re-install cast iron radiators

jonny7117

I posted this in another forum and members suggested I post here as well for the expertise this forum has. Apologies if you're seeing this twice.
This past summer we bought a 1933 house in metro Boston with a forced hot water hydronic system complete with cast iron radiators and a 40+ year old peerless gas boiler. Against my better judgement, we proceeded to remove all the radiators, replace the boiler with a Bosch Greenstar 100k (95% efficient) with indirect DHW, and an Unico high velocity system for heating and cooling on both floors, zoned separately. So hydro-air heating. The air handler is a Unico M3642 (3 ton) unit and the condenser is a Bosch BOVA-060 5 ton unit with heat pump. I'll note that I just had a blower door test which came back at 16 ACH50 for the 2000 sf balloon framed house, and we're trying to bring that down with air sealing.
Fast forward to the heating season and the system is pretty unbearable. The heat pump is set as primary heat down to 40 degrees F, and then the hydroair system takes over. The blower speed is turned down as low as it can go, and it still sounds like a windstorm outside everytime it turns on. The high velocity system just makes it feel drafty with the cool air temp from the heat pump, though the boiler is not much better. At just below 40 degrees F, the boiler heats supply water to cose to 160 degrees F, which I assume means it's never able to condense because I suspect the return water temperature isn't more than 20 degrees lower. The system does modulate temperature, but I've never observed it below 140-something, and I wonder if it's just cooling down then. The system also cycles multiple times per hour as it reaches temperature, stops and the indoor temp immediately drops. I don't know if this means the boiler is close to short cycling. When it was 25 degrees outside last week, the hydroair just never turned off for hours continuously, but at least the exiting air temperature was warm enough to not be uncomfortable. Even when the outdoor temperature was only 35 degrees, it took 10 hours for the indoor temperature to increase from 64 to 70 degrees (setback for a vacation).
In a nutshell, all of my fears about the new system compared to the old have come true. The one thing the system does well is maintain even temperature everywhere, and I suspect that some of it's poor heating attributes will serve it well for cooling.
Am I crazy to want to put CI radiators back in and use the new boiler for what it should be good at doing? We gutted the low height cast iron piping in the basement for future head height, but all the CI piping to the second floor is still in place. In an ideal world, I'd hope the system could be repiped with oxygen barrier pex for home runs and zoning by floor. I'm not sure if these two different hydronic systems would be compatible, or if we'd just have to abandon the Unico heating coil and redo the boiler for radiators.
I did a load calculation usig the Slant/Fin app which suggests 70k btu/h @ +5 degree outdoor temp, 70 degree indoor temp. It's also been suggested this is overestimating -- it's ~35btu/sf. The Bosch 100k boiler feels oversized considering it only has a 5:1 turndown. I calculated the CI radiator output and it came to 118k btu/h @ 180 degree water temp. So definitely oversized to be able to lower water temp into the condensing region for a lot of the heating season. Any way to make the current boiler work for higher outdoor temperatures without short cycling?
This is definitely a costly mistake, but I'm taking the long term view for comfort. Thanks for your thoughts, very much appreciated.

KC_Jones

Couple things in general terms.

If you hear the high velocity system something isn't 100% correct. I've been in houses with properly designed high velocity and you don't hear it. Sounds like a possible design/sizing/install issue.

Second, with a properly designed system it may take many hours to come out of a vacation setback. Do you have outdoor reset on this system? If so that could contribute to the long recover, again if everything is designed properly this would be expected performance. Even if you put the cast iron back, if you do out door reset, you would see something similar. Though 10 hours does seem a bit excessive.

All that said if it was me, I would have never removed the cast iron, but you are going to pay to have it reinstalled.

Do you still have all the old radiators?

The boiler size is what it is, given the noise you are hearing and the poor performance, I would suggest the person you used to install isn't well versed in system design and probably not your best bet for any work moving forward.

Have you contacted the designer/installer about the problems and asked for resolution?

Jamie Hall

There is no good reason -- other than money -- why you can't restore the cast iron radiators. I don't suppose you still have them? Well... new ones can be had. And the result will be more comfortable. There is no reason, also, why the Bosch can't power the radiators. So that's not lost, either.

Also, you can run the radiators, as you are hoping, with home runs and zoned by floor. I myself am none to keen on PEX for hot water, but it seems to work well if properly supported.

What is going to take some thought is the control, piping, pumping and valving strategies to get everything to work smoothly together. Not that it can't be done, but it will take some good creative thinking by your installer.

jonny7117

Thanks for the thoughts. I'm going to reach out to Unico itself and see what they have to say. The installer, while well rated, hasn't suggested it's a problem other than to turn the air velocity down manually with this laptop. I'd like to see what an "independent" expert has to say. I've read they're very responsive, anyone have success stories?

