Boiler Replacement Project (Success! and Thank You!!!)

[Matt Meier]

I'm convinced. The simple simulation says that I can improve the efficiency up to about 11% by varying the system flow based on operating conditions. It's not a completely intelligent simulation. It doesn't try to optimize the flow, or balance the system over time. Those are things I wouldn't try to debug for this fall anyway. It's also only for one boiler/pump/zone/imaginary house combination. However, that's a sizable gain over time; enough to make me believe that it's probably worthwhile in most cases.

The downside is that I can't find suitable pumps. It seems the small residential ones are all boxed-up units, and you have to follow their prescribed operating curves. I can get pumps with speed-control inputs in the larger commercial units, but those are big, and also quite a bit more expensive. I'll work around that eventually, but I have a couple new questions.

First, I keep hearing that induction motor pumps are terribly inefficient, and that the ECM pumps are infinitely better. Why? It's just an induction motor, and we can get well into the 80% range even with older induction motors with proper control. Is this only because we assume that induction motors are full-on/full-off, and that the ECM pumps can modulate?

Second, I've read quite a few times that a large ECM pump with zone valves is much cheaper to operate than using zone pumping. Why? Flow is flow, and pressure is pressure, and they multiply to give the delivered power. For example, suppose we have one pump, pumping three circuits. Given an 8ft drop across the system, and the first circuit flows 2gpm, the second 3gpm, and the third flows 3.5gpm, that gives us a total of 29.4W. Assuming the pump is 85% efficient, that's 34.5W into the pump. Now, if we split the zones with three pumps, each operating at 85% efficiency, then we get the same total. Aside from the fact that the three pumps will be operating at slightly different efficiency points, where's the difference? Adjusting the operating points of the pumps slightly, assuming they're properly sized, won't create the discrepancy that's argued all over the web.

Thanks again. This is getting fun!

Matt

 

[Matt Meier]

Never mind the pump comment. I think I've found one line that will work.

 

[Sponsor]

 

[Matt Meier]

Well, I found a lot of asbestos, so I've been not working in the basement for a while. Now that it's cleared up, and I'm back to the project, I should clarify a few things.

First, I shouldn't have referenced improving the efficiency by 10%. First, that's fundamentally incorrect. For example, a system operating at 92% efficiency can't have its efficiency increase by 10% in either an absolute sense or in a relative sense. In addition, it's like lumping apples and oranges when comparing the BTU input to the energy conversion from electricity to moving water. So, the actual number I came up with was approximately a 10% drop in energy use. It was just horribly misquoted. I complain about this all the time, so I should have known better.

Also, in my question about one large motor vs. several small motors, I didn't factor in the efficiency of the pumping mechanism. It was a brain fart, because my head was spinning trying to figure out why people keep posting on the internet that zone pumping is SO much worse than an ECM with zone valves. The question remains, but off by a factor of 3. At any rate, I think I've resolved that issue too. It does appear to be based on the assumption that zone pumps are single-speed, and have to run at full power. If true, that certainly causes the energy use in zone pumping to skyrocket, because it's hard to find a single pump that's appropriately sized to a single zone when operating at full power. However, there are a handful of very small ECM pumps available now that fill that void. I'm waiting on energy-use data from one such manufacturer to verify this.

Unfortunately, some of the boilers I was considering don't seem to have external temperature controls. What I mean is that my control theory requires the reset curve to be adjustable in real-time by the master controller, based on which zones are operating. It's a consequence of the fact that the heat exchangers aren't balanced with respect to the heat transfer at a given temperature. In other words, some zones require hotter water than others at a particular outside temperature. Not to mention the fact that zones can request different zone temps. I could just set the reset curve for a worst-case situation, but that would push the system energy use far enough from the optimum point that all the additional control effort wouldn't really pay off. So, I'm still shopping for a boiler, but luckily there is still a pretty decent selection with provisions for analog input temperature control.

I have a final issue: Venting. I'm not going to be able to cut holes through the fieldstone, obviously. The rim joist is 6" thick, with what I think is only a 4" brick wall on the outside. I'm likely going to just bottom out a hole saw and chisel out the core, and repeat and repeat, and then use a diamond hole saw on the brick. Does that sound like the simplest and cleanest way to do it, or is there some creative method I'm missing?

