I have a simple radiant heating system in my basement.  It has been there since I bought the house.  It used to work much better.  I can't figure it out. I replaced the hot water heater due to CO2 issues in the old one.  I should have rebuilt the gas burner, but decided cost wise to just swamp it out.  They were both 40 gallon, but the old one seemed to work much better?  Perhaps it was turned up or tuned to a higher btu, but one was not stamped as such on the side of the heater.   

What I have:  A closed system.  40 gallon / 38,000 btu new water heater.  Grunfos (old /original 1991) small pump pumping to approx 750 sq ft. basement.  Water goes out at 160F and returns at about 95F from the system loop.  I has been very cold here lately, but has always seemed to lag over the old system.  If I click the thermostat off, the unit fully recovers to 160F in about 18 minutes.  I have also found if I let it run 10 minutes, shut it down and let it recover for about 15 min, then let it run another 10 minutes, I can get the return water to about 105F.  

Am I just underpowered with the 38K btu system?  We've seen lows the last week in the -5F range that does not help, but it drives me crazy to see that hot water heater running for 2 to 3 hours at a time to keep it at 65F (what we set it at).  

Any thoughts or suggestions?  Thanks.

I assume it was CO and not CO2 issues? If water water heaters are used as dedicated space heating boilers (not accepted under any model code) special care must be taken to avoid the issues you are having, not the least of which is sustained steady state flue condensation. The result of running atmospheric or power-vented domestic water heaters with lower return water temperatures will almost always be deterioration of the flue and burner with predictable degradation of combustion efficiency and as the manufactures would say "serious risk of personal injury or death".

Tank/type water heaters are certified with a single output listed on the rate plate. It is only the target water temperature that can be adjusted in the field. One should no the load presented by the structure and the output of the intended heat source before buying new equipment.

Your current setup is using more fuel than it should, failing to keep you comfortable and surely headed for early failure, normally not covered under warranty.

No one has warned or told me about the issues you mentioned until now. It was installed by a professional plumber and also given passing by County inspector (not saying much).

What would be the ideal situation? Electric boiler? I cannot do tankless water heater as I don't have an outside wall to vent the super hot vent and the flue will not work.

Ideas anyone?

Sounds under powered (based on inability to maintain tank temp) although it might eventually catch up if you leave it on.

A condensing gas boiler, sized by an expert is probably the ideal fix. That same expert should be able to tell you if you need more insulation.

it seems pretty hard to think a 38kBTU water heater would be unable to heat 750 sq ft of basement unless it's basically uninsulated. startup might be tough in colder weather.

1. you should not be pumping 160 degree water into the slab. bad idea. 130 max, I would recommend. Maybe you have a mixing valve? if not, turn down the tank temp. that has no effect on output in this case. will help your efficiency too.

2. You have a huge temperature drop. either you're using a mixing valve, or the slab is damn cold and you are just dumping tons of heat into it. wait for it to heat up the first time before freaking out, and stop turning it off. maintain your temp. ONCE YOU ARE UP TO TEMP, observe the water heater. it should cycle normally and not plummet in temperature. if so, you should be fine.

You must control the return water temperature as you would with any gravity vented gas appliance. Tankless water heaters are rarely the answer and electric boilers usually cost three times as much to operate...except where I live.

Before starting any radiant floor heating a proper Manual 'J' heat load should be performed and appropriately sized equipment specified to meet the predetermined load.

38,000 BTU should be more than adequate heat for even a drafty (1.0 ACPH), uninsulated 750 SF basement. However, as others have said, a water heater is not the best way to provide the heat. Not sure why venting a gas boiler is not possible?

What is the pumping rate? If you are even pumping 1 gpm with a return temp of 95F, then you are dumping 31K BTU somewhere, which might be more than this heater can deliver.

Do you know whether your slab is insulated or not?

I don't have an outside wall to vent the super hot vent

Some are so low temp that they use PVC for the vent.

I will have to check.  I lowered the temp after reading a previous post that said NOT to pump 160F water into the slab.  Rookie / unknowing mistake.  I turned it down to 135F.  The temps are up from -5F to about 10F this morning.  Seems to be much easier to get up and maintain temperature.  My slab is insulated, but my windows are 20 yr old alum and not great.  They get replaced next year.  When it is -5 out, those suckers even get frost on the "inside" of them and you can feel the cold coming through.  I doing a test this morning.  It is 13F out and I'm trying to get the basement up by 2 full deg F.  I'm going to see how well the water heater can heat the circulated water.  I'll let you know and thanks for your help!!!

