gravity conversion

DaveGateway

I'm doing a gravity to circulation job at home. I read in Dan's gravity heating Q/A section (on this site)that an injection system would be a good way to convert this. The system has 2 supplys and 2 returns just like your gravity paper showed. the article said to tie the 2 supplies together and the 2 returns together to form a large loop. Then put a pump in the loop for constant circulation. An injection pump would supply the loop with hot water when necessary. This makes sense to me. I would also put TRV's at the radiators (11 of them) $$$. I bought Dan's primary/secondary pumping book and thought it was very informative and a good resource. But I still have a couple of questions. The pump in the large radiator loop will have differing head pressures depending on the state of the TRV's. How do I compensate for that? I've seen that grundfoss makes a pump that automatically adapts itself for changes in pressure (Alpha pumps). Or I've seen that I could put a device in called a diffential pressure control. What is the best way to solve the head pressure issue? Also if I have TRV's in each room and I would like to use a Tekmar reset control (#364) to manage the system - Theoretically I would not need a room T-stat because of the TRV's since they are mechanical. I do need an indoor air temp sensor for the control unit and injection system so do I locate the sensor for the control unit in a critical location in the house? How do you integrate TRV's with a controller using injection pumps - that's my basic question. The whole house would still be using the old piping. I just cant figure out which way to solve the gain in head when the TRV's are all closed. This would cause the system loop to be blocked. Also when the TRV's are open what makes the heat go up the pipes instead of going through the main loop. My laterals to the radiators have no valves in the basement - only the hand valve at the radiator. One more thing, I have a domestic coil in my boiler that I will keep on using. How will that work with the Tekmar unit? All the drawings i've seen show an indirect water heater and i'm not sure I want to spend for one of them yet if I have a coil that still is fuctional. Also on my system i would take the 2 large radiators in the sunroom off the main loop and zone them with a stat and circulator. The rest of the house would be on the gravity conversion setup.Hopefully this wasn't too long.
Thanks,
Matt

P.S. I've been an industrial electrician for 14 years and since I started studying up to do this project at home I have been fascinated (almost addicted) to finding more info and designing this system. If I can get this to work ,(and I will), it will be very gratifying

Al Letellier

gravity conversion

Matt, you've got a project on your hands, but congratulations on your research and using this site.
You can set a Tekmar for constant heat call and don't need a stat in the house if using TRV's. Use a pressure by-pass and keep things simple. The pressure drop will cause flow to the radiators as the TRVs open up. Your hot water is the tricky part and using a tankless may prove more trouble than it's worth. The reset control will keep boiler temps too low on mild days to make hot water and keeping the boiler hot will require a mixing valve to keep system temps low for energy savings. Your best bet, in my humble opinion is to use an indirect, use constant circulation on the system, and let the TRVs control room temp. You'll get hotter water in the mains when making domestic, but the TRVs will prevent overshoot.
Good luck.

To Learn More About This Contractor, Click Here to Visit Their Ad in "Find A Contractor"

DaveGateway

Thanks for the response Al. By pressure bypass do you mean a differential pressure control valve? Also how do you size the pressure bypass valve and where would that go in the system? Concerning the DHW, I do have access to a new 30 gallon weil mclain indirect tank for just under $500. I'm not sure if that would be large enough for my house. I have 1 shower,washer, kithen sink, bathroom sink and 1/2 bath sink. Is there a rule of thumb for sizing indirect DHW tanks?
So you think the outdoor reset,injection pumping setup would work for my old gravity system? I'm about ready to start ordering components and I want to get these last bugs worked out. I've settled on a Tekmar 364 controller,Bell and Gossett 100 pump for the secondary loop,B&G nrf 22 for the boiler loop,Taco 007 for the sunroom,and I'm not sure what pump to use for the injection pump. I got prices for the TRV's (5-3/4",and 6-1") Danfoss-around $1100 total including the heads. Do the TRV's have to be sized in anyway other than the input pipe diameter and are they reliable? If I decided to not go with the TRV's for cost and put 1 room sensor in for the whole house I guess that would be an option too. I could still use the manual valves to throttle the rooms and the pressure bypass would take care of the rest, right?
Everything I've seen says to 1/2 the main pipe size (2-1/2") and drop 1 size from there. I'd need to put in 1" to the large loop - sure seems small.
Thanks again,
Matt

