solve ticking noise with constant flow system?

RonV_2

Thanks. This will make great reading.

RonV

RonV

can constant flow be retrofitted to fix ticking noises?

My hydronic system suffers from annoying ticking noises when the pex tubing expands and contracts against wood framing. This occurs mainly where tubing passes into walls and through wall studs and floor plates for attachment to manifolds, not in the floors where there is no sound problems at all, despite this being a staple-up system. My control system is a conventional on/off thermostatically controlled setup so when calling for heat, the zone valve opens up and the tubing goes from room temp to hot very quickly as the water circulates. Hence the expansion/contraction of tubing and its movement against framing.

I gather that there are two approaches to solving this problem. One is to open the walls at each noise source and provide some sort of isolation between tubing and framing. While simple in concept, this would call for many many wall openings in finished areas with considerable baseboard and crown millwork. This seems like an awful lot of work and refinishing (likely involves several dozen holes, drywaller, painters and finish carpenter, say $6K?). The alternate that has been suggested to me is to convert to a constant flow system which will dramatically spread out and minimize the frequency of these ticking sounds. My original contractor who suggested this has moved out of town and is no longer available. My new contractor, who I believe is competent, roughly estimated about $8K to do the retrofit, not including any wall repairs which may be needed to replace local loop valves and stuff. Now, I know you guys can't guess if this is good value or not without knowing more details of the system so I won't ask. But do any of you have experience with doing this kind of conversion and from that experience, can it be done economically?

I'm happy to provide details of my system if it would be helpful, but here are the basics:
- primary loop which includes the two boilers is coupled to secondary loop via modulating bypass valve. Modulation to regulate output temperature is via capillary tube tied to secondary loop on its exit from the secondary loop main circulator. This in turn feeds the main distribution manifold
- distribution uses simple on/off type zone valves on 17 zones, all driven by mercury thermostats
- the control system is very simple, using relays, no microprocessor panel involved. Thermostats call for heat and turn on the circulators, and also enables the boilers.
- other loads on the primary loop include hot water heater and pool heat exchanger


any advice appreciated

Constantin

Allow me a simple question....

... does the current system incorporate outdoor reset?

... do you have differential bypass valves if multiple zone valves are attached to one circulator?

The only reason I ask is that a home I am familiar with has the same ticking problem... caused by fast, hot slugs of water hitting the PEX tubing. If the system in this house were on a proper OR curve, most the valves would be open all the time, creating a virtually constantly circulating system. Plus, it doesn't help that this house features one Taco 011 that is feeding 8 zone valves without a differential bypass...

So, if you haven't yet considered OR, now may be the time. Your current control sheme sounds like a form of indoor reset, which can work also. Yet, I wonder how it can set the loop temps to ensure that the reference zone is running constantly? Perhaps I'm missing an important clue here, which isn't too difficult considering that I'm quite unfamiliar with your type of system, and a mere homeowner to boot.

RonV

A mere homeowner? You sound savvy to me. To answer your questions though, my system has no outdoor reset function at all. I'm pretty sure the only "reset" function in my system is the manual modulator setting on the bypass valve down in the boiler room (currently set to minimum). I have one circulator on my secondary loop, the loop that feeds one giant manifold with all of the zone valves, and the zone valves are not differential bypass type.

I guess this means I need to learn more about the benefits of outdoor reset control. I presume that the outdoor reset controls the secondary loop temperature. That's a good idea because it means that my fluid temperature in all the zones would be a lot lower most of the season, hence keeping the zones open longer. And I'll bet that the hardware changes would be simpler. A buddy of mine has a three loop system that uses loop injection to modulate his distribution temperature. I'm pretty sure he has an OR, so maybe I should have a closer look at his setup. I didn't know that this can make my system sort of a virtual constant flow setup. Very interesting indeed.

Ron

Alex Giacomuzzi

More Info. Please

It sounds like you have a relatively large, yet basic controlled system with a large number of zoned areas... 17 per you post.

The questions that I would be asking myself are how many of those 17 zones actually require separate control, or are there several that could be grouped and possibly help simplify the control for example, which would also help lower the temps experienced by some of the problematic tubing runs.

