Two pumps connected in parallel runs continuously

[Julian]

Hello everyone, I'm a Plumber/ pump man from the Philippines. I need advice regarding pump problem. I dont know if this the right forum, but here it is. Two pumps connected in parallel runs continuously. What puzzled me is that whenever I partially closed main discharge valve or any of the discharge valve of both pump the pressure drops. Now this my first encounter with such a pump. Please refer to attached files. Is it perhaps a variable speed/ discharge pump that reacts to the pressure transmitter installed at the discharge. Pump is servicing a four story building. pressure is 70 psi drops to 50 to 60 whenever I partially close any discharge valve. Looking forward to your advice.

 

[John.g]

Re your query of 50hz vs 60hz, yes a 50 hz motor will run at 60 hz but not as efficiently and there are greater electrical losses as well apart from mechanical stresses on the windings due to the greater centrifugal force and if the pump is running at end of curve conditions (flat out) then the motor can trip out on o/load as the power required will be 1.7 times higher, ((60/50)^3)). Motors can be designed for 50/60 hz operation, don't know if the Grundfos one installed is of this type but it certainly has the power, 3KW. I can,t find the pump curve for the (60 hz) installed pumps but the above curve (post #34) shows that the maximum required power at 50 hz is ~ 1.56 kw so it would still be ok at 60 hz as it would be drawing 2.7 kw ((1.56*(60/50)^3) so maybe these motors are designed for dual frequencies. I wouldn't bother changing it withourt at least contacting Grundfos for advice. I reckon that based on that 8.8 m3/hr and assuming running the VSD to maintain a constant pressure of 50 psi (3.5 bar) that the savings would be ~ 0.6 kwh/hr, 14.4 kwh/day or 5256 kwh/annum.
I estimate that at 80 psi the 60 hz pump is supplying 8.8 m3/hr with a motor load of 2 kw.
No comment needed on your tests as everything is adding up now.
You could do a quick test on the gland to confirm the leak by shutting the suction valve and cracking open the discharge valve which will put full system pressure on it.

I would be grateful if someone can post a link to the 60 hz pump as shown in the label photo as I can't locate it on the Grundfos site.

 

[snowhead]

I would be grateful if someone can post a link to the 60 hz pump as shown in the label photo as I can't locate it on the Grundfos site.

If logged in on Grundfos you need to change the Location to the Phillipines to get 60hz info;

60hz, shaft speed, output etc matches, but the product number doesn't;

CR 10-4 A-A-A-E-HQQE - 96503187 | Grundfos product center

Sizing and selection tools, quick dimensioning, pump replacement tools, CAD drawings, pump curves and other relevant data.

product-selection.grundfos.com

 

[John.g]

Thanks for that, for some reason or other it doesn't seem to be accepting my editing but I have the above now anyhow.

OK, able to edit/change now.

 

[Julian]

Re pump 1 leakage, that will be the mechanical seal in the top housing, the top of it can be seen if you remove the coupling guard but the bad news is it is a complete strip down of the pump to replace it. Could this be the reason pump 1 has been taken out of operation? Unless you have experiance and the tool I wouldn`t recommend you attempt the rebuild tbh.

I do have countless experience overhauling pumps, but not of this kind, almost all of them different from each. Thanks for the suggestion. I will think about it. Basically it all depends on the price, if the owners will agree. New vs rebuild cost.
[automerge]1574014813[/automerge]

http://hydro-tech.com.my/wp-content/uploads/woocommerce_uploads/2019/06/Compact-R-Brochure.pdf
50hz, but the basic system will be the same.

Info on CR pumps

http://daparak.com/wp-content/uploads/2013/11/CR-CRI-CRN-CRT-Installation-Operations.pdf

I have read the manuals for the VSD's you provided. Is Rhymebus same as grundfos? also I cant find the programming manual. Its only for installation.

 

[Julian]

Re your query of 50hz vs 60hz, yes a 50 hz motor will run at 60 hz but not as efficiently and there are greater electrical losses as well apart from mechanical stresses on the windings due to the greater centrifugal force and if the pump is running at end of curve conditions (flat out) then the motor can trip out on o/load as the power required will be 1.7 times higher, ((60/50)^3)). Motors can be designed for 50/60 hz operation, don't know if the Grundfos one installed is of this type but it certainly has the power, 3KW. I can,t find the pump curve for the (60 hz) installed pumps but the above curve (post #34) shows that the maximum required power at 50 hz is ~ 1.56 kw so it would still be ok at 60 hz as it would be drawing 2.7 kw ((1.56*(60/50)^3) so maybe these motors are designed for dual frequencies. I wouldn't bother changing it withourt at least contacting Grundfos for advice. I reckon that based on that 8.8 m3/hr and assuming running the VSD to maintain a constant pressure of 50 psi (3.5 bar) that the savings would be ~ 0.6 kwh/hr, 14.4 kwh/day or 5256 kwh/annum.
I estimate that at 80 psi the 60 hz pump is supplying 8.8 m3/hr with a motor load of 2 kw.
No comment needed on your tests as everything is adding up now.
You could do a quick test on the gland to confirm the leak by shutting the suction valve and cracking open the discharge valve which will put full system pressure on it.

