Flow is too high

Joe Mattiello

Centrifugal pumps are variable flow variable torque. As you throttle back on the flow the BHP required is also reduced, causing the pump to run cooler, and more Efficient. The 80 watts on the label is maximum, and not necessarily what you are using.

RoosterBoy

I have a taco 007 for a baseboard loop that only call's for 2gpm flow. and 5 feet of head. the noise I am hearing I think is velocity and not air is there a way to reduce the flow down to 2gpm and keep my pump. I am using 3/4 inch copper.

thanks

Paul Rohrs_4

several options

(Why do you have to keep the 007?)

Options as I see it:

1.Pipe in a differential bypass loop to divert excess flow.

2. Pipe in a balancing valve that will add Head to the pump. In this case you will be on the very high side of the pumping curve which will (technically) shorten the life of the pump. (10.55 feet of head)

3. Change pump to 006 and also balance......

Other options?.....

Regards,

PR

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

hr

A flow setter

would work, but I rather you find a closer sized circ.

With a flowsetter you are just throttling and driving that oversized pump farther off it efficiency peak. And you may just "move" the noise to the flowsetter restriction and the copper will still transmit noise throughout.

Very similar to how a faucer aerator, or a slightly opened bib creates noise that is heard everywhere in the copper plumbing system.

A Grundfos 15-58 on speed one would be a close match :)Handles the velocity issue and less power consumption (60 watts). And power to spare should the system ever be enlarged.

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Paul Rohrs_4

uh.....

I wonder if I should go back and edit my previous recommendations to include a "Flow-setter". I wonder if anyone will notice? That is an excellent alternative.

Good catch Hotrod, Looking forward to buying you your beverage of choice in MN next week. We get in on Wednesday night.


Regards,

PR

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

RoosterBoy

thanks guys
on the grundfos 15-58 what flow rate would it give me on speed one. as compared to what the 007 is giving me now

i used a flow caculator and here is my info

The Fluid Flow Calculator

This is your input:

fluid temp F: 180F
flow rate GPM: 2.6
pipe size in: 3/4
pipe length ft: 130
pipe material: C


These are the results:

velocity ft per second: 1.5674
Reynold's Number: 26931.8724
friction factor fa: 0.0241
friction factor f: 0.0243
head loss ft: 1.7568
pressure loss psi: 0.7400

Jerry Boulanger_2

Speed control

Put a speed control on the 007. Use one that is designed for small motors (typically exhaust fans), not a light dimmer. You will then be able to dial in the flow and save a bit of pumping energy.

hr

Get the pump curve from their website and you

ride the curve to your system spec Make sure you add fittings valves, flow checks, etc to the pressure drop calc, not just the piping length. www.grundfos.com

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Steamhead

My 2-cents

I'd round that 2.6 GPM up to 3 GPM. My three circ choices are:

Taco 006- 5 GPM at 5 feet of head. This one's brass though, which you really don't need to pay extra for, also I don't think it uses standard flange mounts.

B&G NRF-9F/LW Same flow as the 006, but is cast-iron with flange mounts.

Grundfos UP-15-10F, about 3 GPM at 5 feet of head, cast-iron with flange mounts. This one might cut it a bit too close.

The 3-speed Grundfos 15-58F on low speed will give you the same flow as the 006 and NRF-9. The Taco 007 will pump 17 GPM at 5 feet, so that's probably where your noise is coming from. And I'll bet that loop doesn't have much Delta-T with the 007 either!

My first choice would be the NRF-9 for this loop.

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

hr

Back up a step

how many BTUs are you trying to deliver to that space (the calculated heat load)? From this you would calculate pipe size, baseboard length, pressure drop, delta T desired,and finally pump size.

Remember, depending on the piping, you may need to include the pressure drop through the heat source.

Baseboard sizing tables ususlly show output at 2 different flow rates. The lower rate will assure quiet, trouble free operation. Higher output would require more gpm, which increases the velocity, lowering the delta T through the circuit BUT increasing the average temperature across the element for higher output.

So make sure as you downsize the pump you will still get the needed BTU to the circuit.

Currently it looks like you may be moving the water at 3 times the accepted velocity. Design around 4 FPS for quiet operation. Too slow, below 2 FPS will often cause air noise as the fluid may not carry air to the purge point. You could trade velocity noise for a gurgling noise, especially if there are lots of ups and downs in the piping loop

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Maine doug_11

Jerry, what is

the difference between a solid state wet rotor circ speed controller and a dimmer?