Re: long time to come up to temperature, I understand that a mod/con boiler with outdoor reset will take a while (so no large setbacks). What I don't understand is the relationship between supply water temperature and btu output. Do they directly correlate? My potential issue is that at +35 degrees outside, the boiler puts out 160 degree SWT (the max setting for the hydroair coil) and still took 10 hours for 6 degree increase. So I can't see how the boiler is ever condensing (since heat pump takes over at >40 deg F), nor how the boiler will keep up when the temperature is in the single digits.

I don't have the radiators, but am trying to track them back down. I would have needed to sand blast and repaint them anyways, so I'm telling myself that refurbished CI won't have been much more than what I would have paid to clean up the originals in the first place. There's a few local shops that specialize in refurbished raidators.

With respect to PEX being "properly supported", could you explain what you mean? Do you mean physically, or something else to do with the system? Trying to educate myself, and this will help.

mattmia2

If you slow down the air velocity you will reduce the output at a given water temp.

You could run the heat pump when the outdoor temp is mild (maybe with a higher outdoor temp that switches to fossil fuel to reduce the drafty feelings) and run the radiators when it is too cold for the heat pump.

If the reset is right it should be matching the heat loss and the system should run continuously if it is a perfect match. i don't know anything about the unico system, I don't know if it does some sort of outdoor reset also. ideally they would both be reset but under one control so the velocity and water temp would both throttle down. If it isn't keeping up, the design water temp in the reset settings may be too low for the actual system conditions.

The 5:1 turndown along with settings on the boiler such as cut in/cut out differential on the supply water temp and time delay before re-start should be sufficient to prevent short cycling. You can re-calculate the heat loss at the warmest temps it will be operating and see how close that is to 20% of the output of the boiler. If you can't get the return water temps down to 130 or so, the output will likely be about 85% of the input anyhow.

The output is a function of the average water temp, inlet air temp, water flow rate, and air flow rate. A lot more heat will move from 180 degree water than from 120 degree water in to 60 degree air. Because the delta t of the air is a factor the relationship isn't linear related to water temp.

Most boilers have some sort of boost based on time that can be turned on that will boost the supply water temp if the call continues for some period of time to boost output for recovery from setback.

You probably will have to fix the design/install issues to get the unico to work well for ac anyhow.

Pex is fairly flexible, it needs to be supported something like every 2 feet to keep it from sagging and it expands and contracts a fair bit so it needs to be able to move.

Ironman

Here's a link to the thread about this on HVAC-talk:

https://hvac-talk.com/vbb/showthread.php?2213809-re-install-cast-iron-radiator-hydronic-system

jonny7117

Thanks, Mattmia2, I'm thinking the same thing regarding the heat pump: use it when spring/fall temperatures fluctuate between 50s and 60s. This heat pump will work to below zero, but definitely not worth it due to dimishing returns on energy efficiency, defrosting, and drafts as you point out. And totally agree on solving issues for AC -- noise will be a bigger issue for AC because Unico's minimum cfm is higher for AC vs. heat.
Question on observing short cycling: is there a way to visualize boiler cycles? The smart thermostat measures when the air handler is in use, not when the boiler is on or off. I suppose I can watch when the circulator runs, but even then I don't understand if that isn't also linked to the air handler and not the boiler. FWIW the air handler can make many calls per hour (5+) as indoor temp is satified and then drops again. I changed the thermostat to call again after dropping 1 degree (installer set it at 0.5 degrees to help with comfort). The intent was to get the system to "insist" on longer cycles. We have one zone on HW baseboard alone, so I can see when those calls are short (can be a few min sometimes). Not a problem when other zones are calling, but there's no way to ensure that's happening so some calls may be short cycles.
My concern about the system keeping up is that it appears that the boiler reaches maximum operating temperature for a Unico system (160 degrees) at +35 degree outdoor temp. The air velocity can't be turned down any more -- I think there's actually just two speeds. When it's in the low 20s outside, the system actually runs continuously because it appears the settings better match the house heat loss at that temperature, not because the system itself has better modulated. The added benefit is that it's not super loud then since both zones are calling and air flow is distributed among all 25 outlets, and not half of them. I worry about what happens when it's in the teens or single digits -- I'll just have to find out!
The boiler 5:1 turndown ratio probably matches a 40-50 degree outdoor temperature if the whole house heat loss is used, and some wiggle room applied for Slant/fin not being super accurate.

jonny7117

Completely separate question for the uneduated (i.e. me). Since the 100k boiler with a 5:1 turndown gives a min output of ~19k btus, and a 19k heat loss for the entire house occurs in the 40s outdoor temp, is the recommendation to only have a single zone? If that's the case, how would balance be achieved for the second floor radiators vs first floor? The second floor has CI piping (1 inch I.D., I believe) down to the basement, the first floor piping has been removed. If two zones can be done, how would balance be achieved then? Thanks for humoring me with your expertise.

mattmia2

One way to look at when the boiler is firing and not firing is to record the panel with a camera in video mode, there usually is an icon that indicates it is firing. Looking at the supply and return water temps while the t-stat is calling for heat and the air handlers are running will tell you a lot about what is happening.