I would be grateful if someone can post a link to the 60 hz pump as shown in the label photo as I can't locate it on the Grundfos site.

If the motor is designed to run at 50/60 hz, would not it be indicated on the label. I see only 50hz there and not 60/50hz. and the other one 60hz.
Thanks for the computations. I am wondering though, this is a water supply system for tap, for hotel guest use, How is it suppose to enable power savings if it keeps on running even if no one is using the tap. Is not the on/ off system with a big pressure tank and pressure regulators on each floor, more suitable for the purpose.
[automerge]1574015590[/automerge]
"You could do a quick test on the gland to confirm the leak by shutting the suction valve and cracking open the discharge valve which will put full system pressure on it".

Good suggestion this I will do. I am also planning to do a quick test on the motor via emergency switch, what do you think?
[automerge]1574016936[/automerge]
But before I take out pump one I want to reduce the speed of pump 2, as I can sense an increase in vibration and hear what seems water hammering, every time I close fully valve 1. So I leave it crack open to ease pressure a bit. I am thinking I can do speed modulation by adjusting VSD2 frequency. that is why its important that I learn first to program this thing.

 

[snowhead]

If the motor is designed to run at 50/60 hz, would not it be indicated on the label. I see only 50hz there and not 60/50hz. and the other one 60hz.

Yes it should be so most likely isn't suitable for 60hz.

Thanks for the computations. I am wondering though, this is a water supply system for tap, for hotel guest use, How is it suppose to enable power savings if it keeps on running even if no one is using the tap.

The pressure sensor / transducer in the pipework should detect the pressure and signal the VSD to reduce speed or shut down.
Either the sensor isn't working, the VSD isn't working or the settings have been changed so it only runs at full speed.

But before I take out pump one I want to reduce the speed of pump 2, as I can sense an increase in vibration and hear what seems water hammering, every time I close fully valve 1. So I leave it crack open to ease pressure a bit. I am thinking I can do speed modulation by adjusting VSD2 frequency. that is why its important that I learn first to program this thing.

See above, it may be the sensor has failed and the VSD is manually set to full speed or it's set to full speed anyway.

If so you should be able to reduce the speed setting on the VSD.

I suspect this boost system has had several faults which have accumulated over the years with various people attempting to fix and have failed, most likely through lack of knowledge.

You need to be prepared to cap the open ends of the pipes from the pump you are removing , it's quite likely the valves will pass some water when closed.
Or if you can't reduce the other pump speed you may be able to pipe some of the flow to a drain to reduce the pressure temporarilly through the removed pump valves.

 

[John.g]

Yes its strange alright to have a 50 hz motor and as well as that it can run on low voltages between 220 to 240 and high voltages of 380 to 415. I would certainly ask Grundfos if this was installed by mistake as it doesn't make much sense. Can you post a photo of the 60 hz motor nameplate.

I have put together a simple spreadsheet of the VSD which shows the effect of changing the speed which you can input yourself, you may find it interesting.

 

[Julian]

Yes its strange alright to have a 50 hz motor and as well as that it can run on low voltages between 220 to 240 and high voltages of 380 to 415. I would certainly ask Grundfos if this was installed by mistake as it doesn't make much sense. Can you post a photo of the 60 hz motor nameplate.

I have put together a simple spreadsheet of the VSD which shows the effect of changing the speed which you can input yourself, you may find it interesting.

Wow, interesting indeed, I am getting a total hands on education, another proof that formal education needs total overhaul. Our children need not waste too much time in school studying mostly nonsense.
Ok back to topic 🙂
First I wanna make sure that you did not mixed up constant pump speed and variable pump speed. Looking at the spreadsheet, briefly ( I need to go to work) I got the hunch that I should set the system at desired speed, corresponding to the building height, and then put it on "on/ off mode" so that it stops when no one is using the tap and ran on that sweet spot rpm whenever it does needs to ran.

 

[John.g]

Wow, interesting indeed, I am getting a total hands on education, another proof that formal education needs total overhaul. Our children need not waste too much time in school studying mostly nonsense.
Ok back to topic 🙂
First I wanna make sure that you did not mixed up constant pump speed and variable pump speed. Looking at the spreadsheet, briefly ( I need to go to work) I got the hunch that I should set the system at desired speed, corresponding to the building height, and then put it on "on/ off mode" so that it stops when no one is using the tap and ran on that sweet spot rpm whenever it does needs to ran.

