Just placed the heat recovery portion of my system into operation today. It's circulating coolant water through a hydronic loop in the basement floor. Pics to follow soon.
One of the problems I'm running into is that I'm getting big gulps of hot water. On warm up the temp gauge was bouncing between 60f and 140f. I closed the pumped loop return valve way down and that helped quite a bit, but it was still bouncing between 100 and 130 after a 4 hour run.
There is a 180 deg thermostat, with a hole in it, in the engine and the system is plumbed to a 40 gal water heater tank. To get the hot water I tee'd into the tank inlet and outlet. The pump is a grundfos 3 speed set on the low setting that is on the hot side.
That's good to know. Now I'm hoping for bad weather so I can run it under a good load. I'm dumping the heat into a 1400 sq ft 4" thick concrete slab and only saw about 1 deg F temp rise.
Quote from: Jens on November 24, 2010, 12:08:24 AM
The most important point of hydronic heating is it's evenness, nothing happens fast and that is a good thing.
Sure....until your wife says "it's too hot in here" and you can't cool the house down ;D
Jens, being that your system is is liquid to air, you probably have excellent control of the temp. You can always add cooler air into the mix.
Tom, be careful of thermostat "overshoot" because the slab will carry a lot of heat inertia and a room that is too hot will be slow to reverse.
Looking forward to the pictures.
veggie
Aahh, but you can always open the windows. A warm floor and a nice breeze is delightful. :)
Well some over shoot would be interesting. That is why I chose to do the basement slab with the chp system, it has about 5 times the mass of the main floor (4" standard slab vs. 1.5" light weight concrete). This system will only run about 200 hours per year when the bats need charging, so the thought was to store as much heat as possible during the runs.
Eventually the plan is to add a couple of 4X10' flat plate solar collectors on the roof for main floor heat. Maybe next year. I'll see if some pics can be posted tomorrow.
The first 2 pictures are of the tank and pump connections and the second is the manifold connections to the basement and main floor slabs.
Tom
Where are you measuring that temperature swing at? I would say use a smaller pump or put a bypass/recirc line around the pump to really lower the flow, untill you get a more stable temp. As a reference, With 50F input water, a 6/1 is only going to give you about 3/4GPM of 100F water at full load... If you are pumping any more than that thru your thermal black hole, I mean 4" hydronic slab, you are taking away the heat faster than the engine can produce it, and the thermostat will gulp the entire time, never allowing the engine to get to optimum temperature. Well at least untill the slab stops absorbing all that heat...
Temp is being measured at the manifold. I don't think there are much smaller pumps so a bypass may be in order. The pump is a Grundfos ups15-58fc 1/25 hp unit. I'm thinking about setting up the pump to circulate the engine coolant and have the thermostat bypass to the hydronic loop. The thermostat housing on an IH 345 is an external piece that I have a spare of and I may be able to use it to accomplish the job.
Tom as Ronmar points out the pumps capacity, coolant in temperature and thermostats slow and over reaction habits is what is giving the temperature swings.
Back when time was cheap we used to function test a lot of thermostats with the boiling pot of water and a candy thermometer. They react very slowly to open and when they go to closed at room temp condition they compress a lot of spring (lots of lost motion)
Installing a pump bypass circuit with throttling would certainly be the way to go but it is not a simple snap of the finger to accomplish. A much smaller thermostat from a small outboard motor would likely meter much better than that "cornbinder" from the 345 :)
The problem with an outboard motor thermostat is they are low temp. Most of them are only rated to 62 deg. C.(143.6 deg. F). The reason they are low is any higher you risk having salt and other minerals percipitating on the engines cooling passages and blocking them and the end result will be overheating.
I have just the thermostat that would work nicely. I am working on cross referancing one since Wednesday. It is a turbocharger thermostat for a Porsche 944 Turbo. The thermostat starts opening at 83 deg. C. (181.4 deg. F.)
I am trying to find a 90-92 deg. C. thermostat though. Check the picture below. Oh, the thermostats flange is only 25mm in dia.
Henry
That is the type of thermostat I use on the secondary output of my heat exchanger. It is out of a mercury mariner outboard and is rated at 120F, which was perfect to feed my domestic hot water system. It is also the perfect size to plug right into the 1" NPT pipe nipple that comes out of the heatex. I just screw a 1" to 3/4" fitting over the top with a spring inside to hold the thermostat flange in place...
Tom
The more I think about your setup, the more i think a thermal regulating valve may be what you need.
1. IMO, you need the engine to be running at 195F at the cylinder head so it remains healthy.
2. IMO, 195F coolant is a little too hot to be feeding into the slab.
What a TRV would do for you would allow a higher flow, by mixing the 195F coolant with some of the colder coolant to maintain a higher flow of say 95F to the floor.
There is an interesting article over on the build-it-solar website than documents a thermal collection and storage system that feeds a hydronic system. One of the drawings shows a TRV used in such a way so regardless of what the storage tank temperature is, it mixes cold return water with it to get the proper temp to feed the floor. You will still have to adjust the flow so it dosn't take away the heat faster than the engine can produce it, but this would probably allow you to get a little higher steady volume to the floor instead of the gulps it is getting now...
