Inquiry sent to Xylem today - Update with new circ (Grundfoss Alpha)

EricPeterson

Here is a request for technical support I sent today. No response so far:

My boiler system uses a NRF-22 B&G pump. On cold startup the pump is circulating water in a tight loop through the boiler and the near-boiler piping. During this time the pump makes a slight whining noise.
Can you please tell me if this condition is damaging the circulator? Details below,

This is a converted gravity system with a cast-iron boiler.
To protect the boiler from low-temperature return water, a Caleffi 280 thermostatic valve has been installed. This valve mixes heated water from the boiler with return water from the system based on a setpoint of 140F.
During cold startup the valve is closed to return water for about two minutes, during which time the NRF-22 is circulating water through the boiler and the near-boiler piping, which works out to circulating 7 gallons of water through about 20 equivalent feet of 1-1/4" black iron pipe.

Thank you,
Eric Peterson

GGross

Do you have a picture of the installation?

EricPeterson

GGross said:

Do you have a picture of the installation?

Yes.
Flow from boiler supply passes through air separator, then up and over to circulator, then over to tee which sends water to system and to the Caleffi valve.
From the Caleffi valve boiler supply mixes with system return water and then is sent over to the boiler return.
The additional piping and valves can be ignored, those are for use when the Caleffi is shut off and water flow takes a different path.

Hope this helps!
Eric

hot_rod

When the valve is in full bypass, all flow going back to the boiler it has the same opening as when it is in full flow to the system. So I'm not sure how it could make noise in one position and not the other.
There probably is less flow resistance through the boiler, so the circ should be close to run out.

The flow through the valve is always the same opening size, as one side closes the other side opens by the same amount.

In the cutaway, one opening is at the bottom the other opposite is at the top. There is never a time when there is a flow restriction, it is always a 10 Cv valve, regardless of the position of the cartridge.

I'll know more when my sample arrives, to do some flow testing.

EricPeterson

@hot_rod - the ball valves are all full port so that should not provide any restriction.
The only thing I can think of is that with lower flow resistance, the water is flowing at a faster rate through the loop, hence the noise. I can make a video later if that would help.

As I said before, the noise diminishes once the valve starts to open to return water from the system, which takes up to two minutes for the 140F element. This time will be reduced with the 115F element.

My main concern is whether or not this noise indicates condition that might case the NRF-22 to fail prematurely. That said, the noise is objectionable to my wife, so I would like to address it.

If there is a reliable solution that would lower the pump speed below a certain temperature, then I'm all ears. If there are any specific suggestions I would like to hear them, the more details the better.

Thanks,
Eric Peterson

hot_rod

I think we talked about that pump being a bit oversized. If you had a 3 speed NRF 25 you could dial in a better flow rate.
How did you choose the NRF 22?

With the short loop through the boiler and only a handful of fittings you are probably running at the end of the curve. Maybe moving 15 gpm or more through what looks to be a 1" steel pipe. So possibly some high velocity. Above 5 fps you start to hear noise in piping.

I forgot what sized boiler? I doubt you need to move more than 10- 12 gpm? So a NRF 25 on speed one looks like a much better matched pump.

Really, Xylem would need to know more about the pipe sizing and application to correctly answer your sizing or noise question. Did you send them a pic of the near boiler piping?

EricPeterson

hot_rod said:

I think we talked about that pump being a bit oversized. If you had a 3 speed NRF 25 you could dial in a better flow rate.
How did you choose the NRF 22?

With the short loop through the boiler and only a handful of fittings you are probably running at the end of the curve. Maybe moving 15 gpm or more through what looks to be a 1" steel pipe. So possibly some high velocity. Above 5 fps you start to hear noise in piping.

I forgot what sized boiler? I doubt you need to move more than 10- 12 gpm? So a NRF 25 on speed one looks like a much better matched pump.

Really, Xylem would need to know more about the pipe sizing and application to correctly answer your sizing or noise question. Did you send them a pic of the near boiler piping?

