Two temperature boilers plumbing circulator location

Kkane97

I currently have a crown non condensing boiler, two zones both baseboard through my house. Iam wanting to switch one zone to radiant flooring. Doing so I am concerned about return water temperature being to low and need to add a mixing system in to regulate that and lower the temperature loop for the radiant section. I have currently 2 circ pumps in the return sides of my system, looking into plumbing designs I found one that I think will satisfy my needs and be the most simplistic to plumb in my system only using two three way mixing valves, one for the low temperature radiant and the other for regulating return water temperature. Of this design just omitting the second loops mixing valve and continuing high temperature loop. But all the diagrams I am seeing have circulator pumps on the supply sides of the radiant loops, in my application for adapting my current system it would be much simpler to me to leave them in the return side. Is there a reason I commonly see circulator pumps on supply side? 

Jamie Hall

Provided you have independent pumps powering each circuit, yes you can either have the pumps on the supply (after the mixing valve) or the return, before the mixing valve.

However. Doing that you wind up with both pumps being, from the standpoint of water circulation, a long way from your expansion tank and thus being at the lowest pressure point of the loop, rather than at the constant pressure point of the expansion tank. Not good. Not good at all. Unless the overall system static pressure is enough higher to keep good pressure on those pump inlets there is a very real risk that they won't have enough intake pressure at least from time to time -- and cavitate, which destroys pumps.

hot_rod

Do you know the heat load of the two zones you are considering?
What about boiler size?
One rule of thumb is if the radiant is 10- 15% of the boiler output, a return protection valve probablyis not essential. The reason being the boiler can catch up top that load rather quickly.

One mixing valve for the low temperature is all you need. The other loop runs at boiler temperature.

Is the boiler on outdoor reset control?
If so and the radiant is only one zone then a manual 3 way mix valve is a simple option. its temperature output tracks along with the boiler temperature. based on outdoor reset.

A thermostatic 3 way is another option, but it stays at a fixed temperature.

Kkane97

Jamie Hall said:

Provided you have independent pumps powering each circuit, yes you can either have the pumps on the supply (after the mixing valve) or the return, before the mixing valve. However. Doing that you wind up with both pumps being, from the standpoint of water circulation, a long way from your expansion tank and thus being at the lowest pressure point of the loop, rather than at the constant pressure point of the expansion tank. Not good. Not good at all. Unless the overall system static pressure is enough higher to keep good pressure on those pump inlets there is a very real risk that they won't have enough intake pressure at least from time to time -- and cavitate, which destroys pumps.

That makes sense being that the pumps are pulling instead of pushing the radiant loops restriction of pipe size would also be a factor in causing cavitation right?

Kkane97

hot_rod said:

Do you know the heat load of the two zones you are considering? What about boiler size? One rule of thumb is if the radiant is 10- 15% of the boiler output, a return protection valve probablyis not essential. The reason being the boiler can catch up top that load rather quickly. One mixing valve for the low temperature is all you need. The other loop runs at boiler temperature. Is the boiler on outdoor reset control? If so and the radiant is only one zone then a manual 3 way mix valve is a simple option. its temperature output tracks along with the boiler temperature. based on outdoor reset. A thermostatic 3 way is another option, but it stays at a fixed temperature.

I don't know the exact heat load, but estimated high 250 square feet would be about 10,000 BTU boiler is 90,000 iirc

EdTheHeaterMan

Ask @hot_rod if this diagram is better.

it places the boiler circulator after the expansion tank and if you are lucky enough to have a built-in boiler air scoop, then adding a vent in the boiler will be an excellent place to vent dissolved air from the system. The hottest water and the lowest pressure in the system is where dissolved air will separate. If there is a vent location in the boiler casting, that is where the lowest pressure and the hottest temperature water is located.

Kkane97

@EdTheHeaterMan that's exactly what I was thinking as far as the diagram as far as the mixing valves thank you, currently my expansion tank and air vent and filler are on the outlet of the boiler before it tees off to either zone

EdTheHeaterMan

Are you going to use the SEP4 in your repipe? @hot_rod is the expert on that stuff and may have a better or less expensive idea for you. I think he will chimes in sometime tomorrow.

Which cast iron boiler are you using? Brand & model number. We can look up if you have the built in air scoop tapping.

hot_rod

That 10K radiant load should not cause a 90K boiler to condense, so no worry about return protection.

Same with copper fin tube, it is low mass and would not generally require boiler return protection.
If I were a betting man, I'd guess the boiler is also oversized to the total building load? only a room by room calc could answer that.

