WM Ultra P/S

Brad Barbeau

I had a new Ultra 155 installed last year (<a href="http://www.heatinghelp.com/forum-thread/135093/My-new-Ultra#p1221431">http://www.heatinghelp.com/forum-thread/135093/My-new-Ultra#p1221431</a>) and I've been pondered how it was piped.  The layout was executed by out contractors to a design provided by the supply house.  The Ultra manual specifically says: <strong>"Use only primary/secondary piping as shown.</strong> Failure to follow these guidelines could result in system problems." (<a href="http://www.weil-mclain.com/en/multimedia-library/pdf/weil-mclain-pdf/products/boilers/gas-boilers/ultra-series-3-ue/ug3_boiler_manual.pdf">http://www.weil-mclain.com/en/multimedia-library/pdf/weil-mclain-pdf/products/boilers/gas-boilers/ultra-series-3-ue/ug3_boiler_manual.pdf</a>, page 12).  The tees are supposed to closely spaced, which mine are not.  I've attached a simplified diagram of the boiler loop.  The pictures above will provide a bit more detail. 



Now, I'm not having any system problems, but I did notice something last night.  I've added another zone (the coach house) into the stubs that were left for that purpose.  That zone is heated minimally at the moment, so when it turns on, it dumps a whole lot of 50 degree water into the boiler loop which then gets pumped to the house zones due to the piping arrangement.  I could feel the pipes getter cold as this happened last night.  Now perhaps I don't understand enough about primary/secondary piping, but I have the idea it's not supposed to work that way. 



Is this piping arrangement okay despite the fact that it violates WM's instructions? 



Thanks for your comments.

icesailor

W/M Ultra P/S

That is a beautiful thing. I looked at the first comments about the controls. No mention of the piping.

Somehow, it strikes me that that is not how Weil/McLain intended Primary-Secondary piping to be done. It would seem to me that when you add the coach house, you end up making the primary loop a dump loop for the cold returning water. The tees are too far apart. If they are too far apart, the water doesn't circulate through the closely spaced tees but circulates through the boiler.

Maybe I am wrong, but my gut says no, and I would never let them get that far apart. The spacing is a matter of pipe diameters, not an arbitrary dimension. One thing I have from Taco says that it SHALL be no more than 4 pipe diameters. That makes 1", 4" apart. Less is better.

Others more experienced than I will chime in and say I am wrong, or right.

I looked at a house today that had a replacement Combi Mod-Con, connected in the wildest way to a floor system of that 1992 Orange Heatway Entran II tubing with a couple of loops disconnected. I told them not to touch a GD thing and PRAY.

Jean-David Beyer

I am not a contractor, but I do have a W-M Ultra 3.

I do not see why your system would not work. I assume you have a circulator in each of those secondary loops. W-M specify that the closely spaced Ts be spaced no more than 12 inches apart. Mine are 5 inches apart and made of 1.25 inch copper tubing even though 1 inch is enough (80,000 BTU/hour). John Siegenthaler's books suggest the Ts be no more than 5 pipe diameters apart, and he should know.



On the other hand, that is not how W-M say to pipe it. The way they have it piped, there is only one set of closely spaced Ts, in the primary (boiler) loop. In the secondary loop, you may have one or more circulators -- one for each zone, if you zone with circulators, or just one if you zone with zone valves.



Piped as yours is, your carriage house gets first chance at the hot water and after it has cooled warming the carriage house, what is left is available for the rest of your house. If your resets are set so carefully that the circulators run most of the time, you will not be getting a slug of cold water from the carriage house messing up the rest of your house. So you may wish to really fine tune your outdoor reset. In my house I have tuned it very fine and the downstairs radiant zone runs for long periods of time; once it ran 18 hours straight. Upstairs is fin tubed baseboard and not adjusted quite so fine.



If you have preference and do not choose to finess the reset settings, you might pick up first for the house, and give the carriage house the second chance at the primary loop. It does not feel good to me, but I am not a contractor.

kcopp

There is more....

Than one way to Pipe primary secondary... I don't see an issue here. You have the flow rate for the boiler and the tees out to the system are closely spaced. Looks neat and tidy.