The boiler is a Burnham ES27, the net output is 153 MBH, the near-boiler piping is 1-1/4" black pipe,
I didn't choose the NRF-22, it came with the boiler and I installed in one Christmas Day when the coupler on the Series-100 broke. It's worked fine as a replacement.
I'm open to trying a NRF-25, but the difference looks small to me on paper, and I would not be surprised to hear it make the same noise. I haven't sat down and tried to do the calculations for flow rate.

Eric

hot_rod

EricPeterson said:

hot_rod said:

I think we talked about that pump being a bit oversized. If you had a 3 speed NRF 25 you could dial in a better flow rate.
How did you choose the NRF 22?

With the short loop through the boiler and only a handful of fittings you are probably running at the end of the curve. Maybe moving 15 gpm or more through what looks to be a 1" steel pipe. So possibly some high velocity. Above 5 fps you start to hear noise in piping.

I forgot what sized boiler? I doubt you need to move more than 10- 12 gpm? So a NRF 25 on speed one looks like a much better matched pump.

Really, Xylem would need to know more about the pipe sizing and application to correctly answer your sizing or noise question. Did you send them a pic of the near boiler piping?

The boiler is a Burnham ES27, the net output is 153 MBH, the near-boiler piping is 1-1/4" black pipe,
I didn't choose the NRF-22, it came with the boiler and I installed in one Christmas Day when the coupler on the Series-100 broke. It's worked fine as a replacement.
I'm open to trying a NRF-25, but the difference looks small to me on paper, and I would not be surprised to hear it make the same noise. I haven't sat down and tried to do the calculations for flow rate.

Eric

That boiler, like most cast boilers have very low pressure drop, and can run with no flow.
Note the min. flow and pressure drop in the manual table attached.

So the the size of the circulator has to do only with the distribution piping. How many gpm and what head is the piping? That is the best way to size the circ.

What about heat loss, does the home require the full 155 K the boiler can supply? If so at the very most you need 15 gpm. With the large piping we can see I'd assume the distribution has very low pressure drop? So a pump capable of at the most 15 gpm, call it 5- 6' head, unless you have done a piping loss calculation?

Ideally you run a circ at the mid point of it's curve.
So 15 gpm, 5' head is speed 2 on a NRF 25, right at the sweet spot.

The series 100 is a unique circ, low head, high flow. Shut off head, the very left of the curve is only 6'

Give me a few days to get mine installed to see if there is something unusual at that bypass condition.

hot_rod

Mine arrived today. I'll start piping it up.
I have an Alpha 26-99 so I can run a wide range of flows.

Basically that clear composite cartridge moves up and down to vary the ports. When in bypass it is fully retracted. flow comes down around the spring and right to the bottom left port. Not much od any flow restriction to cause noise.

OldawgBryant

@EricPeterson may sound a little stoopid, but have you tried fully insulating all pipes and valves in the bypass? (trying to determine if it is a piping noise vs pump noise). Also, the long flat head screwdriver makes a great stethoscope and you may use it at different points on the piping and components to hear the location of the noise more exactly (or auto parts stores sell stethoscopes)

OldawgBryant

@EricPeterson Also, you have valves on each side of the mixing valve, so you should be able to simulate the condition of the noise again, by throttling the flow, when the mixing valve opens fully, right?

EricPeterson

@hot_rod and @OldawgBryant - I've given this some thought. Please bear with me with my analysis...

First note that there are two scenarios under discussion:

1. boiler running with water in a tight loop with the Caleffi valve closed.
   
2. boiler running with water passing through the system with the Caleffi valve partially or fully open.
   

hot_rod said:

With the short loop through the boiler and only a handful of fittings you are probably running at the end of the curve. Maybe moving 15 gpm or more through what looks to be a 1" steel pipe. So possibly some high velocity. Above 5 fps you start to hear noise in piping.