Another question is..... do you know which of the loops are causing you problems. Hopefully all of them are not. Can you isolate the ones with problems?
You suggest it is happening in the wall cavities. Do you have any pictures of the install especially in those problem areas??? This could help you locate the problems perhaps..

If you could isolate the problem cases, might it be possible to spray some foam into those wall cavities to help stabilize and isolate the tubes from running. Much cheaper than making big holes in the walls. Check with you PEX tube mfg. to verify the foam will not be a problem with the tubing. More manageable holes to patch.

Does your mixing valve incorporate outdoor reset into its control?? If not, this would help..

Regarding constant circulation.....this would definitely help, but with 17 separate zones which are going to be at or slightly above ambient when off (due to the thermostat) will still get hit by the warmer circulating water when a call comes in..

Here is the most simple suggestion I can think of knowing as little of your system as I do... Consider installing a buffer tank appropriately sized in series with the loop that feeds all the zone valves. Allow your 3 way valve to adjust the temp of that buffer tank with OR active. In other words, treat the tank like your domestic HWH with out priority and including OR. Now, if you allow the circ. pump to run continuously on the zone valve loop, it will deliver a lower water temperature and hopefully a slowly changing water temp to all loops which are held constantly open. The zone valves would not call to activate the boiler anymore as they probably presently do. The OR would reset the buffer tank temp and directly call for boiler heat when it dropped to its call temp.

Now comes the desire to simplify.......... If you can keep an many of those zones calling in parallel (zone valves open continuously), provided your emitter design matches your heat losses appropriately for each of the spaces. I am really stretching here given that I have zero info. on your house and living style. If it is impossible to do that in all areas, you may be able to utilize this approach in some or several, yet not required in all.

If you can not keep any of the zones open, you will need a by pass valve installed for obvious purposes in the loop between suppy and return manifolds...

With a buffer tank that is reset, and constant circulation in the trouble some loops, you may be able to minimize your ticking problem. It would not be out of line to also consider indoor reset also.

I hope some of these thoughts help.......

Good Luck........ Alex

Constantin

Interesting...

So you have 17 zone valves on one manifold? Sounds like a huge circulator! As for the pressure bypass valve, there would be only one, at the end of the manifold, joining the supply and the return manifold together.

Whenever the pressure that the pump produces exceeds the setpoint, the valve opens and allows the excess water to go directly to the return manifold. In this manner, the flow speeds within the loops are controlled and the pump isn't woking like crazy to push lots of water through a small pipe. Proper flow speeds in turn reduce pipe noise.

As for the OR retrofit option, I'd encourage you to investigate it further. Retrofitting an OR controller shouldn't cost nearly as much as the other options you have listed since much of the piping, etc. is already in place. Retrofitting OR should also disrupt your house a lot less than opening up walls. There are many different ways to regulate that secondary loop temperature, using injection-mixing, 3- or 4-way valves, etc.

Your statement that 17 zone valves are piped off that manifold still has me scratching my head though. If OR results in most zone valves staying open much of the time, this circulator might make sense, though retrofitting a bypass would be a very good idea. You could also consider using smaller circulators branching off the secondary loop with just a few ZVs on each sub-branch instead. Perhaps a piping schematic would help?

RonV

more to come

Yes, there is one TACO 0012-F4 circulator pushing water towards the manifold. The pipes connecting them are partially hidden in a finished ceiling, but the manifold itself is clearly visible in crawl space and I don't see anything connecting the hotside input manifold to the cold side return manifold. So I guess that means there is no differential pressure relief valve.

To answer Alex's questions, yes I also believe that the number of zones is escessive. For example, four zones include the dining, living, front hall, family room+kitchen and yet these rooms are connected by large 6 foot wide open archway rendering all of these zones into a nearly open plan. Other zones can be also be grouped, so there is much potential for simplification. In addition, I know precisely which zones are problematic after three years of poking around. All the problematic ones are on the second and third floors and all the ticking noises come from the walls and plates of the second floor. This includes a total of only four bedroom zones and one "bonus room" in the attic that is currently unused. If we fixed those four bedroom zones, I would consider my problems solved. Unfortunately, we do indeed run all those four zones at notably different temperatures, so grouping them into one would not be a good idea.