DaveGateway

Thanks for the response Al. By pressure bypass do you mean a differential pressure control valve? Also how do you size the pressure bypass valve and where would that go in the system? Concerning the DHW, I do have access to a new 30 gallon weil mclain indirect tank for just under $500. I'm not sure if that would be large enough for my house. I have 1 shower,washer, kithen sink, bathroom sink and 1/2 bath sink. Is there a rule of thumb for sizing indirect DHW tanks? So you think the outdoor reset,injection pumping setup would work for my old gravity system? I'm about ready to start ordering components and I want to get these last bugs worked out. I've settled on a Tekmar 364 controller,Bell and Gossett 100 pump for the secondary loop,B&G nrf 22 for the boiler loop,Taco 007 for the sunroom,and I'm not sure what pump to use for the injection pump. I got prices for the TRV's (5-3/4",and 6-1") Danfoss-around $1100 total including the heads. Do the TRV's have to be sized in anyway other than the input pipe diameter and are they reliable? If I decided to not go with the TRV's for cost and put 1 room sensor in for the whole house I guess that would be an option too. I could still use the manual valves to throttle the rooms and the pressure bypass would take care of the rest, right? Everything I've seen says to 1/2 the main pipe size (2-1/2") and drop 1 size from there. I'd need to put in 1" to the large loop - sure seems small. Thanks again, Matt

Mike T., Swampeast MO

Differential pressure bypass: generally you would use the same size as the SIZED DOWN mains, 1" in your case. Danfoss and Honeywell (under the Braukman name) are two mfgrs I know of. On a simple system without primary/secondary it is installed between supply/return with the circulator inside the "loop" it forms. I presume it would be installed similarly in the radiation loop "around" the circulator. As I've only seen one (mine) and it's the only one in town my experience is limited however. Differential pressure seems EXTREMELY low in a gravity conversion--I've NEVER been able to see an indication on the integral gauge and had to adjust mine by ear. Since I'm not using primary/secondary it's adjusted for some bypass continually for best fuel efficiency--slight changes have an amazing amount of effect on boiler cycling.

Can't offer anything regarding indirects as they are extremely uncommon around here--I suppose there are some in commercial use or apartment houses but I haven't seen one yet. I do see an occasional tankless coil--LONG since abandoned.

TRVs: I'd use the current size of the pipes as the sizing guide as they will be a perfect fit. You could likely use smaller, but the aggravation of changing the connections wouldn't be worth the small reduction in valve body price.

They seem extremely reliable--mine are about six years old and the only problem was one that clogged and operating the valve repeatedly cleared the problem. I later found (when re-working some of the main piping) that someone had used two sections of galvanized pipe in some previous repair/relocation of that radiator. The galvanized pipes were nearly completely obstructed (about a pencil-width hole left in a 1" pipe) while ALL of the black iron displays little more than surface pitting/corrosion after 81 years. I see them occasionally around here in old apartment houses and the older Catholic schools/churches and they seem to work reliably for decades.

Injection: Haven't seen it on a gravity conversion but of course that doesn't mean much... I do believe it would be a decent way to keep the boiler loop temp high with a conventional boiler (where it's most efficient) without the use of a large buffer. Never forget though that the capacity of most gravity systems to liberate heat GREATLY exceeds the capacity of the boiler to produce heat. Also remember that the radiation loop will hold MANY times the amount of water and have MANY times the mass of the boiler loop. Opening a single TRV on a set-back radiator has a rapid and amazingly large effect on return temperature and I'm not exactly certain how injection deals with the situation.