@hot_rod I think you are on to something here. According to this, 20 gpm through 1-1/4" pipe translates to 5.23 fps.
According to the pump curves, with very low head the NRF-22 is pushing around 22.5 gpm.
Hence the noise when the Caleffi valve is closed?
So once the valve opens the and water is flowing through the system, the head goes up, the gpm goes down, and the noise dissipates.
Note that before the valve was installed, the NRF-22 was quiet and did an adequate job of circulating water through the system.
According to that gpm to fps convertor, all I need to do to get the fps below 5 would be to have a gpm of 19 or less. The NRF-25 on speed 1 or even 2 would seem to meet that.
The only concern, would the gpm be adequate to ensure flow through the system?
My original calculations, based on this page from @Steamhead indicated

• EDR 908
  
• GPM 24-25
  
• Head 3-3.5
  

There was also a chart there which, along with a heat-loss calculation, led me to the ES-27. I not 100% confident in these numbers.
But if that GPM is correct, the NRF-25 might be a bit undersized.
On the other hand, since I have three zones, that analysis would apply only when all three zones were calling for heat, which is not often.

When you net it all out, what I think it comes down to is a circulator that meets two conditions:

1. runs quiet when the valve is closed.
   
2. provides adequate flow to the system when the valve opens.
   

So for example I think this rules out the Series-100, which according to the pump curve would seem to me be even noisier when the Caleffi valve is closed.
@hot_rod - you also mentioned Grundfos, is there a particular model you would recommend?

Thanks for reading,
Eric

hot_rod

Your boiler only has a 155K output, so with a 24- 25 gpm flow rate you would end up with a tight delta I would think. If you design around a 20°∆ you only need 15 gpm?

While you have enough radiation to support more BTU/ hr transfer, you just don't have the horsepower.

You are correct if the system head is that low, below 4', you can see what flow that circ is producing.
Which is why the series 100 was such a good match for a very low head, high gpm, relativity speaking, demand.

When you drop the bottom of the 280 out to try the different temperature sensors, see idf any debris or teflon tape is jammed in that clear composite spool. that too could cause some restriction noise.

The noise you are hearing, assuming you are completely air free, suggests excessive flow. The numbers seem to be point that way also?

I can't quite see if you have a valve at the discharge of the circ? If so you could try throttling some flow away. But that is a parasitic solution. A closer sized circ, or a "cruise" control on it would be another way to fine tube flow.

You are probably sick of spending $$ on the system. I think a circ like the Alpha 15-58 has settings specific to your system. Min., max. or constant flow settings. And it tells you what it is doing on the display. You could probably circulate your system running 20W or less! With a properly adjusted ECM circulator.

EricPeterson

hot_rod said:

Your boiler only has a 155K output, so with a 24- 25 gpm flow rate you would end up with a tight delta I would think. If you design around a 20°∆ you only need 15 gpm?
While you have enough radiation to support more BTU/ hr transfer, you just don't have the horsepower.
You are correct if the system head is that low, below 4', you can see what flow that circ is producing.
Which is why the series 100 was such a good match for a very low head, high gpm, relativity speaking, demand.

Yes I agree. The EDR for the main house was originally heated with a bigger boiler, plus we added on that load with an addition, but all mitigated with better insulation. But bottom line is that we don't need all that radiation at full power, which as you pointed out our smaller boiler can't provide anyway. So I agree also with the lower gpm.

hot_rod said:

When you drop the bottom of the 280 out to try the different temperature sensors, see idf any debris or teflon tape is jammed in that clear composite spool. that too could cause some restriction noise.
The noise you are hearing, assuming you are completely air free, suggests excessive flow. The numbers seem to be point that way also?

I will check for any debris. The system should be air free at this point, as little new water was added during the Caleffi install due to the isolation valves.

hot_rod said:

I can't quite see if you have a valve at the discharge of the circ? If so you could try throttling some flow away. But that is a parasitic solution. A closer sized circ, or a "cruise" control on it would be another way to fine tube flow.

There is no valve for the circulator discharge, and I agree that would not be an ideal way to reduce flow.

hot_rod said:

You are probably sick of spending $$ on the system. I think a circ like the Alpha 15-58 has settings specific to your system. Min., max. or constant flow settings. And it tells you what it is doing on the display. You could probably circulate your system running 20W or less! With a properly adjusted ECM circulator.

I don't spending a bit more money for a good solution, the cost so far has been quite moderate, just the Caleffi valve along with a few fittings and globe valves.