Matt

 

[Dana]

It may be possible to install the venting inside the pre-existing flue cavity, if it isn't being used by other appliances. (A coaxial vent for both combustion air & exhaust can even fit inside a 6 x 6 tiled flue.) Talk to local code officials to verify that it's legal in your area before going that route (it isn't always.)

Alternatively, there is sometimes sufficient space in plumbing stack chases to share the chase for taking the vent through the roof:

It's also probably possible to install the boiler somewhere else in the house (even a first or second floor closet re-purposed as a utility closet), if that makes the venting issue easier or more cosmetic.

If side venting, be sure to terminate both the exhaust & air intake at least a foot above the historical highest snowpack depth- more if drifting snow is an issue. Sometimes that means making something of a snorkel out of it:

In addition to venting you have to consider how you are going to manage the condensate disposal. It's mildly acidic, and it'll be churning out on the order of a gallon of condensate per therm of natural gas burned, which in your house is going to add up to something real. Dumping it into a sump under the basement slab in a not-so-well drained slab isn't advised without neutralizing it first, but it's fine to dump it in cast iron drains.

 

[Matt Meier]

Update:

The condensate is going into a PVC floor drain, so that should be fine.

I moved the mechanical room to the other basement, so everything is venting out an old basement window that had been broken and boarded up. I replaced it with some glass block and ran the pipes through that.

I installed a 100kBTU/h Westinghouse UFT unit. I had two reasons for selecting this. First, I need something that modulates REALLY low, because of the micro-zones. Of all the boilers out there that I could find, this one has the second-lowest minimum firing rate at 10k, second only to the 80k units with a 10:1 turndown ratio. The smaller units all had lesser turndown ratios, so even the 55k boilers (the smallest I could find for sale) the minimum firing rate was 5.5kBTU/h. I went with the 100k for peace of mind, based on my most aggressive 99% heat load.

My final issue is my flow rates. I'm really just looking for a sanity check and some reassurance. Not that I'm going to rip out all my piping and start over before this winter, but if it's worthwhile I might do so next summer.

I have a final design for 12 zones. Some zones don't exist yet, and I have some expansion room for an addition. Anyway, my worst-case heat loss is 89kBTU/h. With a 20° delta T I'm looking at around 0.5-1.5GPM per zone at design temperatures, and much lower under normal operating conditions. Further, I want to increase the delta T in the future. I can't now because the smallest pumps can't provide flow that low, but I have a plan to fix that with my own Frankenstein units.

My concern is that somehow my pipes are way too big. With 0.5GPM, I should apparently be using 1/4" or 3/8" copper. That's so geometrically ridiculous that I just said "screw it" and plumbed everything with 3/4" pipe, which makes the step-up/step-down fittings really compact but makes my flow velocity really low. I'm not worried about air bubbles; I can periodically run the pumps up to higher speeds to push them along until they're completely purged. However, there are so many unexplained warnings about low flow rates that I can't get them out of my head. I know logic says that the 2" pipes in the old system were just fine, so this should be fine, too, but I can't get it out of my head.

Is there any other issue with oversized distribution pipes besides the initial cost?

Thanks for all the help from everyone. This site is the primary reason that I didn't just hire someone to put in a replacement cast-iron boiler, and I'm convinced that it's also the only reason that I don't have a 200kBTU/h unit in my house.

Matt

 

[Matt Meier]

Well, after the past couple months it's safe to say this was a success. I'm still working out the control software, but the system does run the house (even if I have to go into the basement to make adjustments).

The good news is that the boiler hasn't shut down since I turned it on, except when I cut the power to wire in the last few pumps. Even with the micro-zones (even the 7x12 laundry room has its own zone), I've been able to stagger the heating demands so that the boiler fire request never goes out, and the boiler never heats beyond the set point. I'm sure that will change this spring, but I did a "shortest cycle" test with only the smallest radiator loop pumping, and the boiler fired for over 5 minutes before shutting itself down. My only regret is that I can't control the firing rate on the boiler. When it starts up, it tries to reach the demand temperature quickly. I could back that off by holding the firing rate at the minimum when only the small zones are calling for heat. Still, the internal controls do pretty well. I suspect I'll get some consistent cycling in the early fall and late spring, but the "nice" thing about my part of Michigan is that we transition through the 40s and 50s pretty quickly. The other good news is that the average power use by the pumping array has been less than 30W. I know I can do better by rigging a circuit to the motor's speed control so I can adjust the motor speeds in the control software, but I'm not sure I care enough to save $5/year.