Here is a link to the article. I found it very interesting reading.
http://www.builditsolar.com/Projects/SpaceHeating/SolarShed/Controls.htm
My strategy, to reduce the incoming water temp, is to mix it with the cooler water at the top of the tank by drawing it through the tee. It seems like it would work ok if warmer water was going in the engine. I'm trying to come up with something simple because the engine only runs about 200 hours per year, it will only save me about 1/5th of a cord of oak per year. Since we're now burning only about 2 cords per year, that is not a huge savings since the wood is already payed for.
The math to support this is 17000 btu available from the engine times 200 hours, divided by the 27m btu in a cord of oak.
Sounds like pretty good math if all the heat goes into the floor:) If it gets to a normal op temp, some heat is going to be lost thru the tank cooler... I think if I was trying what you are, I would put in a heat exchanger in place of the tank. I could then take what ever temperature I wanted off of the secondary side of the heatex(as long as it is less than the 195F op temp) either using a thermostat or by adjusting flow. If I didn't need the heat in the floor, I could run it thru a radiator first to dissipate the heat to the atmosphere. By using a "T" you are allowing the coolant, and the heat to go to both places, instead of into the slab.
Using my system as an example(look at my avitar), my engine under load hangs at right around 200F at the cylinder head which feeds the heatex. The flow thru the secondary heatex loop is modulated by a 120F outboard thermostat. At 1KW load, I get 1 quart per minute of 120F water. At 3KW load, I get 3 quarts per minute(with about 70F return water), but the engine temp hardly varies. That 120F water feeds the top of a storage tank, and the cold water from the bottom returns to the heatex. Eventually the tank botom water temp rises as the hot water replaces the cold water in the tank. The return water passes thru a fan-coil unit, and as the return water temp gets above ambient air temp, it starts to transfer heat to the atmosphere. But because I am sending all the heat to one place first, that place, the tank, gets first crack at all the heat. Now if I didn't need the heat in the tank, i could just as easilly send the heat into the fancoil first and dissipate it to the atmosphere, and the tank would only get water near air temp... I took the modular approach when setting up my system. My components are fitted with garden hose fittings. That combined with washing machine hoses allow me to re-configure the system in a matter of minutes with no special tools... The heatex allows me to maintain a constant 195F-200F temp at the engine, and easilly convert to 120F to feed the domestic storage. it also allows me to run corrosion inhibitors and antifreeze in a small volume in the engine.
Tank cooling does work, but it really only makes sense someplace where you don't have something more sophisticated like a radiator or heat exchanger available... In a cold climate, you are also going to have to deal with the evaporated moisture condensing on every cold surface it comes in contact with after leaving the top of the tank...
Does your slab have insulation underneath?
The original plan called for an insulated hot water tank to eliminate the heat loss. And since I use a fair amount of distilled water for the batteries the thought was to just allow the tank to boil into a copper coil to condense the steam. A new hot water heater is not in the cards, so I may scrounge one from the dumps. I tossed a magnesium rod in the tank to stop the corrosion and I've never even seen a skim of ice on the water tank in the cold of winter so anti-freeze has never been in the system.
That's a really tidy installtion Tom; although I always look at those multi-way manifolds with a shudder of fear - so many pipes.... so many opportunities for leaks.... still, I'm sure I'll end up with something similar one day.
How warm is the slab now?
I still think you need a heatex. By mixing the flow from the two systems in a hot water tank, they will sort in the tank by temperature, so the coldest water will still be feeding the engine from the bottom of the tank, causing it to gulp, and you will still be sending potentially too hot a fluid to the slab... You will also loose any secondary cooling that the open tank would have provided once the slab gets up to temp. The only way I can see to convert from a 195 engine temp to a 100+ slab temp is with a heatex or metering valve. IMO, you need something like this: If the slab no longer needs the heat, the radiator fan can come on to dissipate the heat to the atmosphere...
(http://i270.photobucket.com/albums/jj85/rmarlett/th_listeroidtank3.jpg) (http://s270.photobucket.com/albums/jj85/rmarlett/?action=view¤t=listeroidtank3.jpg)
Wow I've got a warm, ok warm spots in the basement slab!!! We're at the tail end of a long hard run today and I can feel the warmth in the slab. The temp cycles have really settled down and we are running around 110f into the slab and 80f out.
These temp gauges are really handy for diagnosing the running condition of the engine too. The earlier today the incoming temp was up to 160f at times and we were blowing black smoke, pulled the injector and found a nice chunk of carbon on the tip. Cleaned it off and the exhaust cleaned right up.
Quote from: Tom on December 04, 2010, 07:18:20 PM
.... I can feel the warmth in the slab.
.....pulled the injector and found a nice chunk of carbon on the tip. Cleaned it off and the exhaust cleaned right up.
Nice.
What were you burning to get carboned up ?
The waste hydraulic oil does it. Only have about 5 gal left and then it is on to an almost unlimited supply of WMO.