High pitched whining noise in 1 zone of radiators when zone pump is running

SamDaley

I have a Navien combination boiler that serves 3 individually pumped zones of baseboard radiators. When the pump in zone A is running, there is a high pitched, steady whine that can be heard coming from the radiators in that zone. I swapped pumps from zones A and B and the whine still happens in zone A. Zones B and C have no issues. There isn't any pressure loss in the system that I've seen.
Any ideas what might be causing this?
Thanks!

hot_rod

Do the radiators have TRVs? Thermostatic radiator valves?

EdTheHeaterMan

Are there any grandchildren in the home? Are they near the radiators that are whining?

Since there is no reply from @SamDaley for some time, I just had to ask.

I have experienced moaning and other noises in systems with thermostatic valves that were located on old radiators. I am wondering if there are orifice plates in the valve unions on the problem zone. Is this an old gravity system that may have been converted to a zoned system? The easiest way to convert a system like that to zones is to leave the top floor connected to the old pipes and run new supply and return to the lower floor zones.

That means that if there were orifice plates inside the radiator valve unions, they might still be there.

Without more info on the system, we are only guessing

HomerJSmith

Are we talking fin tube baseboard? Wet rotor pumps are very quiet. Are you getting heat in that zone? Perhaps it is the way that zone is piped. Pic of your sys would be helpful.

You can take a mechanic's stethoscope that you can buy at any auto parts store and use it to try and locate where the sound is coming from.

SamDaley

Good question, no the system does not have any radiator valves and these are fin tube baseboards. It was originally a boiler with 1 large pump and a valve per zone. It was converted to the 3 pump (no valves) configuration a few year ago when the combination boiler was installed and has been working fine since then. We are unable to locate the exact location where the noise is coming from since we can hear it from all of the radiators in that zone, but the whining is noticeably louder at a couple of the radiators vs the others in that zone or at the pump in the basement. Other than the noise, the system seems to be working properly.

HomerJSmith

You should be pumping into the zones not away from them. I like pumping into the highest pressure loss. When using pumps for zoning I like to use balancing valves on the zone returns to regulate the flow. I want the flow to be about 4GPM per zone to prevent velocity noise and pipe wear. What are the kind of pumps you are using? Is the noisy zone the shortest zone?

You could have used one ECM pump with your old zone valves. Pics of everything would be more helpful.

SamDaley

I had thought the same thing about the pumping direction but contractor that did that part of the install had some reasoning why he did it that way. I have debated changing it around but haven't since the system seems to be working properly otherwise. I am using Taco 077-ZF5-9 1/25hp pumps.

hot_rod

3/4" copper tube? What is the approximate length of the loops, or the noisy one.
If the loops are under 50' total, you may need to throttle back the pump.

Seems the single pump and zone valves was a better option. Why the change?

Could the noise be from the boiler running on it's low fire end of the modulation?

HomerJSmith

hot-rod, "Could the noise be from the boiler running on it's low fire end of the modulation?"

I find that statement very insightful.

SamDaley

The noisy loop is probably over 150' and is mostly 1/2" and 3/4" copper. The noise stops as soon as I shut off that zone's circulating pump so I has something to do with the line itself (since swapping pumps didn't stop the noise). The contractor that did the original install decided to do the pumps vs valves and it's listed as an option in the Navien manual so I didn't question it, other than the direction.

Dave H_2

I have found in some systems that have a noise that is hard to diagnose, add an extra pipe support, clamp, whatever in another part of the pipe.
It sounds weird, but there are some "harmonics" based upon the distance between clamps, length of pipe, size.....Everytime I hear of a project like this, I think of the Tacoma Narrows bridge.


Sometimes it works and is the easiest and cheapest way before repiping. Worth a shot.

pedmec

What circulator are you using on the system. With only one zone operating all your velocity is in the single zone which is not only one zone but is the shortest zone. You also said that you have 1/2" piping on that loop? that would absolutely not help and could be the cause as you are reducing the area and creating a high velocity area

SamDaley

Thanks for everyone's help so far. I know that this isn't a very straightforward issue. Unfortunately I'm guessing that Dave is probably right and there was some tiny change in the system due to gravity, line build-up, etc. and now it has hit some random resonant frequency that's causing the noise. Which makes me think that this has to occasionally happen to other people and the only solution for them can't be to cut up a ton of drywall and add start adding supports or replace pipes until the noise goes away. I wonder if there is another plan of attack that someone would do if that's the case (like a variable speed pump)?

hot_rod

Are there isolation valves on the circulators? Maybe try flowing down that zone by partially closing the valve on the discharge
Fin tube design is typically 1 gpm, with the 4 gpm table also shown. I suspect you may be pumping a bit too much in the zone.
It’s an easy inexpensive test, assuming you have valves.