The system has been running 1/2 hour now. I've put a thermometer on the pipe coming out of the bottom of the water heater and the outgoing temp of the water after running is about 91F. The water coming back is not much lower at about 87F. The question is, how long will this system take at that rate to get the room up to 64F air temp? I'm thinking the water heater will increase the outgoing temp a bit as time goes by...I hope? I'll let you know. The pump is an old Grundfos UP26-99F. I don't know how many gallons per minute it pushes is this closed gravity fed system?

Update for diagnosis: After 2 hours the air temp has gone up 2 deg F. The output temp from the WH is 94 now and the return is at about 90. So, the slab and also the returning water is heating up / but slowly. Pretty darn slow if you ask me.

Thoughts: If I had a boiler or gas powered tankless (that could vent in my current venting or pvc) that pumped water at a continuous125-130, wouldn't it heat things up much quicker and save time and energy use?

Please comment and help if you can.

Your pump is to large you are over pumping. Check the flow chart. Grundfos UP26-99F at about 7.75 feet of head will push 20 GPM. A 300 foot loop of 1/2" pex pipe represents about 7.75 feet of head.
Assuming you have on the out side 3 300' loops of pipe in your 750 square foot basement you would need to be moving considerably less water for those little heat molecules to jump off in to your slab. About 3 gpm or less.
The best solution is to replace the pump with something like a Taco 005 or better yet the new Taco Bumble bee, a delta T pump. The Bumble bee is your best choice as it will ramp way down on flow to hold a delta T setting. We are using these now on all our design work as the pump is smart, it adapts to the building needs regarding water flow.
Alternately you can throttle back your Grundfos to get the return water temp in the 15 degree differential range (100 out 85 back)
But you will really need to choke back on that pump, you will hear the water constriction on the valve and know you are about right.
Dan

you would need to be moving considerably less water for those little heat molecules to jump off in to your slab

Reducing the flow rate is not going to make the entire slab heat up any faster (but it will save some pumping cost). Let us know where you are at after 48 hours of operation.

If you want fast response with a cold slab, you need many more btu/hr.

Jonr,
You are sadly... wrong.
I have a hard time wrapping my head around the concept as well, but reality is slower moving water will transfer more heat.
Pumping faster will distribute less heat, 20 gpm or 6.5 gpm per loop more or less is way over pumped. The heat transfer is compromised.
Dan

Prove it. Perhaps start with turning off the valve completely :-).

Delta T
The amount of heat released as the water moves by, the faster the water moves the less delta T.
There are others on this blog that have a better handle on the physics of how it works. As I said I have a hard time wrapping my head around it, but the reality is that moving water faster is not the solution. It is counter intuitive but slowing the water down will release more heat. That's why I have become a big fan of the new Delta T pumps, they help with the proper pump velocity as the building needs change. Making my job easier, I hit the mark on water flow based on delta T letting the pump do the math. We ( installers, designers, distributors) are actually lucky to have this new line up of pumps, our systems can be much more fuel and energy efficient. For instance the Taco Bumble Be is an ECM motor magnets drive it, it will crawl if needed and suck little juice or ramp up to close to a 008 pump curve.
C Butlers numbers reflect the concept, out at 91F back at 87F you are only dumping 4 degrees in to the slab. Like I said, I am a simple person and the concept confounds me but it is of a truth.
As was explained to me fast moving water is like a fast moving train through the station, no one gets on or off.
Perhaps Dana, Rob or Morgan can better explane why this is so,
Dan

Your are correct that moving water faster is not the solution, but this not the place to design DIY radiant floors either, since no such place exists. The pump is grossly over-sized. still guessing about heat load and now running the water heater at a temperature that nearly insures short life and CO producing conditions in a matter of a few short years. Perhaps the tank will start leaking before anyone gets hurt.

There is no fossil-fuel boiler available in the US with a low enough output to properly serve what I can guess is the real heat load for this project. Like most radiant floor problems, it is not the equipment but the designer, or lack thereof. There are three negative aspects of over-sizing pumps for hydronic systems. First, the initial cost, second, the increased cost of operation and finally the risk of severe erosion as the pump takes inner surface off pipe and system components.

Ahh,
The fourth element, positive and perhaps most important rule and one that might extend the appliance life, Delta T.
Holding to the delta T = Slow water, better heat exchange. Then the other three negative aspects will diminish, even the appliance will hold higher temperatures possibly extending it miss placed life.
Dan

There are three negative aspects of over-sizing pumps for hydronic systems. First, the initial cost, second, the increased cost of operation and finally the risk of severe erosion as the pump takes inner surface off pipe and system components.

Agreed. Lower heat transfer isn't one of them.