Boiler return pipe size and circulator replacement

JeremyG

I'm trying to keep my 1983 gas cast iron boiler going for a couple more years until I'm ready to finish the basement and make heating upgrades. I'd also like to improve comfort where possible. Here's what I've got:

1935 house with a converted gravity system
167 MTBU/H input, 137 output Weil-McLain HE (fan assist) boiler
94 MTU/H @180º cast iron radiators (found a copy of the American Radiator catalog)
60 MTU/H heat loss (Manual J performed earlier this year) on a design day, 17ºF outdoor, 72º indoor
B&G 100 series circulator, estimated 5 ft. head loss using B&G System Syzer for Windows

I've become aware that the boiler spends a lot of time below 140º because the boiler is way oversized and I have plenty of radiation. My band-aid was to set the boiler minimum runtime to 15 min, so at least the boiler cycle is ending at about 140º. Rather than plumbing a bypass loop, I'm considering a Taco VT2218 Delta-T with its "boiler protection" feature to keep the pump speed low until it reaches 140º.

Weirdly, the copper return piping near the boiler reduces to 1" before going into the circulator. The boiler input and output are 1.25", and the copper supply pipe is 1.25" as well. (Circulator to boiler is 1.25".) Steamhead's article suggests that I should have a 15 gpm pump, that's about the max for a 1" copper pipe, right? Is there any value to having a smaller pipe on the return side?

My current circulator is rated for 22 gpm at 5' head, so maybe the 1" pipe was intended to throttle an oversized circulator? I did the head loss calcs at 15 gpm based on Steamhead's chart and because head loss goes way up with that 1" pipe above it. So perhaps the circulator is actually delivering about 15 gpm due to the restriction of the 1" pipe?

On the comfort front, the two larger bedrooms upstairs overheat. They are each at the end of the two branches from the boiler, and the radiators are big. I will add TRVs if necessary, though wondering if trying a variable-speed circulator first makes sense?

Thanks for any and all insights!

EDIT: My first priority is to avoid flue gas condensation damage to the boiler; anything else is icing on the cake.

EBEBRATT-Ed

with a boiler output of 137,000 you should have 137000/10000=13.7 gpm which would be 1 1/4 copper or pipe. 1" is good for about 8 gpm but a small amount of 1" will not restrict a properly piped 1 1/4" job

Your heat loss of 60,000/10000=6 gpm but you need to stay with the 13 gpm to keep your oversized boiler happy

Zman

Slowing the boiler circ will not make the return temps hotter. In fact they will get colder. The water will spend more time in the radiator dumping heat. It may also result in uneven distribution. Now don't start thinking increasing the speed will help, it won't. It will just tighten the delta between supply and return temps.
I think your temp solution is about as good as it gets until you replace the boiler or repipe it.

JeremyG

Thanks for the info. I checked the supply and return temps where the iron pipe starts with an IR gauge and they were both around 135º after a 5º recovery from setback due to being away. That got the radiators up to 135-140º and the boiler up to about 160º. Sure feels nice and toasty! I'll check with a thermocouple for curiosity's sake, though the small delta-T corroborates the over-pumping hypothesis. My hope is the delta-T pump keeps the boiler itself hotter for more of the heating call. I can satisfy a design day with 150º water, so lower flow should be adequate almost all of the time.

Zman

The output of the boiler and the size of your emitters is fixed. The laws of thermodynamics take over from there.

If you are piped direct, changing the flow rate will have little effect on the average water temp. I will however change your delta t. With a direct piped setup lowering the flow will increase the delta T and lower the return temps.

If you change your piping to primary/secondary and add a boiler bipass loop, you can get true boiler protection.

SuperTech

The VT2218 in boiler protection mode is not enough by itself, trust me ive tried it myself. Your situation sounds exactly like my own. I can easily get by with 150 degree supply water almost any day, but keeping the return temps high enough was impossible. Even with a bypass.

My solution was to use the VT2218 as a variable speed injection pump with primary secondary piping. I added a Caleffi QuickSetter with flow meter on the return side of the crossover, this allows me to have complete control over primary and secondary loop temperature and time between burner cycles. Currently the burner might have one cycle per hour with the Grundfos Alpha on the secondary delivering constant circulations. I could even get away with down firing the burner with no worries.
These changes worked better than expected. Complete boiler protection, standby loss elimination, no possibility of thermal shocking the boiler and improved efficiency all by modifying the near boiler piping.

HVACNUT

@JeremyG
If you really plan to replace in a couple of years, why do anything now?
IMO use that $$ to save for the system you want, or with Christmas coming, and you and me being BBF's...

JeremyG

Great info — thanks y'all.

@HVACNUT: I am reticent to spend any money now, though HX failure would be a real doozy before I'm ready to upgrade to a mod-con and relocate the boiler and DHW. There was evidence of condensation on the vent pipe when we moved in last year and it look me a full heating season to catch on to how many of the heating calls were short, with boiler temps at or below 130º.

@SuperTech: Your account is very helpful, and along with @Zman 's input I take it there isn't a shortcut to correct boiler piping. Time to do some reading and find a good local contractor if I don't take the cheapest and easiest solution of burning a little more gas to avoid short firings.

JeremyG

Well, I couldn't help myself at bought and installed a Taco VT2218. The outgoing B&G 100 is only three years old, so it'll probably still fetch a decent price on eBay or whatnot.

It turns out the the boiler protection feature is only available in the setpoint heat mode -- and that only makes sense when it's installed in a secondary loop. So, I tried the Delta-T mode and fixed speed mode. Interestingly, when running in the lowest fixed speed, I get about 25º delta-T over a 15 minute minimum run. I played around with delta-Ts from 15-25º and settled on 25º for now. After 15 min, the boiler temp jumps to about 145º after a few seconds of heat soak, indicating that the HX should be out of flue gas condensation range.

I'm sure it's not as good as P-S piping, though it does a few things for me:

• Keeps more heat in the HX and supply piping and minimizes heating the return pipes of considerable mass (my previous delta-T was 5º). Not the end of the world in the previous scenario because the heat is still in the house, but might as well emit the heat from the radiators.
  
• Warms the house more quickly and evenly. The old pump moved so much water that the radiators with the longest runs were really cooking -- because they were designed with bigger pipes to work in a gravity system.
  
• Gives ~200 W more headroom on my modest inverter for backup power.
  
• Has sensors and a digital display -- so much more fun for the hopeless tinkerer Joking aside, it is helpful for this relative newbie to be able to see what's going on with temps