Empire_2

Paul R,...

I have seen your diagram and was wondering where you got the program from. Cost and all that stuff. I am currently using a Demo from Taco, seems to be ok but they want $1,700 for the whole thing.. Kind of hi.

jerry scharf_3

Different jerry

Doug,

A dimmer simply reduces the voltage supplied to the motor. The motor then "slips phase" and turns slower with presented load. Some solid state units do the same, using a triac instead of elecromechanical methods.

A solid state controller like you find on an injection pump is a very different animal. The motor doesn't slip. Instead, the controller synthesizes a frequency lower than 60 Hz and the motor turns at a speed that is output-frequency*max-RPM/60Hz. If the output frequency was 40Hz, the motor would turn at 2/3rds of the rated speed. This is known as a Variable Frequency Drive.

hope this made sense.

jerry

Jerry Boulanger_2

Up here in Canada...

...about eight bucks. I don't know the technical difference, but the ones that are rated for fan speed control will handle more current and seem to have a wider operating range than the less expensive light bulb dimmers. I've used both extensively for demonstration purposes and the fan speed control units do work better. My guess is that it will provide longer service as well.

Paul Rohrs_4

Hydronic Pro's

It's from Hydronic Pro's and the mind of John Siegenthaler. It is the HDS software package that let's you do a multitude of things. Pump sizing, expansion tank sizing, my personal favorite is the injection mixing simulator. I attached a JPEG of that module. This software has been out for a couple of years. Money well spent.


http://www.hydronicpros.com/software.php?AD=3661381c285e26337362ca90c2204678




Regards,

PR

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Joe Mattiello

007

The Taco 007 will do a nice job for you, are you sure its velocity noises your hearing? You can balance the system using a ball valve to restrict flow, and have the pump back up its curve. This will cause the pump to run more efficiently in most cases.
Taco, Inc.
Joe Mattiello
Technical Service Technician
joemat@taco-hvac.com
401-942-8000 X 484
www.taco-hvac.com

hr

Joe, I'm not clear

on how throttling the output could increase efficiency. Let's say a circ consumes 80 watts of power and moves 10 gpm at 7 ft of head. This being the knee of the curve or the middle 1/3 of the curve.

Now if you balance down that output down to 2 gpm wouldn't it still consume 80 watts but accomplish much less work, in terms of BTUs moved?

Or does the power consumption drop as the balance valve is closed down?

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Steamhead

I'm interested to hear the answer too

I have never believed in throttling circs. If the circ is sized correctly to begin with you don't need to mess with it.

And Joe- I really do like Taco circs. But the smallest flange-mount unit you guys have is the 005, which would still be too big for RoosterBoy's job according to his figures. I take the same position regarding circulators that I take with boilers: a good circ is a good circ- proper sizing and installation count more than anything.

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

bob_44

Efficiency

Water HP or the amount of power put into the water is GPM*FT HD*sp.gr./3960. For 2gpm@5' that's .0025253 HP. Efficiency is WHP/BHP*100. A Taco 007 is 1/25 hp or .04. .0025253/.04*100=6.3%. A 006 would be 10%. These calculations are made with Mfg. listed HP. In reality, as a pump is throttled the watts consumed should decrease, so you should measure the watt draw with a watt meter(not an amp meter) and figure out how many BHP you are actually producing at operating conditions. 746 watts =1 HP The smaller pumps get less efficient because bearing and internal friction etc. are a bigger percentage in proportion to pump size. A 007 pumping 12gpm@7' is approx. 52% eff. bob

hr

I come up with @ 20%, bob?

If I use the formula I was shown, although I'm not sure about the watts consumption of the 007.

Also efficiency curves seem to agree? Are we talking two different types of efficience. One being energy consumed to work provided "Wire to Water".

Also you said as the the pump is throttled power consumption SHOULD decrease. So it does, or it should

My point is.. since the manufactures have gone to the trouble of making dozen's of sizes of small circs, as the market demands, why would one not chose the pump the matches the application as opposed to throttling one that is two to three sizes too big for the job?

On larger sizes don't the manufactures actually shave the impellers to match the pump to the job?