Dave H_2

@SamDaley A variable speed circ can affect it, the rpm of the motor will be completely different. And if you chose a circ that you can select the speed or setting on the dial, even better. Not saying it definitely will fix it, but it can help.

And I wasn't necessarily saying to open sheetrock and start poking around, sometimes adding a support 6" or 12" away from an existing support can help.

EdTheHeaterMan

Are there any valves that can isolate the circulator pumps or at least the zones? If you are lucky enough to have a valve on that circuit at the discharge side of the pump on the whining zone... can you throttle the valve closed while the pump is operating? See if the noise changes or is eliminated.

SamDaley said:

The noisy loop is probably over 150' and is mostly 1/2" and 3/4" copper. The noise stops as soon as I shut off that zone's circulating pump so I has something to do with the line itself (since swapping pumps didn't stop the noise). The contractor that did the original install decided to do the pumps vs valves and it's listed as an option in the Navien manual so I didn't question it, other than the direction.

I would like to see a loop that has ¾” then ½” pipe included. If the 2 different sizes are in series, then effectively you have a loop with only 1.5 to 1.7 GPM flow before you develop velocity noise. Read page 4 of this booklet to understand this. https://www.xylem.com/siteassets/brand/bell-amp-gossett/resources/technical-brochure/fh-z100b-bg-zoning-made-easy-2.pdf

hot_rod

Here are some flow/ velocity charts for 3/4 copper .

CDA suggests 4 fps maximum velocity for water 140F or less.
2 fps above 140F. If that matters

Ideally you would calculate the developed length of the loop the circ is attached to, plot a system curve, overlay the pump curve to determine what exactly is the OP operating point.
Balance valves of variable speed circs are best to getting the flow rate correct.

I'd try some flow adjustments before ripping out sheetrock

HomerJSmith

You have no idea what is happening inside that circuit as to flow and pressure which of course are related. That's why I use balancing valves when I zone with pumps. These wet rotor pumps are all high head pumps compared to a B&G 100 that you had before. You just can't slap on a pump, brush your hands and just walk away. Ya, sys curve and pump curve charts are great if your doing a new job, but in retrofit jobs it's difficult to know what is going on behind drywall. I have seen ways to estimate it, but it is still an estimate. You still don't know. Yes, you can install a pressure gauge on the intake and discharge of a pump to get an idea of the sys pressure loss, but who does that?
It's just better to install a balancing valve on the return of the circuit and adjust the flow to the size of the pipe you are pumping thru. It saves the pipe from excessive flow wear, yes pipes can wear out, noise issues and possible pump cavitation.

SamDaley, just install a balancing valve on the return and get an idea what is happening with the flow. It wasn't a problem with the B&G100, but it is now. You will at least eliminated one possible cause of your problem.
https://www.supplyhouse.com/Caleffi-132552A-3-4-NPT-QuickSetter-Balancing-Valve-w-Flow-Meter
https://www.supplyhouse.com/Caleffi-127359AF3G5-3-4-Sweat-FlowCal-Automatic-Flow-Balancing-Valve-3-5-GPM

SamDaley

I do have isolation valves so I turned one 45 degrees (90 is closed) and the noise went away. With that said, I don't know for sure if that just reduced the flow some and shows a path to a solution to the issue or if 45 degrees basically closes off the flow completely or mostly.

Dave, sorry I didn't mean to imply that you were suggesting that I would need to cut open the wall just that it would be a tough to justify doing that even if the issue was 100% due to needing support added to an unknown location that ended up being buried in a wall.

HomerJSmith

You just answered your own question. Yes, closing a valve lower the flow rate. Just put in the balancing valve, bite the bullet, and do it.

You obviously have a ball valve partially closed off. You still don't know what the flow is thru that circuit.

mattmia2

Measuring supply and return temp with the valve fully open and at 45 degrees will tell you if you have enough flow at 45 degrees.