Caleffi 280 Thermostatic Mixing Valve

tcovert83

I think its correct to say my radiators are "bigger" than my boiler: there's 2100 sq ft of EDR in old cast iron radiators, and using Formula 2-2 from iDronics issue 35, that tells me I'd get (need?) over 300,000 btu/hr with an average water temp of 170F and a 70F room. However, I should emphasize that during the cold snap the house was plenty comfortable with (if I'm remembering correctly) 130-140F water temps, which is ballpark 200,000 btu/hr using that formula. That plus my massive exposed supply and return lines in the basement probably matches the 235,000 btu/hr output of the boiler. The boiler didn't run the entire day, but plenty of it (I think 16-17 hours according to the ecobee logs).

Yes, during recent service calls the boiler does cycle off, and I verified (on the control board's LCD screen) that the boiler temp during these short off cycles is near the setpoint (178-179F).

I have ordered the 130F cartridge and will give that a shot, but your previous message about the flow rates I need to get sensible delta T in my system makes me think I will probably need a different circulator. Is that right? One step up from the NRF-33 would be the NRF-36, I think. Would that be a good choice? What else could I do here?

hot_rod

I suppose it depends a bit on how the radiators are piped. If you end up with a wide delta the radiators at the end of a loop may not output enough heat?

That was the one design criteria with gravity systems the length of the run outs that would allow enough gravity flow. This is one reason B&G developed the "booster pump" to allow conversions, or performance improvements on gravity piping systems.

I would try the lower cartridge first. My concern is the mixing valve has added just enough flow resistance to push you to the limit, maybe beyond of what that pump can do.

But if the system ran those low SWT/ RWT without the valve, you really need a protection valve. you do not want to run that boiler constantly at 130- 140 supply temperatures. Did you note the return at the boiler at those temoerature conditions?

Without being able to calculate the piping circuit, it will take a few trial and error steps. If the pump is not up to the task you could upsize or add another 33 in series to double the head.

If we assume you in fact need to move 23 gpm to get the job done (230,000 @ 20∆)

Looking at the pump curve the 33 will only get you about 5'. Two in series would get you closer to 10' head.

Attached is a 23 gpm flow in a 17 Cv valve, 4.23' head.

Another check would be to pull the top off the air sep, we have seen a number of cases where the mesh or media inside gets plugged with teflon tape , rust, sludge etc. That will add a bit more flow resistance.

But you want to get to a design condition or near it to get the info you need for the next step. It is always easier to get heating systems to perform adequately on mild days :)

tcovert83

I did use a (cheap/digital) strap-on temperature probe that I wrapped around the return pipe with some electrical tape and a piece of pool noodle for insulation prior to the install of the protection valve. That was giving me readings way below 120F nearly all the time, including at the end of heating cycles. I'm not confident that it was super well calibrated, but I would be surprised if it was 20F too low…. This is why I decided to install the valve in the first place.

regarding circulators: I think an NRF-36 fits exactly in the space that my nrf-33 currently occupies. If I find that I'm still getting (some) cold radiators after swapping out the 140F cartridge for the 130F cartridge, do you think this circulator upgrade would help?

hot_rod

More flow always = more btu output from the heat distribution. And yes even in cast iron radiators that were on a gravity system.

I've tested 4 different cast radiators from 1/2 gpm - 8 gpm. Increased flow always sped up the heat output and raised the average temperature across the radiator.

That being said once over 2-3 gpm the output increase is not very significant. See how the below curve flattens at @ 2 gpm. So depending on the pump size and power consumption the last few % may not be worth chasing. head loss increases with the cube of the flow rate. If flow is doubled head increases 3 to the 3rd power.

Graph 2-13 is pretty accurate for any type of heat emitter.

Fig 2-14 is a 250' loop of 1/2 pex in concrete, SWT costant 110°

With radiators it is mostly a surface area game, the greater the surface area the steeper the slope.

So to answer your question the PL 36, considered a high head circ, would move more gpm. I doubt with large diameter piping and wide open radiators that you will get velocity noise. This red valve looks to be a gate valve? If it is a globe valve it would be a perfect balance valve. Even the gate valve would allow you to do some balance.

The PL 45 is a bit flatter curve, but a bigger body pump 8-1/2" flange to flange.

If money were no object I'd consider a small Grundfos Magna that you could vary speed and save 50% or more on power $$ It would involve flange changes however.

This small Magna in the 20- 25 gpm range is operating at 83% efficiency.

At the B&G site you can enter system spec and it shows pump options. I put 23 gpm at 10'. The PL 36 has a wire to water of 31% Where the black and green lines cross is the operating point OP with the numbers 23 gpm, 10' I entered.

The PL 45 is on a better spot on its curve so 37% wire to water efficiency.

Any Y strainers in the system?

tcovert83

I do not have a Y-strainer, at least I haven't found one anywhere yet.

Is there a reason you picked the PL- line of circulators instead of the NRF- line?

hot_rod

I just entered your spec at the sizer program and it gives the selection. I'm not sure why no NRFs show up?

Performance is similar. The PLs are open frame motors, a bit more efficient compared to a wet rotor where the motor is sloshing in the fluid.

This pump curve chart below shows more of the NRF 3- speed selection, at various speeds.

If you want 23 gpm at 10', Speed 2 on the NRF 36 or NRF 45 is on the money. Since you want to work with the flange to flange spacing you have the 36 is the choice.

tcovert83

thanks a lot for your help! I will post again in a few days when I've got the 130F cartridge to describe my (short term) experiences.