and speed, pump B does it's thing, increases PSI/flow and exits. Now is there an over-all gain in this setup? IDK, especially b/c of the properties of water.

Now do the exact same thing with air (and with 2 pumps that can handle higher than stated pressure/flow rates) and I think you will find that the pressure coming out of pump B is of higher PSI than pump A. The thing with this example b/c air can be compressed, is that pump A may need an output flow rate equal to that of the input required of pump B otherwise pump B will be in short supply of pre-pressurised air.

streety - 25-7-2018 at 19:12

I think this is going to depend on the type of pump. If you are looking at something like this I think you would be better with putting them in parallel.

This is a positive displacement pump meaning it moves a fixed volume of fluid on each cycle. If your outlet is closed it will keep pushing on the liquid until its maximum pressure is reached at which point either something breaks or the motor stalls. I'm not sure but I don't think the pressure of the inlet liquid will help as it is blocked off while the pump is discharging.

For the aspirator you need at least 50 psi but you also need a relatively high flow rate which for a positive displacement type pump is a hard limit. Using them in parallel will increase the flow rate.

For velocity type pumps (e.g. centrifugal) running them in series would help with the pressure.

The pressure washers sound like compelling options.

elementcollector1 - 25-7-2018 at 19:37

To give a general answer, yes: Putting two water pumps in series will give a little less than double the pressure. Running two pumps in parallel will give roughly double the flow rate. Of course, the former assumes that all relevant parts in your water pumps can handle the increased pressure.

This is true for centrifugal pumps, but I'm not sure if it's true for every type of pump - I would imagine certain configurations inherently limit the amount of pressure they can output, and were designed as such.

streety's absolutely correct in that you need both high pressure and flow rate. I abandoned my aspirator in favor of a cheap $70 vacuum pump due to this very reason (I was unable to find a water pump that met both qualities), and have frankly never looked back.

Twospoons - 25-7-2018 at 20:32

The risk you take is that the housing and seals of the second pump will experience double the pressure they were designed for. With a high quality pump I would not expect this to be an issue, but with a cheap pump (ie built to minimum specs to save on materials) ... i'd be standing behind a wall when you fired it up.

You might also find a displacement pump (ie piston type) wont work at all, as the non-wetted side of the piston is going to be at atmospheric pressure, so the motor is going to be working against double the force it was designed for.
Centrifugal pumps don't have that limitation.

[Edited on 26-7-2018 by Twospoons]

Fulmen - 25-7-2018 at 21:56

The two last posts cover the salient points. It will work on a centrifugal pump, but not on positive displacement pumps like the ones discussed here. Since the piston works against atmospheric pressure it is limited by the pressure difference between the atmosphere and outlet pressure.
One way to sidestep this would be to use a submersible pump as the second unit. If contained in a closed vessel pressurized by a first pump it will work. One possible problem would be the risk of water leaking into the pump as the casing isn't designed for high external pressure.

Sulaiman - 25-7-2018 at 23:03

If you want vacuum for distillation then I suggest a (relatively) cheap HVAC rotary vane pump,
e.g. https://www.ebay.co.uk/itm/2-5CFM-Industrial-Professional-1-...

If just for vacuum assisted filtering there is no need for extreme vacuum,
too low a pressure is actually a problem as it will cause solvents to boil.
I really like my little vacuum pump for filtration;
https://www.ebay.co.uk/itm/DC-12V-24V-80Kpa-Micro-Piston-Vac...

wg48temp - 26-7-2018 at 03:52

Quote: Originally posted by Sulaiman

If you want vacuum for distillation then I suggest a (relatively) cheap HVAC rotary vane pump,
e.g. https://www.ebay.co.uk/itm/2-5CFM-Industrial-Professional-1-...

If just for vacuum assisted filtering there is no need for extreme vacuum,
too low a pressure is actually a problem as it will cause solvents to boil.
I really like my little vacuum pump for filtration;
https://www.ebay.co.uk/itm/DC-12V-24V-80Kpa-Micro-Piston-Vac...

A detailed perspective on gravity filtering versus vacuum assisted filtering

Say you are gravity filtering a liquid and the initial depth of the liquid is 100mm with only a few mm of precipitate. The filtering rate is approximately inversely proportional to the pressure differential across the filter and filter cake. So for a water solution 100mm deep that is a differential pressure of 10mbar (10m=1bar) or an average of 5mbar. Lets assume it takes 20 minutes to filter.

Suppose we then use a cheap piston vacuum pump or home made venturi pump that can only pull a vacuum of 100mbar (relative to atmospheric pressure) that’s a differential pressure of 105mbar so it will now take approximately less than one minute to filter. Most vacuum pumps even cheap plastic ones or venturi pumps can easily pull a vacuum of 500mbar which would take about 20s to filter, Even with a perfect vacuum the maximum possible of 1000mbar it will still take about 10s to filter and can freeze the filter as the water will boil/evaporate, not a good idea.