Thanks for you help in understanding.

http://www.pmmag.com/CDA/ArticleInformation/features/BNP__Features__Item/0,2379,3964,00.html

hot rod



To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Steamhead

Hmmmmmm

I know it's possible to damage a pump by dead-heading it (pumping straight at a closed valve or something similar), which makes it overheat due to the water churning in the volute. Wouldn't excessive throttling do that too?

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

hr

Zen and the art of throttling

I'm sure nothing in excess is good when it comes to throttling.

My other headscratcher is the fact that you should not flow against a partially closed ball valve. Yet some circuit setters are ball valves??

It seems they should all be a cone shaped or tapered plug type of valve. I know some brands are.

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

bob_44

Good Call

Hot Rod, as usual an excellent article by Siggy. The formulas I used are from B&G publication TEH375 which uses shaft HP, that's why I mentioned the watt meter bit. I agree with you 100% about pump sizing. I was surprised the formula in the article is not the same one as in Siggy's book. I love that term "wire to water" very explanatory.

I,ve never heard that you shouldn't pump against a partially closed ball valve. They sell memory stops for them. The old B&G circuit setters were square head plug cocks, the new ones are ball valves. I believe Tacos are also. bob

Steamhead

Ball Valves

like gate valves, can wear unevenly if left partly open. This can prevent them from shutting off completely. But in a circuit-setting application it doesn't need to shut off completely.

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

hr

Certainly there are times when

a flow, or circuit setter, are required. Injection mixing is a classic example. Doubt we will ever see a circ for a 4gpm, .5 head application

Some flowsetters I have used, Danfoss, T&A, Istec for example, have an equal proportion type of valve. A cone shaped valve with a cone shaped seat. Usually with a fine thread stem for real accurate adjustments.

I suspect that sharp edge a ball valve, partially closed, presents to flow in the 4 FPS velocities we commonly work with, will not be a wear problem for many years. Maybe in DHW applications with 8 FPS or more it would be more of an issue?

Unless you are pumping sand

Seems every year pump manufactures add a handful of pumps to their small circ line up, as well as the clever multi speed models. We should be thankful, and take advantage of the ability to get it (the circ pump) sized as close as possible.

I understand pump design gets pretty involved as you change speeds but not the impeller geometery. I've heard this is one reason we don't see multi speed high head circs on the market yet. It's easier said than done

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

bob_44

High Head

I guess that depends on how you define "high head". A three speed Grundfos UPC80-160 will do 50gpm@50' on speed three. I know of a mechanical engineering firm that loves these pumps, they design for speed 2 and then have 1 and three to cover their ###. bob

hr

How high

for my, mostly residential work, needs the Grundfos 26 series gets me by, in the high head range.

The 15-58 is an excellent first step. Perhaps crunch together the 26 series 64, 96, 99, maybe the 100. This would cover the ever increasing need for high head to drive "choked up" condensing boiler HXers, of various outputs Would handle the micro tube QuickTrack type applications also.

Then maybe two pumps to cover low head, high flow. A small model for a variety of injection mixing applications. Then a upper gpm model for primary loops and reverse indirect tank operations.

So a total of 5 different 3 speed pumps.

Same for the B&G PL series. Scrunch together about 12 of those into 4- 3 speed models. I like the PL series for commercial applications where noise is not an issue. The cost of that open frame motor looks better than wet rotors of the same spec!

In a lot of cases I would rather have a fixed speed than a variable speed circ. Takes the cost of the drive and R&D $'s that will be packed into them.

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Joe Mattiello

Cast Iron 006

> I'd round that 2.6 GPM up to 3 GPM. My three circ

> choices are:

>

> Taco 006- 5 GPM at 5 feet of

> head. This one's brass though, which you really

> don't need to pay extra for, also I don't think

> it uses standard flange mounts.

>

> B&G NRF-9F/LW

> Same flow as the 006, but is cast-iron with

> flange mounts.

>

> Grundfos UP-15-10F, about 3

> GPM at 5 feet of head, cast-iron with flange

> mounts. This one might cut it a bit too

> close.

>

> The 3-speed Grundfos 15-58F on low

> speed will give you the same flow as the 006 and

> NRF-9. The Taco 007 will pump 17 GPM at 5 feet,

> so that's probably where your noise is coming

> from. And I'll bet that loop doesn't have much

> Delta-T with the 007 either!

>

> My first choice

> would be the NRF-9 for this loop.