Hopefully all this chit-chat is useful to those other two that are installing the Caleffi valve - @TommyA and @OldawgBryant, and anyone else reading this exchange.
Thanks a bunch @hot_rod for all your help!

Eric

hot_rod

Nice thing about being over radiated you can go lower, more efficient SWT. I've run cast radiators down 120 or a bit lower and they still radiate a nice comfortable heat.

So ever there were a HP or mod con in your, or the next owners, future, you might cover a design day load at 150, 140° or lower?

I'm short 1 piece of 1" tube to fire up my 280. I have good instrumentation including a BTU meter to "watch" how it performs. Plenty of pump to overflow it also. Stay tuned. I'm letting my shop slab cool down also so I have a large, hungry, consistent load.

OldawgBryant

@EricPeterson I have a ways to go yet, I'm almost done with the black pipe section of the return, the threaded supply is finished, then I go to 1-1/4" copper sweat to finish both the tie ins, I appreciate all the discussion on this and will be waiting for @hot_rod results of the tests. I have installed an additional mixing valve in case mine squeals also... all of my interconnected close nipples made the unions a real tough install (to get everything lined-up just right) but now it gets easier.

TommyA

In my mind, the 280 valve (sort of, kind of) mimics many zone valves opening in succession (or even better, TRV's). This is exactly what the variable speed constant pressure (or proportional pressure) circs were designed for.

As @hot_rod mentioned, you (and I) will be heating our homes with 120-140 degree water in those rads. Given the volume, I'm pretty sure that 85% (95? 100%?) of the time, that Caleffi valve will *never* fully open before the heating cycle ends. For me at least, having a boiler very closely matched to the load.

EricPeterson

hot_rod said:

Nice thing about being over radiated you can go lower, more efficient SWT. I've run cast radiators down 120 or a bit lower and they still radiate a nice comfortable heat.

So ever there were a HP or mod con in your, or the next owners, future, you might cover a design day load at 150, 140° or lower?.

@hot_rod - yes it's something I would consider in the future. I still have one un-insulated BR and four/five under-insulated rooms that need higher temps. If those are dealt with then a modcon/HP would make more sense. But those considerations at install time coupled with the desire for a simpler, time-tested design led me to the ES27.

TommyA said:

In my mind, the 280 valve (sort of, kind of) mimics many zone valves opening in succession (or even better, TRV's). This is exactly what the variable speed constant pressure (or proportional pressure) circs were designed for.

As @hot_rod mentioned, you (and I) will be heating our homes with 120-140 degree water in those rads. Given the volume, I'm pretty sure that 85% (95? 100%?) of the time, that Caleffi valve will *never* fully open before the heating cycle ends. For me at least, having a boiler very closely matched to the load.

@TommyA - as indicated above, I'm not quite there yet. Also since we use setback thermostats, in the morning the system will definitely heat up sufficiently to fully open the valve since it takes a while to warm up the house. I've not had the issues you mentioned even when the valve is partially open. Also I plan to replace the 140F element with a 115F which will fully open at 133F. The 115F should work better in that regard, and since my ES27 boiler accepts a 110F return water temperature, the boiler should be protected.

@hot_rod - I see that Grundfos has a 15-58FR and 15-58F model, but I don't know the difference.
So if I were to switch circs it would be either one of these or the B&G NRF-25.

Eric

Robert_25

Eric, the difference in the Grundfos 15-58FR and 15-58F is the orientation of the flange bolts. Based on the picture you provided you would need the "FR" model.

The B&G NRF-25 will have the flange bolts 90 degrees off from your NRF-22, so unless you have rotating flanges it is not a direct swap (at least not without getting out your pipe wrenches).

Depending on how much flow you are looking for, the B&G NRF-9F/LW might work.

hot_rod

I’d rather you get one of the electronic ECM type circs. Grundfos 15-58E or Alpha 15-58. Taco and other brands have comparable versions

They have more adjustability, digital display to see flow rate, head, etc. Same with any brand they offer either flange ordination. Some Tacos have a dual flange, so either mounting position is possible.

