New boiler?

G Twitchell

I know that over the past 10 to 15 years or more, a lot of radiant floor systems have been put in using tankless water heaters as a heat source for the system and I have installed at least 6 such systems using Takagi TKD 20 180 kbtu units as a heat source and I did my personal unit first and it failed the heat exchanger just after the cold weather was over this spring. In disecting the unit, I found the heat exchanger was leaking near the edge of the lower stage 1 burner, which I concluded that this is the burner which is burning almost continuously during the heating season for the last 8 years. The Engineer in me has also concluded that this is the wrong heat source and that this type of installation needs to have a full condensing boiler that will handle higher flowrate and has staged (turn down ratio). The Takagi units have the staged burners, but I thins that the heat exchanger being snigle stage and copper is not the right configuration for this application. Takagi has a newer design (still a tankless water heater, not boiler) that has 2 stage heat exchanger (2nd stage is Stainless steel) but the primary is still copper. By doing the 2 stage it is now higher eff and condensing according to their data sheets.

Am I thinking along the right lines about boiler vs tankless water heater??



Any thoughts and comments?? Thanks:

George Twitchell, Laramie, WY

Snowmelt

Your kinda right

Your right about the tankless used for radiant floor. That we can do, the hex on the tk-20, is a copper hex, the newer ones have a different copper on the primary then the tk-20.



I have a unit doing 3 loops of onex tubing in one room and a master bedroom with commercial base board. This is a second heat source, the house has forced hot air.



Here are a list of questions I have.



What size pump do you have?

What size piping?

What is your gpm

Do you have an air separator on the system?

Ever fresh the unit.

Have you ever descaled the system

How long did it last?



I don't install but I fix them all the time...



Your right about a boiler being a boiler and a water heater boing a water heater. You can use a tank water heater also for low temp application.



Boilers came out with combo units.



Also NAvien has 2 stainless steel hex on there unit.

G Twitchell

New boiler?

Snowmelt, My system is currently :



1" manifolds (copper) to 3/4 feeders to 1/2" Pex in floor loops (no more that 220' long)

Spirovent air seperator

Main pump Taco 011 series on the supply manifold and each of the 5 zones have their own Taco 007 series and the large pump comes on when any one or more Sones come on demanding heat.

When all zones are on, the flowrate is 4.8 to 5 gpm

Flushed system once or twice in 8 years and drained water was clean and had no particulates. The Takagi has an input filter and I clean that routinely when I see the flowrate go down slightly.

Looking at the inside of the heat exchanger as I have disected it, there is little to no scale as I would expect since this is a closed system.

The Takagi 2kd 20 has lasted 7 3/4 years.

I have pictures of the bottom of the exchanger that I can post if I can figure that out (being computer challanged as I am)



George Twitchell

Snowmelt

No need for pic

I have seen about 50 leaking heat exchangers, getting 8 years for heating isn't really that bad.

The copper heat exchanger is going to eventually give up on you.



I'm going to assume your not going above 140 degrees. I would just update and put a tk jr or a tk 4, there cheap enough, and you have the ability to swap it out quickly.



The tk-20 has water and gas on the side, now everything is on bottom. So it involves a little bit of plumbing but I think now everything would be on bottom.



Your question was copper the correct material to use for this application, answer is no. But it's the cheapest that's why everyone is doing it. (Copper hex for tankless heating application.)

Harvey Ramer

Problem

The single biggest problem is Tankless Water heaters (the whole lot of them) are designed to be run with a wide temperature difference between the supply and the return. That is the nature of DHW heating. It comes in cold and goes out hot, normally 70° difference. It takes a very large burner to accomplish this. 3-4 times the size required by a (for example) 2,000 sf. single family home.



The nature of a typical heating system is a lot less difference in the supply and return water temp. Normally, american systems are designed anywhere from a 10° to 20° delta-t. In real life operation they often drop below this design. A typical heating system requires more flow than the heat exchanger in these units is designed to deliver. A typical pump rated at the flow required for the heating system, will only produce a big enough pressure differential to get a pathetic trickle going through the heat exchanger of one of these units. Remember, in DHW you will typically have around 60 psi pushing on the one side of the heat exchanger and an open faucet on the other side. That is what they are built for.



If you put one of these units on a heating system, one could expect a multitude of failures. The one you experienced no doubt came from rapid heating and cooling of the heat exchanger due to a grossly oversized burner and not enough flow. Thermal stress is a killer to almost any given heat exchanger. A modulating unit of this same pedigree, might give a little bit better performance on a heating system, but I wouldn't expect much! After all, the water heater's built in temp sensor is not in the heat exchanger but directly after. I would expect burner surging and the same thermal stress occurring on the heat exchanger. This burner on a heating system is like a cat playing with a mouse. It's just too big.



I'm sure someone will read this and say, I have one and it heats my house just fine.

I'm not arguing that point, I'm trying to give you some insight on why it will break.



Harvey

Snowmelt

The reason

Why it leaked.



There's a few reasons and there simple reason



The flow of water erodes the walls of the copper.



If any air bubbles get trapped in the hex, same thing happens, it erodes.



The third would be if the flue gases condensate back on top of hex fines, eventually the acid will eat the hex.



There is one more that they tell me, if the machine goes on & off a zillion times it will expand and contract the copper which will eventually give at a joint,



Do you have the usage time vs the on off cycles?