Here are three options using a 3 way mix valve. The first, direct pipe assumes the boiler can operate with 1 gpm flow. Most cast iron boilers can.

Primary secondary series loop costs one more circ.
The sep is another option.

EdTheHeaterMan

Lowest cost design. I would use it for one of my customers.

Kkane97

hot_rod said:

That 10K radiant load should not cause a 90K boiler to condense, so no worry about return protection. Same with copper fin tube, it is low mass and would not generally require boiler return protection. If I were a betting man, I'd guess the boiler is also oversized to the total building load? only a room by room calc could answer that. Here are three options using a 3 way mix valve. The first, direct pipe assumes the boiler can operate with 1 gpm flow. Most cast iron boilers can. Primary secondary series loop costs one more circ. The sep is another option.

It definitely is oversized I believe, 1300 square foot about 65 ft of baseboard is all it was doing

Kkane97

EdTheHeaterMan said:

Lowest cost design. I would use it for one of my customers.

On all these that use a mixing valve there is a extra circulator pump to control the flow of water through the valve when the valve is bypassing, what would you recommend as far as controlling that or should I just wire it to come on with either zone calling for heat through a relay?

Kkane97

In the future if im happy with the radiant and can provide sufficient heating as my primary source the other zone I would like to switch over to it as well as I redo parts of the house.  If this is the case would it be better to have a boiler protection loop?

hot_rod

So the 90K boiler could in fact be twice the size you need? There will be some short cycling of that sized boiler with a 10K or smaller load that the radiant demands.
A 50K condenser would be an ideal swap, someday. Especially if you go all low temperature radiant.

The direct pic that Ed shows does have circs in series, so it will be doubling the head, possibly over-pumping a bit?

Maybe a delta P circ as the boiler circ would minimize that. I think it would work okay without the boiler circ.

Kkane97

I got back this evening and checked it's a crown twz100 100k btu. So sounding like it's definitely oversized, house isn't to badly insulated I redid the crawl spaces and sealed a bunch of windows up a couple years ago. As far as short cycling I don't recall it running for less than 5 minutes at a time which I know is a little low, is the radiant going to cause to to short cycle even worse than that? Lower temperature heat output over a longer duration for the radiant. 

My rough numbers could be off but I agree it's definitely big it seems like, for the last two years on it I heated the whole house for about 375 gallons of fuel

hot_rod

Not so much the low temperature, but when you start zoning on a fixed output, oversized boiler, expect to see more short run times. Shoulder season is where it will be most noticeable.

EdTheHeaterMan

After closer look. and reading Hot Rod Bob's comment. It appears that I may have made my second mistake on HH. How could that be? This is a lower cost option:

hot_rod

We are keeping track

Kkane97

EdTheHeaterMan said:

After closer look. and reading Hot Rod Bob's comment. It appears that I may have made my second mistake on HH. How could that be?

So for now, consensus is simplest design, shouldnt be a issue for now as for as condensate, one mixing valve for the low Temp zone

My simplest way to plumb in would to be leaving my circulator pump on my returns so after my radiant loop in the diagram, my circs on the return side

EdTheHeaterMan

Kkane97 said:

EdTheHeaterMan said:

After closer look. and reading Hot Rod Bob's comment. It appears that I may have made my second mistake on HH. How could that be?

So for now, consensus is simplest design, shouldnt be a issue for now as for as condensate, one mixing valve for the low Temp zone

My simplest way to plumb in would to be leaving my circulator pump on my returns so after my radiant loop in the diagram, my circs on the return side

If it is easy enough to do it, try to put the high temp circulator pump on the supply side after the expansion tank on the supply side. like the diagram above

This would be the least friendly arrangement for air removal.

hot_rod

You need  to pump away from the mix port of a 3 way valve. This allows it to pull from the H &C port of the valve

Kkane97

hot_rod said:

You need  to pump away from the mix port of a 3 way valve. This allows it to pull from the H &C port of the valve

So should be no issue to leave my circs after the radiant zones?