I considered simpler ways of isolating the tubing from the framing. One idea was to make myself a simple viewing device, much like the doctors use when doing laporoscopic surgery (I'm in electronics so this doesn't seem all that tough to me), and use it to spray silicone lubrication onto the tubing at those noisy spots. Silicone lubricants are usually very compatible with most kinds of plastics (of course I would check with the PEX mfr and do a long term test myself too). Trouble is, I don't believe this would last more than a year or two, for some reason.

No, my mixing valve does not incorporate outdoor reset control.

Your buffer tank idea is interesting and I will study that some more. In the meantime, I will take as my homework to prepare a system schematic and post it sometime tomorrow. Thanks so far.

Ron

Weezbo

slower acting zone valves or modulating zone valves or....

sub injection off primary to these zones, would all be an answer to off on high heat supply and associated noises...

Unknown

Personally, I'd add outdoor-reset variable speed injection first.

Secondly, I would consolidate zones as much as possible, then upgrade the T-stats to models with PWM.

Between PWM and outdoor reset, your system will operate very smoothly indeed.

RonV

heres more info

The attachment is my system schematic. Its my first one, so apologies for mistakes.

Ron

Alex Giacomuzzi

Thanks for the schematic........

Ron ---you draw a nice schematic --- Thanks.
A few more questions ----

What are the sizes of your two boilers? What is the manufacturer and type of boiler? What type of control, rotation of load, or sharing of load do they do..... how are they integrated into the heating loop? Describe as best you can.

What is the heating load on your Main Residence excluding the pool, DHW, and building 2. What is the load on building 2 alone. Describe the cycling of the boiler... on time versus off time presently --- again as best you can. What temp is the thermostatically modulated bypass valve set to? I presume this is a fixed value and it could be adjusted if required to a lower value. Is there a temperature indicator after the pump somewhere? That would be handy to have to be able to read circulating water temps.

You really have much to work with already and that is a plus. The 17 zones could drive the cycle time on a boiler crazy. This would be dependant on the type, size, & style of boiler that you have and if it has any modulation capability (or not).

The first thing that I would try is to turn down the modulated by pass temp which is controlled by the modulating valve to a lower value. Turn the thermostats up in several of the common areas such that the pumps start, the mixing valve starts working, and probably a boiler starts. Leave the thermostats set high enough in those common areas that they will not be satisfied for awhile. I would expect that the boiler would start cycling on high limit.

Now go into one of the problem areas and set its thermostat up thereby calling for heat. Listen to if there are more, less or equal amounts of noise within the house. Try the other problem areas also and note your findings. Leave the pumping system run continuously if possible. Now make some slight adjustments (and I mean slight) to the modulating temperature control valve. These adjustments could be as low as 5 or 10 degrees at first or possibly closer to 10 -- 15 degrees..as you are collecting your data or observations. Try even up to 20 degrees difference. Increase it slightly and after a brief period move it back to an ambient condition or near to it while keeping the pump running. Cycle the problem areas up and back down with their thermostats.

You are manually testing both continuous circulation and outdoor reset (semi -injected if you wish) manually. Literally --- you are the controller. Determine if the noise level meets your expectations or not. This would be a great place to start and would give you an immediate idea of what to expect by adding more controls or buffers.

Once I know your boiler situation, I will offer a thought on buffering and where to put it.

Regards Alex

RonV

more info (sorry its so long)

Thanks for the interest Alex
To answer your questions..

The two boilers are identical Slant/Fin Victory V-150 boilers rated at 150,000 btu input natural gas (output 127,000 btu). The boilers have been carefully tuned for correct firing rate (that's a whole other story). The boiler control is configured so that one is the main boiler and the other is a backup for high load situations. Each boiler has an internal Aquastat as well as an Aquastat mounted externally on the exit pipe. The internal aquastat is set for emergency overtemp protection only. The external aquastat is the one used to control the boiler normal operating temperatures. On the primary boiler the setpoint temperature is 180 degrees F and the difference is set to 16 degrees. On the secondary boiler, the setpoint is also 180 degress, but the difference is cranked up to 25 degrees. The point of this is that the secondary boiler kicks in when the loop water temperature drops down to 155 degrees or less indicating that the primary boiler needs some help. However, during the fall and spring heating seasons I normally leave the secondary boiler powered off and use only the primary unit. I rotate which is primary and which is secondary once a year to distribute wear and tear.