Tim P._3

Two things worth investigating

How may BTUs does each zone require?



What are the pump sizes?



If the system pump isn't variable speed it needs to be putting out more GPM (typically more than the total of the secondary pumps) given your scenario of 50 degree return water.



Seems to me the boiler is putting out 10gpm, the CH is taking all 10gpm, then returning low temp water, then the house is taking 10gpm of low temp water.  If the system pump was pushing 20gpm, it would give 10 to the first loop and push 10 past the first set of tees.



Would the CH pump and the system pump happen to be the same size?

Ironman

P/S Piping...

Can be done several ways and we are sometimes vague with the terminology. What you have is indeed a form of p/s where the boiler loop is the primary and your zones are secondary. The W-M manual shows a primary loop where the boiler injects into that primary loop and the secondaries pull off of it.



To clarify, my understanding of what is the primary loop: the circuit that passes straight through the run of the Tee's is the primary; that which comes off the bull of the Tee's is the secondary, whether it's extracting (the zones) or injecting (the boiler) heat. Thus, you do have p/s, it's just that the boiler loop is the primary and not a secondary injecting into the primary as the manual shows.



There's nothing wrong with piping a boiler the way they did yours as long as it matches the temp requirements of each zone. I just did one like that last week. Doing it the way they did yours is primary/ secondary series. The temp in each descending loop is decreased. This OK if each loop down stream requires a lower temp than the previous. But, if they both require the same temp, then it should have been done p/s parallel like the manual shows with the "moose antler".



You can reverse the secondary connections so that the main house comes off first or you could put a strap-on aquastat between the carriage house Tee's and the main house Tee's that would hold off the main house circs until the loop temp catches up.



4 pipe diameters between the closely spaced Tee's is ideal design, but many manuals simply say 12" maximum. Either should work.

Brad Barbeau

Thanks for all your responses

Yes, they are both the same temp. I suppose that I could repipe in the moose antler at some point, likely when I change to valves which I am considering. The house has four circ zones (Grundfos 15-58) which I would like to replace with one Alpha and valves - unless someone really thinks I should go four Alpha's instead, seems cheaper to do it with valves. The coach house has a 15-58 feed it with two valves off that.



Load in the house is ~80K btu, CH is ~35K btu. There is no system circulator, only the boiler circ which is a Taco 0014. Since the 15-58 are variable speed, I can keep the CH on low which does seem to work, my math indicated that given the head requirements it needed to be on high.



I guess I'm most curious about this in case there is a warranty issue with the boiler and WM says nope, it's not piped properly, our instructions weren't followed.

Ironman

No Boiler Issue...

That I can see in this case. The manufacturer wants it p/s to assure a proper flow rate through the boiler. You should have no problem there. Your problem is a "secondary" issue. (Pun intended).

Jean-David Beyer

what is the primary loop

I have tried to figure out what is the primary loop, and what is the secondary loop. As far as I can tell, there is no universal agreement on this.



My definition of primary loop is the one with the boiler in it. The other loop(s) are secondary loops. If they have sub loops, they could be called tertiary loops, but at some point the stuff may work, but understanding is difficult.



I try to be more clear sometimes by calling them the boiler loop and the load loop, but I forget.



As far as the sort-of original question, it seems to me that what the original poster in this subthread has could surely be called primary-secondary. And assuming his boiler is big enough, and the flow in the boiler loop is high enough, he should have no problems. Since he experiences problems, he could change circulator sizes, or make his reset close enough that the circulators run most of the time.But as Ironman pointed out, this system is acting as though there is insufficient flow in the boiler loop so the carriage house secondary loop (the first one) is hogging all the heat, at least when the carriage house circulator is first turned on.

icesailor

P/S

OK, tell me I am wrong as to how I understand the physics of it.

The reason for the closely spaced tees, not to exceed 4 pipe diameters, that's a 4" nipple on 1", is so that whatever flow is in the primary loop or the boiler loop, is whatever it is. Keeping the closely spaced tees as close together as possible, will cause local circulation on the secondary loop THROUGH the primary. There is where the mixing tales place. In the boiler piping pictured, the primary flow is circular as it flows down the pipe. The secondary flow being horizontal, is laminar. The extreme turbulence causes the mixing and balancing. I don't think that the goal if P/S is the have the secondary circulator pump the water back to the boiler, only pump It through the primary loop and flow in the secondary loop. The wider apart the tees, the greater the chance of loosing the circulation through the loop and it returning back to the boiler.