In some areas there are rebates. Available to upgrade to high efficiency stuff.

Www.dsireusa.org for incentives available in your area.

EricPeterson

@hot_rod and @Robert_25 - I am definitely interested in these Grundfos models (ALPHA 15-58 or UPSe 15-58).
They look to be quite similar with the Alpha having a more sophisticated user interface.
I am not averse to changing the flanges.

These products have different control modes.
My question now is which control mode would work best with the Caleffi valve and my system, so operating with:

• the valve closed (startup)
  
• the valve partially open and delivering heat to up to three zones
  
• the valve fully open and delivering heat to up to three zones
  

The zones are:

• Z1: 33' of 9" baseboard (112 EDR)
  
• Z2: 14' of 9" baseboard and three free-standing radiators (300 EDR)
  
• Z3: 44' of 7" baseboard, 27' with TRV, six free-standing radiators, three with TRV (500 EDR)
  

Which control mode would work best?

1. Constant curve (zone pump)
   
2. Constant pressure (zone valve)
   
3. Proportional pressure (thermostatic radiator valve (TRV))
   

I am thinking either #2 or #3.
I see that the the ALPHA has an additional control mode:

• Proportional pressure (thermostatic radiator valve (TRV)), AUTOADAPT
  

Maybe this would top them all?
Also, each control mode also has three settings (I, II, and III) - not sure what those are.
The question is which is the optimal combination of control mode and setting?

Thanks,
Eric

hot_rod

Well you have an unusual mix of heat unit.
With a zone valve and TRVs in zone 3
, I’m not sure what mode covers a dual control like that. You may just need to try the different settings.

A goal would be adequate heat with one or multiple zones open. All while flowing the lowest gpm needed to cover loads.

You could go to the trouble of developing a system curve for each zone,lay it over the pump curve, see where it falls on the 3 different zones.

Those pumps also have multiple fixed speed settings, but it would be nice to find a modulating setting that best suits the system

TommyA

I'm not pushing any brand over another, but Taco has tons of great videos covering all the deferent circulator modes and speeds (almost as good as the Caleffi series ). The concepts are the same across the brands.



The newer ones are at the bottom. I've found 68, 72, 73 and 90 interesting.

TommyA

Sorry, that didn't paste as I thought. Just click the 1/93 link in the upper right corner.

Robert_25

@EricPeterson This one would bolt right in your current flanges: https://www.supplyhouse.com/Grundfos-92603108-ALPHA-15-58FR-Circulator-Pump-w-Rotated-Flange-Mount-115V-1-20-HP-GF-15-26

The I, II, and III settings are either the three different fixed speeds, or the three different levels of constant pressure - depending on which mode you are using.

I agree with hot rod that you will need to experiment to see what settings work best for you. "Auto Adapt" is intended for TRV's, so that seems like a good place to start.

EricPeterson

@hot_rod - the odd mix with Z3 was done to balance the heat load. The TRVs are in 2nd floor BRs that were overheating and are sometimes closed off. The thermostat is in the third floor workroom with an unregulated radiator. The other two unregulated radiators are in bathrooms, no one ever complains about too warm bathrooms right? There’s also unregulated baseboard on the third floor BR, I shut that off unless it’s really cold. I could add a TRV or put in a 4th zone valve since it’s a dedicated pipe run. But time & money….
Probably way more than you wanted to know!
@TommyA and @Robert_25 - thanks for the links and suggestions.

Eric

hot_rod

So basically the zone with the TRVs will always have a varying flow. That is how the TRV manage the temperature. Typically you would not vary flow in the fin tubes do you have a bit of a quandary as to which setting.

The auto adapt is now on all the various settings, so in one of them, maybe the TRV setting, it may learn the sweet spot.

EricPeterson

Here's a bit of follow up on my Caleffi valve install.

The valve has been working flawlessly and I am very happy with its performance.

As I indicated, the only issue has been the noise when the valve is closed and the water is being circulated through a small loop with very low head. The problem is not with the valve, but with the excessive speed at which water is flowing through that tight loop when the valve is closed. As I indicated before, this flow was in excess of 5 FPS (by my calculations) which apparently is going to generate some noise.