The rest of the control system is fairly simple. When no thermostat or aquastat is calling for heat, the entire system is powered down. When any stat is calling for heat, the boiler is enabled, the boiler circulator is powered, the primary loop circulator is powered and the circulator specific to the load calling for heat is powered. As far as the house heating is concerned, that sensor attached to the modulating bypass valve does all the output water temperature regulation.

So, speaking of that, the temperature regulation is quite sloppy in my opinion. It is currently set to as low as it can go. I'v monitored it off and on for two years and I know that throughout the season the temperature varies from 100 degF to 145 deg F and averages around about 125 degF. I had a look just now and found it to be only 100 degF which kind of suprises me as that is the lowest I have seen it in two years. Once, right in the middle of a winter freeze, it was up around 155 degF. The setting is done using a knob on the bottom of the Oventrop bypass valve, with a scale of 1 to 7. The temperature sensor tubing feeding this valve has a tag on it that says "40 to 70 degC" so we can guess that this is the range of the valve scale, 104 to 158 deg F. I get these values from a thermometer mounted at the input to the main distribution header.

The schematic is a bit oversimplified in that one of the 15 zones for the main residence is actually routed to a workshop building which is a separate structure and not the same as building 2 which is the garage with loft.

Now, for those hearing load figures, the only info I have is from the original installing contractor and its a pretty fuzzy photocopy. Perhaps it would help if you could prompt me with the units that you are expecting this figure to be in (ie. btu/hr or ?). However, I think we have the main residence at 126,000 btu/hr, the workshop is 18,500 btu/hr and the garage (building 2) total is 30,600 btu/hr.

The boiler currently cycles fairly short in my opinion at about 2 minutes on vs about 10 to 15 minutes off (the on time I'm pretty sure about, the off time is a rough guess and varies with load). There are many times when the boiler on time is terminated prematurely when the zone calling for heat is satisfied. This seems to happen a lot.

As a side note, I set the anticipators on all my thermostats to the longest value because I believe my zone time constants are fairly long. The difference setting of 16 degress on the primary boiler was also an attempt to increase the on time of the boiler as the boiler was short cycling a fair bit back then (a long story).


We tried some of the experimental procedure, that you describe, some time ago. At that time, we found that a lower temperature slightly mitigated the ticking noises by spreading them out in time a bit. The difference was very small. Another thing that we tried was to partially close the ball valves on the return side of the noisiest 2 zones. This also helped to spread out the ticking a bit, but the worst zone was still a circus of ticks sounding like squirrels were doing battle with machine guns somewhere in the three walls surrounding my bedroom. When the zone starts up, this ticking goes on for about 2 to 3 minutes and is rapid fire. Anyway, that's how the control setting ended up at its lowest setting. Its been that way for two years now.

However, the procedure you describe goes much further than we did in that you are trying to mimic a constant flow system, so I plan to give this a good try starting tomorrow.

Hope this extra data helps.

regards
Ron

RonV_2

more background

Hi Alex
Further to my post of a few days ago, I haven't been able to do the test yet that you suggested. Today, however, we have been running the system with a steady load (turned up a zone that was cold) and I notice that the distribution water is sitting up at 150 degrees most of the time. This strikes me as hotter than expected and just goes to show that the output temperature regulation isn't very good since I have the manual knob set at its lowest possible value. I guess that the Oventrop Bypass valve (the one that is thermostatically variable) that is used in my system doesn't close completely to simply recirculate the return water. Which means that it has no way of limiting the maximum water temperature if that is true. Since the primary loop in my system seems to spend most of its time at about 165 deg under full load, while the boilers themselves are around 175 deg average, I get the impression that the output regulation isn't really functional under heavy load, so we get 150 deg or thereabouts. Do you have any experience with this type of valve? I've attached a photo showing the valve, and another trying to show how its hooked up.

thanks in advance
RonV

Constantin

Allow me a dense question...