The only bad news is that the 100k unit is, as we realized two weeks ago, a little bit too small. It couldn't keep up when the temperatures were in the low teens (the boiler couldn't meet the required water temperature to heat the living room or the dining room/foyer areas), but the house wasn't terribly uncomfortable. We just turned on a space heater in the living room and it felt just fine, and we don't really use the foyer or dining room anyway. The worst-case scenario is that we would put in a second unit (I left room to pipe a second boiler on the primary loop), but I seriously doubt it will ever come to that. Adding a second radiator in the living room should allow the living room to heat with a lower water temperature, and the second floor shouldn't suffer too much because those rooms will get back some of the lost heat through the ceiling. Basically, it should balance out the heat distribution enough that we'll never notice.

Again, I can't express enough how much I appreciate everyone's advice and PATIENCE as I stumbled through this process. I know I can be difficult, but you've all been wonderful.

Matt

 

[NY_Rob]

Nice of you to post your results!

If your Westinghouse UFT mod-con is based on the HTP UFT series.... have you enabled "step modulation" to prevent the boiler from throwing all 100K BTU's at the CH loop on calls for heat? Rather than starting off at 100K BTU's, it starts off at the boilers lowest fire rate (10K BTU's for your boiler?) and ramps up to 100K BTU's in six steps at 1min intervals. It certainly helps on those short zones.

 

[Matt Meier]

Yep, I've enabled the step modulation, but I set a 10° temperature drop before the boiler resumes firing (the factory was set to 30°F), just because I didn't want the temperature to yo-yo while I was adapting the software and taking some measurements. It's obviously not necessary, considering the boiler is always firing, so I'll loosen that up before spring. I just wanted a worst-case scenario, and even that isn't too terrible.

You mentioned that the boiler starts out at minimum firing rate which is, indeed, 10kBTU/h. However, according to the manual it starts at 40% combustion and then increases by 12% each step. I assumed that was 40k to start, but maybe I'm missing some sort of conversion?

 

[NY_Rob]

My manual (HTP UFT-80W) doesn't mention a percentage of max for the starting rate with step mod enabled- see attachment.
BTW- I believe there is a typo in the description- it states "Step modulation will start at the last modulation rate of the boiler and work up one minute at a time." where that is actually impossible because if the "last" modulation rate of the boiler was 100% it would be starting off from there.
I believe it should read "Step modulation will start at the lowest modulation rate of the boiler and work up one minute at a time."

From many hours of close observation (much to my wife's displeasure) of my boiler- it indeed sounds/feels like it is starting off at it's lowest fire rate of 8K BTU's for my boiler and increasing in six one min steps to max after that.

 

[Matt Meier]

I missed that. I guess I didn't read that part closely enough. I wouldn't be surprised about a typo. There are a couple places in the manual that aren't very clear. It looks like our manuals are identical. If they are, there's a chart on page 62 with the full description of step modulation:

Based on the description, it looks like the burners start up at 40% combustion. I assume the heat output is proportional to combustion %, but I do admit that I don't know for sure.

Either way, it definitely slows down the heat delivery on start-up, so that's a big plus for extending the combustion times during warm weather.

 

[NY_Rob]

^ the bottom of my page is blank where step mod fig. 51 diagram is found in your manual.

I also see some differences in the plumbing diagrams between the two manuals...
In the Westinghouse direct piped schematic they added a flow check on the spaceheating return in addition to still leaving the IFC in the CH pump where the HTP schematic only indicates the IFC in the CH pump.

 

[NY_Rob]

Just updated my 2/3/2016 HTP manual to the 6/2/2016 version that you have... we now match!
Interesting they have now added additional flow check valves.

IMO- there's still a problem w/the above diagrams concerning air removal if you install during the summer and only use DHW for the first few months.