Recapping:
Gravity filltering 1200s
Cheap vacuum pump assisted filtering(£15) 60s
Average vacuum pump assisted filtering(£25) 20s
New rotary vane pump assisted filtering(£100) 10s

Sulaiman - 26-7-2018 at 04:44

plus, it takes time to collect batches of filtrate, filtrand etc,

If you do spend a significant amount of cash on your vacuum system then
have a vacuum switch to rest the pump when not required
(which is most of the time if no air leaks)
I consider my little pump to be almost disposable so just leave it running
(up to days continuously for some purposes - it won't die !)
--------------------------------------------------------------------
Related to the above question,
I destroyed the bearing of a pump like the one above by putting two in series,
also the motor could not self-start if full vacuum was present.

weilawei - 26-7-2018 at 06:54

Assuming you stay within the max pressure of the pump, you can get better suction by going from a larger inlet to a smaller outlet (preferably making entire change in diameter at the latest possible point). When you decrease the diameter, you increase the fluid velocity which is conserved by a drop in pressure--your vacuum source thanks to Bernoulli's principle. Also, re-expanding the flow in a controlled fashion will minimize pressing against the atmosphere/outlet liquid uselessly. See de Laval nozzles for more on that idea.

woelen - 26-7-2018 at 07:06

It could work if the second pump takes a flow, lower than that can be delivered by the first pump. At the output of the second pump (or in the pump itself) there must be some flow limiting mechanism. If the pump itself limits flow, then the output pressure will drop and may go even below the output pressure of the first pump.

A system of two pumps in series is not as simple as a series connection of two voltage sources (e.g. two batteries). In general you cannot easily put things in series in all cases. E.g. putting two current sources in series (in the electrical domain) does not work, current sources can only be put in parallel. Pumps are not ideal current sources (flow sources), nor ideal pressure sources. They are a mix (electrical equivalent would be a current source with parallel resistance or voltage source with series resistance), which can be described by means of a Thevenin equivalent which can be seen as pressure source, in series with a flow resistance (thin pipe). When you put such things in series, then the final pressure and flow may be a more complicated function of pump characteristics and properties of the network, attached to the output.

zed - 26-7-2018 at 13:23

In series? Sure. Why not? Provided the pumps and parts, can survive the pressure differential with the outside world.

Pressure washer?

Made for water. Some, might even have problems with hot water.

Might have parts, that can't withstand solvents.

The airless paint sprayer, may be more useful. Less volume per minute, but more immune to chemical attack. Same high pressures. Some units, have stainless steel guts.

Dr.Bob - 30-7-2018 at 17:25

Quote: Originally posted by wg48temp

Say you are gravity filtering a liquid and the initial depth of the liquid is 100mm with only a few mm of precipitate. The filtering rate is approximately inversely proportional to the pressure differential across the filter and filter cake. So for a water solution 100mm deep that is a differential pressure of 10mbar (10m=1bar) or an average of 5mbar. Lets assume it takes 20 minutes to filter.

Vacuum filtration does not work that way. Maybe changing the vacuum from 380 psi down to 190 psi would double the speed in some cases, but the speed of flow is simply related to the difference between the ambient pressure and the vacuum in the filter flask. So there is little difference between 10 and 20 psi in most cases. And as mentioned, once you get to the vapor pressure of the solvent, it will boil, and little if any increase in speed will occur.

If you have a tough filtration, the key is to use a larger diameter filter, so that you get more flow. Changing the vacuum can't do much, as it can NEVER be over 1 atm of difference. You can also pressurized the input, which can allow much higher pressure differentials than 1 atm. But that takes special equipment.

wg48 - 31-7-2018 at 08:00

Quote: Originally posted by Dr.Bob

Quote: Originally posted by wg48temp

I am not certain what you think is incorrect. If you mean the flow rate of the filtrate is not inversely proportional to the differential pressure across the filter cake and filter you are correct. It should have read the flow rate of the filtrate is directly proportional to the pressure across the filter cake and filter (Darcy’s Law). It’s the time for a given volume of flow that is inversely proportional to the differential pressure.

What I was attempting to do was explain that because the pressure differential due to the height of the liquid in a typical laboratory set up of gravity filtration is so low(less than 10%) compared to a vacuum assisted filtration, almost most any vacuum pump will reduce the filtration time very significantly (less than 10%) compared to plain gravity filtration.

Yes the filtrate flow rate is directly proportional to the surface area of the filter (again Darcy‘s law). I have a 80mm diameter filter funnel to achieve the same gravity filter time as my vacuum pump (achieves about 0.5 bar) in the example of gravity filtration I gave would require a filter funnel with a diameter of about 300mm !!!

PS 1 bar is 15psi or 10m head of water

macckone - 31-7-2018 at 22:57

A surge tank with relief valve in between the two pumps will solve plethora of problems. In series with this setup the only requirements are that the second pump be able to handle the inlet pressure and that the flow rate of the second pump not exceed the flow rate of the first pump. This setup will increase pressure but not flow rate.

Sidmadra - 31-7-2018 at 23:32

I would worry two pumps in series could cause the second pump to become damaged depending on whether it's mechanism is producing a continuous stream or a pulsating stream. A less worrysome approach would be to set up the pumps in parallel with check valves, that way, you don't need to worry about one pump applying pressure to the internals of the other. It would require some additional fittings but would be safer for the pumps I think.

weilawei - 1-8-2018 at 01:54

This is why centrifugal pumps are nice. You can often pump utter junk for a fluid, and running dry isn't immediately damaging (though it'll overheat the pump). You can also run with the outlet blocked, for a short time, with the same overheating caveat. They're not positive displacement, so nothing's going to have a mechanical meltdown on flow being stalled.

And there shouldn't be much issue running them in series as long as they're compatibly sized. My only small centrifugal pump is a sump-style pump I have as the coolant pump for my chiller. Mostly I've used the larger monsters sized for 3" pipe and up, but I believe fountain pumps are often centrifugal design.

[Edited on 1-8-2018 by weilawei]