>

> _A

> HREF="http://www.heatinghelp.com/getListed.cfm?id=

> 157&Step=30"_To Learn More About This

> Professional, Click Here to Visit Their Ad in

> "Find A Professional"_/A_

Joe Mattiello

Cast Iron 006

FYI....Taco has recently introduced the Taco 006, cast iron construction, flanged conection. I'm at a remote location so I can not attach the submittal data information. However you can visit taco's latest, and greatest web site at www.taco-hvac.com for all literature on Taco products.

ALH_3

Google

Affinity laws for centrifugal pumps.

Flow varies by the ratio of the rotor speeds.

Head varies by this ratio squared.

-Andrew

hr

my understanding of centrifical pumps

is.. As the efficiency curve I posted above shows, the pump is most efficient at mid curve. I believe the industry calls this BEP (best efficiency point) SOME varience from this point is acceptable.

From my Grundfos training manual.

" The geometry of an impeller is designed for the flow capacity at BEP. When flow rate is decreased below the design capacity, there is excess flow area between between the impeller vanes and flow seperation occurs. When the flow rate is reduced beyond the inflection point towards shut off, eddy type flow patterns occur near the leading end of the impeller vane and also near the exit end of the impeller vanes. This eddy type flow pattern of recirculation can cause severe erosion in the impeller"

I think this would present itself as a noise often associated with cavitation?

My point being, the original question asked about the correct pump for a @ 3 gpm flow at @ 5 feet of head. I don't agree that throttling the pump he currently has is the correct, or even acceptable answer.

I don't think throttling down to near "shut off" has the same result as lowering the motor speed via a pump that has 3 speeds built into the motor.

I'm still not convinced that the pump "becomes" more efficient as it is throttled down. At least the formulas, and the factory efficiency curves don't seem to indicate this?

Reminds me of the movie "Office Space", where the employee taking funds from his company bank account is telling his girlfriend that he is not really stealing that money. The money just "becomes" his

hot rod

To Learn More About This Professional, Click Here to Visit Their Ad in "Find A Professional"

Weezbo

when the Wilo P differential pumps are in our repitoire we

wont be having this disscussion *~/:) and who better than our buddy hot rod has eluded to this very subject from time to time

ALH_3

Pennies for Everyone

Throttling does not have the same effect as reducing the rotating speed of the impeller.

Throttling increases the loss coefficient. It alters the system curve.

Reducing the rotating speed of the impeller changes the operating characteristics of the pump.

Here's a website that illustrates it better than I can.

Controlling Centrifugal Pumps


The "inefficiency" in throttling comes from using an inappropriate pump that has a greater power consumption than is needed and then "losing" it in the valve.

-Andrew

Unknown

Flow too high

Interesting discussion - lots of replys. I think we are all partially correct. The pump effeciency will increase by throtteling it down but only to the point of the pump's BEP. Once we go past that point it's effeciency actually decreases. Power comsumption (wattage) does decrease with lower flows (put your hand over a vacuume cleaner nozzle and the motor speeds up, due to less work).
I like the idea of running the pump on a lower speed (like the WILO Star 16

Unknown

High Flow - good ideas

Interesting chat line. Here is some clarification on some of the points...
1)It's true you can increase the pump's operating effeciency by throttling or backing up the curve but only until you get to the BEP then the pump effeciency drops off.
2)Less flow, less amps, less HP - check it with an Amprobe - it is true!
3)Small amounts of throttling with a ball valve might be OK but a ball valve is not linear making it difficult to adjust. Using this valve for high decreases or high flows is not a real good idea (noise, cavitation etc).
4)Regardless of type, no centrifugal pump likes extremely low flow rates (seals and volutes heat up in dry runners and the cooling/lubrication fluid in a wet rotor can can get boiling hot). This is dependant on how little the flow is, how long it occurs and the size of the pump (bigger pumps create more heat). HI suggests min flows of something like 5 to 10% of the flow at the pump's BEP - some mfg actually publish min flow rates.
5)I recommend slightly throttling all pumps to ensure it stays on it's curve (only my opinion). That's why we have gas peddles on our car - for control within the operating range of the devise. Sure, if the pump is sized right no need for throttling but when was the last time the calculated friction loss was the real one?

The WILO Star S 16 wet rotor pump, on speed 1, at 2 USGPM we deliver slightly over 5' TDH.

See you all at Wetstock...