The check valve in your system with one zone and no stored heat is unnecessary.

I went with a Taco 009 - more of a high head/lower flow model.

I just took some readings without the restrictions. I'm getting 117 degf out and 81 degf return. That's a delta t of 36 which is better than the 50 plus. Still need a better pump.

Before you spend $ on a pump try turn your mixing valve down to 90F. This should-

-Increase flow  thru the loops
-Decrease flow thru the tankless
-Decrease deltaT across your loops
-increase deltaT across tankless
-Even out slab temp swings with longer circ run times
-Decrease tankless on times
-Increase comfort and efficiency

90*f is not a magic number just a starting point to fine tune your system. Observe DeltaT's across the loops and across the tankless. This will give you an idea of your radiant system flow and tankless flow, two different things with your mixing valve setup. When you report back make sure you differentiate the two and keep track of tankless on/off behavior.

You have already proven the tankless has the capacity by maintaining temp. I assume your tankless will try to maintain a nonadjustable 120F and turn it self off at lower flow rates as temps increase. You will want to avoid short cycling so as not to wear out the contactors. This can be accomplished by increasing flow across the tankless (by either increasing mixing valve temp or introducing the proper sized pump).

Let us know.

there is no way you should have 18' of head in the loops you talked about. you mean through the heater?

I'll try adjusting the mixing valve and see what happens.

The 18' is the head loss number from the Taco pdf.

you should post your math. I'll repeat there is no way you need 18' of head for the loops you described. I don't know what the pressure drop is across your heater, that might change it, but the loops themselves don't need that for sure.

a 250' 1/2" loop would need at least 1.5 GPM of flow to be at 18 ft of head. that's 15kBTUs/hr PER LOOP at a 20 degree dt. Do you really need that much heat?

Short answer, no. The system is heating just fine the way it is right now. I only need 20k btus to heat the whole building. So I only need 5k per loop. I'm trying to heat everything with one loop!

I used this formula from the Taco pump selection guide;

The formula for estimating the head loss for piping circuits constructed of smooth tubing such as copper, PEX, or PEXAL-
PEX is:
HL = k x c x L x (f1.75) k=.0374, c=1.095, L=230, fl1.75= 2

I think maybe my hl should be closer to 2.8

as I thought. 2 GPM per loop is a ridiculously high flow rate. if you're shooting for a 20 degree dt, that's 20,000 BTUs/hr per loop, not realistic.

flow = load/(dT x 500) for natural water, approximately.

I agree. If you just want a more even temperature in your loops, then let some of the flow bypass the heater. If you need more output from the heater (because it reaches max temp and cycles off), then you need to force more flow through the heater with a better pump. If the heating is sufficient and even enough, then I wouldn't change anything.

I adjusted the mixing valve to 90 deg. The loops jumped up to around .4 gpm and the output temp on the heater jumped up to 120. I'll run it like this and see. My only fear is the heater will never shut off, but run longer, we will see.

could the mixing valve have been in backwards? Or could the "cold in" leg on the mix valve have a closed valve or obstruction of some kind?

that would have choked flow off to your zones.

Otherwise, the mixing valve action should not affect your flow rates, only your temperature.

All other components being the same in this system the thermostatic mixing valve has absolute control over flow thru loops and thru tankless proportionally.

If you set the temp at the mixing valve to max system temp (in this case 120F) the hot side of the valve (from the tankless) is wide open and the cold side (from radiant return) is closing up tight. In this case most of the flow is trying to pass thru your biggest restriction (the high head tankless). This increases the total system head and decreases system flow.

When you turn the mixing valve down it opens up the cold side (also closing down the hot side) and allows return water to bypass the tankless increasing radiant side flow (by lowering head).

What I am not sure of is what your final system head is until the head of the tankless is figured and what percentage flow rate you want thru it.

The tankless controls are not designed for this application. You will have to find out if it works for you. If it is controlled by temp only it should be able to turn it self off above 120F. I have to assume it will try to maintain 120F. Your flow will determine if it can stay there. If it is way overpowered try disabling one of the elements. This will also help short cycling.

even full open the flow should have been higher with that pump than less than 1 GPM, I think. and every time the valve closed a (heater water temp exceeds his valve setting) and he was pumping only through his loops and the valve, he should easily have had 1 GPM+ per loop.

we're not hearing anything about a variable flow like that. but if he gets the full flow when the valve is in "full recirc".. say, set to 90 and the water heater is putting out 125... but not when the valve is open, then we know it's the tankless head loss and pump sizing.

but if it's a consistent problem, I'll wager the valve's in backwards, and only cracking to let the tiniest bit of flow through. turning up the valve would allow a little more flow through.

Time will tell. It will take the high mass floor maybe many hours or more to even out temps until you can get a true deltaT across the radiant loops.

I do agree, something is off with his initial headloss calcs across the radiant loops. Again it will depend on flow rates.

I just checked the valve, heater output is connected to the side marked H, the retrun is connected to the side marked C, and the mix side is piped into the suction side of the pump.

The heater has a 1/2" inlet that goes to a 90 fitting into the top side on the first chamber, then turns 90 at the bottom side thru 1/2" pipe into the bottom side of the second chamber, then out of the top side and 90s into 1/2" pipe down to the outlet. This is where I removed the .175" restrictor. Nothing in the heater for varing the flow.

if you crank the tempering valve nearly all the way to the lowest temp... such as 90... with the heater running, does your flow increase?

It's not clear why you have a mixing valve instead of a simple heater bypass valve.

The mixing valve was installed in case this heater did not work and I had to add a bigger one. I was told at one point that I never wanted more than 120 degf supply into the slab. I was also told to leave it out, but I had it, so it went in. The heater I have is supose to supply 125, but i have my doubts.

Yes, it went from almost no movement on the flow meters to .4 gpm