The solution, as @hot_rod has pointed out, is to use a different circulator.
I was all set to go with the NRF-25, but have had second thoughts, and plan to go with the Grundfos, probably the Alpha. Despite my hybrid combination of TRVs and zone valves, I think that I will be able to find one the modes with the Alpha that works with my system. But deep into the heating system, with some subzero weather approaching, working on the boiler is the last thing I want to do right now.

So I cheated for now - by cracking open a valve that bypasses (just a little) the Caleffi valve, which allows some return water to go directly back to the boiler rather than through the valve. I opened this valve just enough so that the whining sound stopped. It seemed like a reasonable thing to do for the time being.
My better half has advised me that I am not to work on the boiler right now. 2

@hot_rod - did you ever run any tests with the valve as far as checking for noise is considered?

@TommyA and @OldawgBryant - did you get your systems up and running with the Caleffi valve?

Happy New Year,
Eric

hot_rod

Yes I built a demo set up in the shop. I piped a small 1” copper and Pex loop. Unfortunately I dropped the Grundfos Alpha 26- 99 and it won’t adjust off the one speed. I did install it with a Quicksetter so I can observe flow rate. I can hear velocity noise in the piping. I had hoped to try one of the delta P settings to see if the pump would respond to the valve as it modulates the openings on the various ports.
If there is more flow resistance in the bypass position, maybe the pump will respond to that?

On Monday I will hit my local Grundfos rep up for a new pump with working delta functions. I have some Alpha 15-55 but I need a higher flow to try and duplicate the noise you are hearing.

I’ll get back on the project this weekend

hot_rod

Here is what I can do now. I have 10 gpm going through the bypass now. This is a 1" valve with a 10Cv.
The pump is rattling a bit, probably from the drop, stuck on speed 3, no 0-10VDC signal applied so it runs full speed.
No hissing or velocity noise.

With all the fittings and flowsetters I'll call it 40' of pex. That puts me at 5.5 fps velocity.

I'll start heating it up to see if the noise occurs as it is modulating the ports instead of wide open bypass.

hot_rod

At 120F the valve starts modulates and you can see bypass flow start dropping, system flow starts moving. At 140F bypass is 100% closed, all system flow.
I did some adjusting and got almost 13 gpm, so over 6 fps

I have a short loop, so this happens quickly. Pump noise, boiler ramping up noise, but nothing I can hear in the valve?

I'll still try a delta P circ to see how flow responds as the valve modulates. But as one port closes the opposite opens that amount, so there is no increase in delta P I suspect.

EricPeterson

Thanks for all that detailed information @hot_rod.

To be clear, I have not associated any noise with the valve itself.
Instead the noise seems to be a local vibration in the boiler loop that is transmitted throughout the system piping.
In my case the noise is present when the valve is completely closed, then lessens as the valve starts to modulate, and once it's completely open at 141F the noise is gone.
When I said that I cracked open a bypass, what I meant was that I slightly opened a valve that allowed some of the return water to flow directly back to the boiler, instead of going through the Caleffi valve.
I believe that opening this bypass slightly increases the head enough so that the FPS falls below 5 and thus the noise disapppears.

Eric

hot_rod

Yes metallic pipe or tube will transmit and carry noise. Sometimes throughout the building. So it may be hard to determine which component is causing the noise. A hissing sound is most often related to excessive flow velocity.
Vibration is more like a pump impeller out of balance, something stuck in the vanes of the impeller. A single solder ball or rust particle is enough to get them shaking.

If your bypass valve is working and still allowing the valve to protect the boiler, may as well go with it.w

EricPeterson

Well that answers one question regarding the Grundfoss Alpha pump. At startup when water is recirculating through only the boiler and the Caleffi valve at 11 GPM - there is no noise, because the flow rate is sufficiently low.

I set it for zone valves and AutoAdapt - will see how that works for now. The other option available through the user interface on the Alpha was zone pumps.

There are additional modes available through the GO app, but I have not tried those yet. That is also where you can get additional information regarding the operation of the circulator.

—Eric