No, I didn,t mix up the two, my spreadsheet (and graph) is dynamic and changes as long as you replicate (by inputting the speed) that the VSD will need to output at the CHANGING flow requirements for any given setpoint (SP) head.
If you set the VSD speed to give you just the desired static head (height) then the top floor will have zero flow, if the building height is say 40 ft or 17 psi you must then add on the required pressure required by the resident say 30 psi so the pump head must be 47 psi, say 50, so the top floor resident will have 30 psi available at his taps/shower and the bottom floor resident will have 50 psi available, pipe friction losses must also be taken into account so IMO the correct way to run this set up is to run on pressure setpoint and maybe also use some logic to start/stop the VSD when demand is zero but VSD is primarily designed for a constant varying demand. however, yes, you can pick and set a sweet point speed based on the "average" flow demand and start/stop the pump using the pressure transducer or simply installing a pressure switch, all very very interesting.
I have also modified the spreadsheet to give a little more info, it graphically now shows the pump+motor efficiencies, just use the flow (x axis) as the reference, the efficiency will be directly above it, on the red trend.

 

[Julian]

No, I didn,t mix up the two, my spreadsheet (and graph) is dynamic and changes as long as you replicate (by inputting the speed) that the VSD will need to output at the CHANGING flow requirements for any given setpoint (SP) head.
If you set the VSD speed to give you just the desired static head (height) then the top floor will have zero flow, if the building height is say 40 ft or 17 psi you must then add on the required pressure required by the resident say 30 psi so the pump head must be 47 psi, say 50, so the top floor resident will have 30 psi available at his taps/shower and the bottom floor resident will have 50 psi available, pipe friction losses must also be taken into account so IMO the correct way to run this set up is to run on pressure set point and maybe also use some logic to start/stop the VSD when demand is zero but VSD is primarily designed for a constant varying demand. however, yes, you can pick and set a sweet point speed based on the "average" flow demand and start/stop the pump using the pressure transducer or simply installing a pressure switch, all very very interesting.
I have also modified the spreadsheet to give a little more info, it graphically now shows the pump+motor efficiencies, just use the flow (x axis) as the reference, the efficiency will be directly above it, on the red trend.

I think you are correct to run the system on pressure set point rather than speed set point, because of the wide varying demand. With this I think the VSD won't be intelligent enough to control or stop via pressure transducer, because it will be running on constant pressure. Unless minimum RPM can deliver a pressure higher than set cut out pressure. All in all I think it would be simpler for me to install a pressure switch in between pump and VSD. And install a bigger pressure tank to minimize start stop cycle. Or if I choose to go further I can also combine a flow switch. As long as water is flowing in the pipe, pump will continue running even if cut out pressure is already attained.
If I may ask how do you compute efficiency based on the given data.

 

[John.g]

There are probably tried and trusted ways of starting/stopping the VSD as there are tens of thousands of these systems about I would think and eventually we will find the answer.

Re efficiencies. Ref Grundfos link in post #66. If you click anywhere in the graph you will all the see the numbers that I inputted in the spreadsheet in columns B to F "under constant pump speed" I clicked on flowrates 0 to 15 ms/hr in 1 m3/hr increments to get all the actual pump conditions in these 16 flowrates.
I then did the calculations in the 16 column table under "variable pump speed" and you can see the formula's used in each of these columns, B to E, and are based on the pump affinity laws (well known) where head is proportional to speed squared, flow is proportional to speed and power is proportional to speed cubed. The efficiencies will remain the same in each row as all the other conditions even though changing will give the same efficiency for each equivalent row as in the full speed table and that's why there is no calculation done in column F.

 

[Julian]

There are probably tried and trusted ways of starting/stopping the VSD as there are tens of thousands of these systems about I would think and eventually we will find the answer.

Re efficiencies. Ref Grundfos link in post #66. If you click anywhere in the graph you will all the see the numbers that I inputted in the spreadsheet in columns B to F "under constant pump speed" I clicked on flowrates 0 to 15 ms/hr in 1 m3/hr increments to get all the actual pump conditions in these 16 flowrates.
I then did the calculations in the 16 column table under "variable pump speed" and you can see the formula's used in each of these columns, B to E, and are based on the pump affinity laws (well known) where head is proportional to speed squared, flow is proportional to speed and power is proportional to speed cubed. The efficiencies will remain the same in each row as all the other conditions even though changing will give the same efficiency for each equivalent row as in the full speed table and that's why there is no calculation done in column F.

Tried the spreed sheet. this is fun I am gonna be late for work again hahaha. It certainly is magic to an unlearned person like me. How did you do it. Well anyway anything that you / I know is simple the opposite is true as well.

 

[quality]

what does 1440 refer to?

rpm .

 

[John.g]

Thanks, the pump will deliver ~ zero flow at 1440 RPM to the top floor if its ~ 31 feet (13.6 psi) from the ground.

 

[Julian]

I was looking at the spreadsheet of grundfos 60hz and I notice how efficiency and flow rate increase/ decrease with just a slight change in pressure. I think I know already what to do.
I was there today tinkering with the VSD but before I did that I switch the emergency switch on so that pump runs continuously if I f*** up. Then suddenly the pump stopped. Luckily I had the presence of mind to push the run button. Is that supposed to do that? is not the emergency switch meant to bypass the VSD? I also flip the emergency switch for the standby to check if its running and nothing happened. So i wired it to bypass the VSD and when I switch on the motor run. Did the previous technician F*** up the installation or am I missing something?