... perhaps Oventrop valves are different from all the other valves out there, but usually the mixed output goes out the "bottom" of the "T" that most such valves have. The picture with the piping shows a manual bypass past the Oventrop that is currently open. As such, why would any water flow through the valve?

RonV_2

You would have to see more of the system, but perhaps the schematic would help (attached). The way this works is that the straight flow through that Oventrop valve pushes water into the primary boiler loop, and since water isn't really very compressible, what "gives" is that water is forced to flow from the primary boiler loop into the output circuit. It wants to flow this way because the pressure is slightly lower there on account of there is less flow through the "T" connection of the Oventrop valve. So, that pipe that appears to be a bypass completes a loop to allow water to flow from the primary to the distribution circuits.

When the oventrop is pushing all water to its T connection (the one that isn't straight through) then there is no differential pressure on the straight through side and nothing encouraging water to flow through that "bypass" pipe that you see, so no water flows from primary to secondary circuits.

At least thats the way I think its supposed to work. My problem is that I get the impression the Oventrop valve never fully closes the straight through connection and from what I can see of the Oventrop technical literarture it isn't clear if it should, but I think so.

So now I'm trying to figure out how to test my theory.

RonV

RonV_2

one more thing

I should have mentioned that this Oventrop thing isn't a mixing valve, it is a bypass valve. So there is one input and two outputs. Water is diverted from one output to the other depending on thermostatic setting. I think that is key to this thing. The valve doesn't have two inputs.

ROnV

Cliff Brady

Look at injection mixing

Do consider injection mixing and possibly injection off of a buffer tank. This will give you an outdoor reset based temperature to your zones with greatly reduced spikes in temperature which should quiet things down.

At minimum, a Tekmar 356 Injection Mixing Controller and the properly sized pump and pipework will get you there. Hopefully your piping arrangement is recptive to this kind of alteration. Depending on the design of your current control system, you may also want to look at other Tekmar controllers that incorporate injection mixing with system control.

Get familiar with these Tekmar 356 Injection Mixing Controller documents and you should be able to come up with a plan that you and your contractor can agree on:

http://tekmarcontrols.com/acrobat/d356.pdf
http://tekmarcontrols.com/acrobat/a356.pdf
http://tekmarcontrols.com/acrobat/e021.pdf

Alex Giacomuzzi

By-Pass Valve; Diverting Valve; Mixing Valve??

Hi Ron

I tried going to the Oventrop web site and I was not able to locate your particular valve. Reading your comments on "How it is presently working" --- I would have to conclude that it appears not to be working. That in and of itself is not a good thing. Temp varies much too much without any adjustment change to the dial.

First of all...... regarding any testing, you could use the shut off valve downstream of the check valve after the bypass valve (close it all the way) and watch the temp. Them slowly crack it open and somewhat modulate that valve.
In all sincerety........ I am not exactly sure what the temperature input element does for this valve. I thought initially it was a mixing valve and that made sense. Sometimes I do not always agree that the schematic matches how a valve should be or is piped --- you know what I mean. But your schematic and pictures have nail it ......nice job.

So ---- I am not sure you have the correct valve for the application especially if you want to consider constant circulation and there appears to always be some flow by passing the valve. This maybe normal for the valve by the way. I believe you have a 3 way diverting valve or call it a by-pass valve.. You have 1 input and 2 outputs. As you explained, it is not a mixing valve (which has 2 inputs and 1 output). In either case, you may have to modify your system slightly and probably loose the valve....

As we talked earlier....regarding the buffer tank, here is what I would suggest. You have a significant load to begin with ---- I would consider a 50 gallon to 80 gallon buffer tank piped identical to your DHW tank with its own sensor or aquastat that incorporates outdoor reset. Pipe this tank counter flow.... hot into the bottom and return from the top. Now this would un-couple your boiler starts and stops from any of your zone calls. The boiler, its pump, and the boiler primary zone pump would only run when the loads in that zone would call. Not any of 17 zones. Your present boiler control scheme appears sound. I would hope that each of your boilers is piped with a by-pass capability as an added level of protection.