I have an article in a piping manual about how Gil Karlson figured out Primary Secondary piping. It had to do with a power plant built at Notre Dame University. The boilers and piping for the plant were designed by one engineering company and two or three new dorms and related piping were designed by another company. The circulator pumps in one side (maybe the power plant) was grossly over sized and the pumps on the other side (dorms?) were not. The mis-match caused the water to not flow from the power plant to the dorms. It was a huge mistake and expensive to fix. The idea of the closely spaced tees came up and it was found that inducing whatever flow in a portion of another pipe would correct the gross mistake. Hydraulic Separators or Low Loss Headers are a better route but just another form of P/S piping.

Or how I understand it and it works for me.

Jean-David Beyer

tell me I am wrong

Since you are a pro and I am just a very interested homeowner, I hesitate to say you are wrong. The trouble with typing stuff on a computer instead of a face-to-face discussion is that misunderstandings are so easy to make and so difficult to straighten out.



"The reason for the closely spaced tees, ...., is so that whatever flow is in the primary

loop or the boiler loop, is whatever it is. Keeping the closely spaced

tees as close together as possible, will cause local circulation on the

secondary loop THROUGH the primary. "



I would say this differently, and perhaps I am contradicting you (I am not sure). I would say, the reason for the closely spaced Ts is to make sure the pressure drop between one T and the other is so low that flow in the boiler loop will not induce flow in the load loop(s). The only flow in the load loop should be caused by a circulator or circulators in the load loop(s). Similarly, the only flow in the boiler loop should be caused by the boiler circulator.



This pdf by Caleffi, bottom of page 2 shows nicely how they work.



http://www.caleffi.us/en_US/caleffi/Details/News/files/01076_na.pdf



I agree if you say that a low loss header and closely spaced Ts are equivalent. An LLH may be economically better than closely spaced Ts if the labor costs and the fact that many LLHs contain air removal and dirt separation, simplifying plumbing.

Mark Eatherton

Primary, secondary, tertiary, quaternary, quintenary...

This argument has been around since, forever... In my minds eye, I consider where the PONPC is located to be the primary, only because THAT is the one place that all pumps MUST pump away from...



Other well respected authorities state that where ever the HEAT source is, is considered the primary.



Personally, I think it is just semantics, and so long as the system is piped correctly, it really doesn't matter.



Back in the good ol' days, there was one primary loop, with its own (oversized) pump that did nothing other than move water in a circle. Connected to that were secondaries. Input secondaries and output secondaries. .Then, the industry has an epiphany. There really was no need to have water doing a circle by itself for no real good reason, other than being considered the primary loop.



These days, the boiler loop is a part of the primary. In fact it IS the primary. Secondaries can also be heat sources (think solar and geo source) so it is still confusing.



The critical things to remember are that the boiler MUST (in most cases) be on a loop all by itself. This is somewhat critical in order for the control logic (modulation portion) to work correctly so it can "see" changes in the load as they occur and respond accordingly. If its flow is a constant, then the only variable would be the load, which affects the boiler loops temperature, positively, or negatively.



As has been previously stated in different ways, the reason behind having closely spaced tees is so that the pumps can't "see" each other, and have ZERO influence over one another. That's all.



To the original poster, other than having secondary take off tees too close to an elbow, I see nothing wrong with your drawing.



ME

hot_rod

a high mass emitter

like a concrete slab can pull the boiler temper down and keep it there until the mass warms up. If the coach house zone is a concrete slab of any size it will take some time for that boiler loop to increase in temperature. A copper fin tube would be one example of a low mass emitter that would respond faster, and allow the return temperature to increaser faster.



If you want multiple secondary zones to receive the same temperature you would need some cross over bridges with balancing valves to dial in the temperature.