Kkane97

So I have changed everything over and added my radiant floor loop, the PEX lines you see aren't connected and valves are closed for their respective feed and returns. I have a issue with the other two zones, I split on my my existing zones which was too large and made it into two halves with their own thermostat and pump. My issue is when  the zone calls for heat it is also flowing through the zone next to it and vice versa. I recognize I messed up something with the flow and have a venturi flow situation from the pump drawing water through the other zone when either pump is switched on  I believe but not sure where the problem is and how to reconcile it

Kkane97

I think I found my issue, I have taco 218 check valves I mounted them horizontal and they aren't able to seal flow so replacing them with spring type should remedy my unwanted flow, or should I change piping configuration to mount them vertical 

Kkane97

I changed our to spring check valves, all is good. Radiant loop is running 130° return temp is 115 average. We'll see how it does

hot_rod

Pumping down into the return?

Where is the expansion tank connected?

The 3 way thermostatic needs the return from the radiant loop, and a pump on the mix port.

Kkane97

hot_rod said:

Pumping down into the return? Where is the expansion tank connected? The 3 way thermostatic needs the return from the radiant loop, and a pump on the mix port.

I followed this diagram, is it that much different having the pump on the supply? The pressure on the return side of the mixing valve is the same that I see? 

My expansion tank is on my supply pumping away from it

hot_rod

This pic here, it looks like the circs are pumping down into the return of the boiler? Maybe a pic from a bit farther back would show more.

Kkane97

hot_rod said:

This pic here, it looks like the circs are pumping down into the return of the boiler? Maybe a pic from a bit farther back would show more.

They are pumping down into the return this is how I have it all set up now and it's working great. Do you see an issue? I have to finish up the wiring but put it together to test it out yesterday 

Kkane97

hot_rod

It is better to have the pumps on the supply header pumping away from the expansion tank. BUT that would be some major piping change. Systems will work with pumps on the return. Usually.

The mix valve zone needs some more work. It needs to have its own pump to pull water from the hot and cold port to mix properly. Tie the mixed return into that threaded header not into the return from that other zone. You don't want it tied in above the check valve.

Add another tee into the return header, or you have a plug at the end of the header, add another tee there to return that mix loop into. With a purge valve.

EdTheHeaterMan

@hot_rod is this not going to do the same thing as your original drawing

Kkane97

EdTheHeaterMan said:

@hot_rod is this not going to do the same thing as your original drawing

Some fancy picture editing there I like it!

Kkane97

hot_rod where I have the mixing valve placed above the check valve my thought process was that it would pull only it's own zones return to mix, as your have drawn out it would be better to pull water in to mix from the rest of the system on the boiler return side after each respective zone? What is the purpose of which side the check valve it's on? Sorry for the questions I'm just curious as the the difference it makes. My first big project like that and I like to understand as much as I can

hot_rod

The mix valve really wants the pump pulling from the mix port. I think all my chicken scratch drawings show the pump on the mix port?
I could be hallucinating

Kkane97

@hot_rod I'm pulling on the mixed port from the valve, just doing so after the radiant loop. The picture in the last screenshot shows incorrect placement, which has the circulator pump mounted after the cold side of the mixing valve which is not what I have

I think you are hallucinating?

Kkane97

This is how I have it plumbed as my chicken scratch drawing

The initial point of this post was to verify it was ok to have my circulation on the return side of my zones 🤣

hot_rod

The pump pulls from the mix valve, at the mix valve, then pushes the mixed temperature into the loops

It will circulate like you have it, but will not mix properly or accurately

This drawing shows the best way, pumps at the boiler supply pumping away from the expansion tank. And the mixed zone pumping away from the mix valve and expansion.

EdTheHeaterMan

hot_rod said:

The mix valve really wants the pump pulling from the mix port. I think all my chicken scratch drawings show the pump on the mix port?
I could be hallucinating

You are not hallucinating Bob. I doctored the chicken scratch to ask if the pump could be pulling from the mix port after the emitter. the difference being the pressure drop of the emitter will have an affect of the mix pressure at the cold port to the mix port, knocking off the accuracy

You leaned me somethin' new today

hot_rod

EdTheHeaterMan said:

hot_rod said:

The mix valve really wants the pump pulling from the mix port. I think all my chicken scratch drawings show the pump on the mix port?
I could be hallucinating

You are not hallucinating Bob. I doctored the chicken scratch to ask if the pump could be pulling from the mix port after the emitter. the difference being the pressure drop of the emitter will have an affect of the mix pressure at the cold port to the mix port, knocking off the accuracy

You leaned me somethin' new today

The high temp pumps on the return go against the mantra here “pumping away” but I suspect the system will work once adequately purged.

Kkane97

Ok thank you for explaining, since I am able to maintain my mixed temperature would it be ok to leave as is until I decide to change the rest of the house to infloor heat? Or switch around now.