I would suggest consolidating several of the appropriate zones as mentioned above and leave several of those open all the time to run constant circulation. Only regulate those that absolutely require it. You can experiment with those.

Now.... your circulating pump loop to the residence loops, building 2 et al.. You can remove the bypass valve, and install 2 closely spaced T's. Use this as an injection point for a variable speed mixing control. This control piping would be piped back to the buffer tank.... supply from the top and return from the bottom. Add the necessary flow check valves and globe throttling valve into this injection piping and you should be good to go. Cliff Brady above mentions the same thoughts along with a recommended Tekmar controller. This is worth looking into.

Good Luck..........
Ciao...... Alex

Constantin

I was dense, after all...

... because that's not a mixing valve, as you pointed out. Looking through the Oventrop literature, the bypass valve as designed seems to be designed to maintain a constant temperature in the secondary radiant loop. Yet, if you want to minimize ticking and expansion noises, it won't help much, because the zones will stay shut most of the time, only to be hit by a slug of comparatively hot water when the t-stat calls for heat.

I have enclosed a small strawman proposal for you and others to burn. I've put a couple of changes into it that you and your installer can consider. For example, I piped the two boilers combined pri-sec, so that they always see similar water supply temperatures. That way, a modern staging control system can wear and tear them equally.

I have also moved the take-off for the IDWH ahead of the primary loop circulator. The idea being that you can save yourself the energy to run the primary loop circulator if the IDWH is on priority anyway. If the pool is never heated when the house is heated, you might also consider moving it to be parallel to the IDWH, I suppose.

Lastly, I put in a mixing system that is meant to cover your heating loops. This can be either a 4-way valve, an injection system, etc. your choice. The mixing system can be run by a controller like the Tekmar series, or you can go for a ready-made mixing block like the ones made by Buderus, Taco, etc. Then ramp the seconday loop temperature up and down with the season and that should minimize ticking.

The last heating item on the list is the pool because it can benefit from even the least amount of heat left in the primary loop. Putting in a buffer tank for the circuit on the other side of the mixing station is a great idea, Alex, as it'll definitely uncouple the boilers a bit and allow them to lengthen their burn time. I've amended the drawing to incorporate your great idea. the only question in my mind is whether perhaps an additional pump and loop would be justified.... it's there for now.

Happy burning!

RonV

Follow up and then its off to the races

I've got a pretty good picture now of how this is meant to go together. I'm a bit fuzzy on why to pipe hot into the bottom and cold out the top but I'm sure its just to go along with the simple physics of it. Anyway, the biggest remaining issue on my mind is the question of whether my system efficiency will go up or down or not really change much at all. I bet its the latter, because the only additional losses that we are adding are the stray heat losses from the buffer tank and any extra piping. This isn't much compared to the heating load itself. And in addition, the boilers should be able to cycle longer which may help improve their efficiency slightly (although thay stay hot all the time now anyway). At first I thought efficiency would be worse because we are heating up all that additional mass of water, but i think that's only an issue at the beginning and end of the heating season when the system might idle all day keeping the buffer a little bit warm. But with the outdoor reset, that shouldn't be a problem I guess.

All in all, I like it. So, what I'm gonna do is draw it all up real nice and neat and go run it past a local hydronics engineer to make sure I didn't do anything stupid. Sure wish I knew somebody competent locally. Oh well, that's a whole other problem.

By the way, I should mention that I talked to the local expert on the Oventrop diverting valve that I've got now and he confirmed that they never completely close. This means that it cannot simply recirculate the distribution return water and isolate the primary loop, and so the distribution water temperature does indeed keep going up until it hits about 155 deg or so when the sytem runs for a long time like under heavy load. So theres nothing broken in my system. He suggested simply lowering my boiler temperature from 180 down to about 160 or so. Yeah, i guess I could do that, but it won't solve my ticking noise problem.

Thank you very much for all your work on this Alex, Cliff and others. I appreciate it very much. If you have any remaining comments on system efficiency, please post.

Ron V