There are a number of devices and fittings on the market to enhance P/S installation and operation. I think Webstone and a few others have added P/S components since this article hit the street in 2003.



www.pmmag.com/Articles/Column/0d004a4adffc7010VgnVCM100000f932a8c0____





The close spacing between the tees is intended to keep the pressure drop as low as possible, although it can never be zero.



It is still a good practice to have checks on those secondary take-offs. A pump with an intergrel check covers one side, a spring check would assure positive shutoff on the other. Any time you pipe straight off the top of a horizontal piping run you have the potential for ghost flow. A check on supply and return can prevent un-wanted heat migration.



Also the primary piping size needs to be calculated to handle the GPM flow required for the largest load.



hr

Brad Barbeau

Just CI Rads

Everything is CI rads.  I've got the reset curve adjusted to pretty closely match the loss of the house but the CH is different because the temperature is lower in there right now.  I had wondered about using one of the separate temp inputs on the Ultra for that area.  I'm not sure that will make any difference because the AQ control won't turn off the house zones when the CH calls for heat unless I've missed a setting.  Also doesn't work too well given how close I have the ODR curve set to the loss.



Glad to know there isn't a problem with the boiler loop though!

Jean-David Beyer

Resets and thermostat inputs.

I have a radiant zone (copper tube in slab) and a baseboard zone (Slant/Fin 2000, oversized). I heat these to about the same temperature, but to do that I must run hotter water into the baseboard than to the slab. The slab gets water between 75F and 120F; 112F at design temperature. The baseboard gets water between 110F and 134F; 130F at design temperature.



What I do may not work for everybody. I use all three thermostat inputs on the W-M Ultra 3. Priority 1 is for indirect hot water heater. Priority 2 is for the radiant slab, and priority 3 is for the baseboard. Let us ignore the indirect.



If ONLY the slab is calling for heat, it gets it; circulator 2 (boiler circulator) and circulator 3  (for the slab) are turned on, the reset curve for the priority 2 is applied and the water temperature is what is set up for the slab.



If INSTEAD, ONLY the baseboard zone is calling for heat, only the boiler circulator is turned on, and the reset curve for priority 3 is applied. A separate relay listening to the thermostat, turns on the circulator for the baseboard circulator.



Now if both the slab and the baseboard are calling for heat, there are two possibilities. The difficult one is if the slab started first, and everything is running for that. What will happen is that when the baseboard calls for heat, its circulator will turn on, giving the baseboard zone water at the temperature required by the slab. This is better than nothing, but not enough to do a good job. For me, that zone has very low heat load, and this will not be a problem. But others might not find this acceptable. Now the controller is set so if the slab zone is still running 30 minutes after the baseboard called for heat, the priority 2 conditions will be interrupted for 20 minutes (turning off the slab circualator), and the baseboard will run at baseboard temperatures. After the 20 minutes have expired, the baseboard zone will yield control back to the slab. If the slab no longer needs heat, the baseboard will get all it wants.



I ran it this way the first two years. Since the slab runs a lot of the time (not today when it is 58F outside), the slab gets most of the heat it wants (at least 30 minutes every 50 minutes), and the baseboard gets 20 minutes every 50 minutes. This is almost enough on cold days, but I have now enabled the boost feature where if the baseboard is still not satisfied after 2 hours, the reset temperature will be raised 10F so recovery will be quicker. And after another 2 hours, the reset will be raised another 10F, and so on (though it will never go above the maximum setting for that zone). I just started doing that this year, so time will tell, but is was almost close enough without using the boost the last two years.



Note that I set the temperature of the baseboard zone down 4F every night so at night the slab zone gets all the heat it wants. I have not decided if resetting that zone actually saves any money. My guess is that it is pretty close either way.

Gordy

Agree with ME, and Ironman

 Primary being the PONPC which usually is in the boiler loop so air removal is getting the hottest water. Not always that way though.



 I think OP needs to analyze how the heating habits of the carriage house will be. As long as the house is not calling for heat then that secondary circuit will not see the cold water rush from the carriage house secondary when it is calling for heat. But when both are calling it will always be an issue as specially if the carriage house will be in setback often. If that is the case I like Ironmans suggestions on control strategies for your issue. Suggestion #2 being like domestic hot water priority.



Gordy