TRVs and balance: TRVs create a dynamic situation--HOW dynamic is quite easily controlled with reset. Reset adds stabilization. The output of standing iron is quite linear--with an appropriate reset "curve" the reset will be reducing the BTUs available to be liberated by the radiation at the same time that the BTUs required by the space are reduced as well. Because of this, the amount of flow through the radiation will remain surprisingly constant in most conditions. Before you get the idea of establishing a "perfect" reset curve that supplies just the amount of heat needed at any time remember that such creates a condition with very little headroom for raising space temperature and such response will be extremely sluggish. A reset curve set about 20° above the average requirement (assuming 20° delta t) OF THE SPACE REQUIRING THE HIGHEST SUPPLY TEMP seems a very good compromise. For example: if space (a) needs 135° water at outdoor design temp set your reset curve to about 155° (10° for half of the delta t, and 10° for headroom).

To find your supply temp requirement: Reduce Manual "J"-based calculations SPACE-BY-SPACE by 15%*. Each square foot of EDR of the radiatior will give of 1.5 btu/hr for each degree of temperature difference between ambient and temp of the radiator. Using that formula find the average surface temp requirement for each space and use the HIGHEST requirement for your reset "curve." If you do this for 10° changes in outside temperature and put everything into a spreadsheet you can calculate your ideal reset ratio as well--just use linear equations and averages. If you're the curious type start playing with the spreadsheet. Balance your radiation perfectly; add LOTS of EDR to curious effect on reset ratio, etc. You'll have to take my word for it that everything that matters IS linear--the only significant deviation is at the low extreme and is the very reason you need differential pressure bypass. There is deviation at high temps as well (approaching steam temps) but your system won't need these temperatures to begin with!

The table I have attached is accurate enough to truly predict the temperature of radiators with amazing accuracy. So accurate in fact that I can actually see (and predict) the effects of sun and wind!

Sizing down the mains: I know it looks strange, but it works. Just take your total heat loss and compare to maximum flow tables and you'll most likely see that there won't be a problem.

* That 85% number was from an EXTREMELY thorough HVAC-Calc calculation using "average" infiltration. I arrived at the reduction by using ACTUAL radiator temps with the TRVs maintaining temp in unoccupied space at known, low and steady outdoor temps on quite still nights. By "old house" standards my house has little infiltration, but I'm sure it's quite "loose" with some modern constructions. I'd adjust it a bit based on how well your windows/doors control infiltration--mine are RESTORED originals with modern storms, spring bronze weatherstripping in the jambs and pile weatherstrip against the lower sash. The weight pockets insulated with 1" Styrofoam and caulked as best as I could manage in such a tight area. The casement windows have original (and amazingly effective) interlocking weatherstripping.

The "required H20 temp" section of the table is NOT adjusted as stated above, but what was used for reset ratio calculations.

DaveGateway

Gravity conversion

Thanks for the reply. I found the Honeywell diff. bypass valve and they aren't that expensive. Did the EDR calcs. on my radiators and the load comes out to around 86000 BTU/H. A 1" pipe should in deed be big enough. I am working on the head pressure calcs. now. When I get done with them I will size the circulator for the secondary loop. I think I am going to buy an indirect water heater for my domestic. It should be easy to pipe in. I also have a wood/coal boiler that I will make provisions for when I pipe my primary loop.

Al Letellier

matt

Looks like someone beat me to your answers, but that's OK. that's what we're all here for. I spent the day investigating an insurance claim 2 hours away and got to enjoy my wife's company for the ride...don't get that much quality time very often!!!
As to the indirect, the 30 gal will work just fine for your load. Run it at 140 degrees and use a good quality tempering valve and you should have plenty of hot water. I have a +30 and never run out...good luck with your project

DaveGateway

Al,
Can I run the tank at 120 degrees and forget the tempering valve or is that not possible because of the smaller tank size? Are you running the 30 gallon tank at 140 and tempering it to compensate for the smaller tank? Do I need a vacuum breaker for the dhw supply if I have a backflow preventer on it?
Thanks
Matt

Mark J Strawcutter

TRV sizing

I agree about using the same size TRV as the original valves. Not only does this make the physical changeover easier, but the TRVs have a _much_ lower cv than the originals and you probably don't want to lower it even more by downsizing.

Mark