Unequal circulators in series

Turbo Dave

Hi all. I am being asked to add a Grundfos 26-150, in series, to an existing Wilo Star 21F, to overcome excessive head.  These pumps have dramatically different performance curves. My inclination is that the Wilo Star would be immediately overcome by the Grundfos and it's existence in the system rendered irrelevant.  I am skeptical of this recommendation from the "engineering" dept. Shouldn't circulators plumbed in series, or parallel, have matching performance curves? Your input please.  Thanks, Dave

Jamie Hall

The combination

can work, but figuring out just what each pump is doing and creating the combined flow vs. head curve is a little tricky.  What needs to be done is to add the heads for each pump at the same flow rate across the performance curve.  That will give you the combined head.



A caution, however (I haven't looked at the pump curves): if one pump's maximum flow  at low heads is significantly different from the other's, and the system operates in that range, you will have trouble.  The one with the lower maximum flow must be downstream, as there will be a significant headloss across that pump (which will be acting as a turbine and overspeeding) which could (probably would) cause cavitation in the other pump -- and may cause problems with the motor as well.

Zman

Engineering?

I assume you are referring to the fancy new name they have for Maintenance Guys?

I would think it is a bad idea. Why not evaluate the system and put in the right size circulator.

Carl

HDE_2

Whenever

I read or hear of these situations I picture 1,000 to 2000 foot radiant loops buried in concrete. Why? Because it usually is. "They told me to put 1,500' of tubing in the basement"

SWEI

One valid application

is for drainback solar in a multi-story building.

Gordy

Parallel or series

When pumps are plumbed in parallel their flow rates will be additive with both pumps at the same head.



When pumps are plumbed in series their head will be additive when both pumps have same flow rate.



What do you need to do? Increase flow rate, or overcome more head.



Or just make it simple and get the right size pump.

Mark Eatherton

Horseless head men...

I HAVE seen it done before, although none of the pump manufacturers recommend it.



I've seen a 007 pumping into a 009.



If I absolutely had to do it, and had no other alternatives, I'd probably put the smaller pump into the larger pump to avoid the possibility of turbineing the smaller pump.



Sounds like someone screwed up and are trying to cover their collective butts.



Again, it is NOT recommended. Too many variables.



ME

Turbo Dave

flow vs. head

The Grundfos moves triple the amount of flow at a given head, on it's lowest speed, vs the Wilo. Overspeed is a guarantee. This is for an evacuated tube solar/shell & tube pool HX application. The solar company sold the customer the Grundfos pump to correct a compliant  of low heating performance. I believe they have false data and/or an incorrect pump sizing formula. I'm going to obtain the actual Cv and see what pump best matches the desired flow rate. Thanks for confirming my concerns.

hot_rod

two pump drainbacks

are very common. determine the lift and required flow and series two identical pumps to give you the required flow and head.



Most solar controllers have variable speed functions and usually one high head circ will work as ir speeds up for fill and modulated after the siphon is established.



A Grundfos 15-100 covers a lot of DB applications.

Turbo Dave

Pump application

The application is a non-drainback solar pool heating system. The pump(s) provide flow through a shell & tube heat exchanger. When the system was commissioned nearly 2 years ago, the Delta T and flow rate were within mfg. specifications. The mfg. now says their engineering dept. has been updated and they have discovered that they sold the customer several incorrect components, resulting in unsatisfactory performance. I am flying across the country to install the new items and restore the system to operation in 2 months. Their current recommendation is the 2 pumps I listed in series. I strongly question this choice. Once I get Cv of the heat exchanger, I can calculate total head loss and start over with pump selection. I just wanted to start by verifying that any 2 pumps plumbed together should be equal in performance curves rather than widely off as they propose. I believe this is basic, but I am questioning their engineers, so want to explore the possibility that I am wrong

SWEI

Still don't see the benefit

of two pumps in series, unless conditions require one pump some of the time and both at other times.  Is the head so high that no single pump will do the job?  That's another valid application.



I would think the labor to plumb, mount, and wire an additional pump would exceed the cost of simply replacing the existing pump with one of the proper size.

remodel

pumps

Parallel installs are done all the time, especially on larger water delivery systems.  Always match curves to get an additive flow rate, so one pump is not fighting the others head output.  Once flows are added you have to recalculate the head requirement at the additive flow rate and iterate.  Rarely in series at the main pump station, but in series as a booster station.  Good to question this, it seems like two bigger pumps in parallel might be a better solution if there is not one large pump to overcome head and get the flow you need.      

hot_rod

cost

if in fact the job does require high head, two 15-58 @$150.00, one 26-99 @ $250.00 (online prices)



Double the 15-58 you get about 34', while a single 26-99 is 29'



But these days you should be able to find a single pump solution, and plenty of VS options.

Turbo Dave

thanks

Thanks for all your input. I feel much better about challenging their choice and selecting a correctly sized pump. I'll report back with the final result. Dave

EBEBRATT-Ed

Pumps in parallel double the flow??????

I don't think so. When the 2d pump is brought on the flow will increase but so will the head on both pumps so the original pump has its flow reduced by the rising head. If their the same pumps they should pump equally but not at double the flow of one pump.



Remodel has it right

Gordy

This explains it well.... Good site

http://www.engineeringtoolbox.com/pumps-parallel-serial-d_636.html