Author Topic: Hydronic heating update and request for sanity check  (Read 442 times)

a_scanner_brightly

  • 5 O'Clock Shadow
  • *
  • Posts: 49
  • Location: Eugene, OR
Hydronic heating update and request for sanity check
« on: October 11, 2018, 12:05:00 PM »
So this is part something of a continuing saga to resurrect a hydronic water heater system in my house after I bought it.

Previously discussed here:
https://forum.mrmoneymustache.com/do-it-yourself-forum!/is-it-correct-to-expect-a-system-that-will-be-circulating-water-to-be-airtight/msg2136926/#msg2136926

So after ensuring the under-floor piping was air-tight (above thread), I went ahead and assembled a bit of what looks like a Rube Goldberg machine (to me) to do the water heating and circulating.  It also has a 2.2 gallon expansion tank (overkill, the whole system only holds about 5 but it was the smallest one I could find) and a TP relief valve.  It's more or less the MMM diagram from his Radiant Heat Experiment, except I use a small point-of-use tankless electric heater (since it's for one room only) and it's closed loop.  I even use the same variable speed model circulator pump too, mainly because it was the most cost-effective since it has a built-in relay.  Will post some pictures soon!

Before adding water I thought I should apply air again to test for leaks.  I pumped it to 20 psi yesterday and by this morning it had... dropped to <10.  Darn.

I'm now doing a divide and conquer approach of the system by closing valves and pumping up to see if I can isolate the leaks to a particular section before I start dish soap spraying joints to look for bubbles.  In the meantime I thought I'd float some sanity checks past y'all.

Sanity check question number 1: could the circular pump or the water heater (neither of them powered on yet) be venting air but not water?

Sanity check question number 2:  also while applying some PEX crimp connectors and assembling the whole Goldberg Machine, some of the joints did swivel a bit in-place.  Does this break the PEX crimp seal, or can they do that within some tolerance?

a_scanner_brightly

  • 5 O'Clock Shadow
  • *
  • Posts: 49
  • Location: Eugene, OR
Re: Hydronic heating update and request for sanity check
« Reply #1 on: October 12, 2018, 11:23:04 AM »
Solved.

The spray-bottle test revealed that out of the 30ish PEX crimps I did, where some of them are even kind of crooked and swivel in place, the leak ended up being from the TP valve not being screwed on tightly enough.

PEX roolz!

J Boogie

  • Pencil Stache
  • ****
  • Posts: 749
Re: Hydronic heating update and request for sanity check
« Reply #2 on: October 12, 2018, 12:16:07 PM »
PEX is great. I prefer the solid copper rings with the big pincher instead of the steel crimper rings, but I used both during my bathroom remodel - I used the crimper rings in tight spaces or where I connected into a solder joint. Reason for that being I am total newbie when it comes to sweating copper, and I wanted to be able to easily remove the pex joint quickly in case I needed to re-do the solder joint. If I would have used the copper rings, I would have had to cut the fitting off, losing both a fitting as well as a potentially crucial length of pipe.

BDWW

  • Bristles
  • ***
  • Posts: 416
  • Location: MT
Re: Hydronic heating update and request for sanity check
« Reply #3 on: October 12, 2018, 01:06:47 PM »
Are you not planning on adding an air scoop?

Just curious, the whole idea of testing air in your system seems  a bit odd, unless you have a scoop on a valved circuit.

a_scanner_brightly

  • 5 O'Clock Shadow
  • *
  • Posts: 49
  • Location: Eugene, OR
Re: Hydronic heating update and request for sanity check
« Reply #4 on: October 12, 2018, 01:23:26 PM »
Quote
Are you not planning on adding an air scoop?

Should I add one?  I'm not totally convinced I need it and they're pricey, so I decided to re-evaluate for v1.1

Quote
Just curious, the whole idea of testing air in your system seems  a bit odd, unless you have a scoop on a valved circuit.

Someone here suggested it a few months ago and I thought it was genius.   It's better than testing it with potentially property destroying water right?
« Last Edit: October 12, 2018, 01:25:08 PM by a_scanner_brightly »

BDWW

  • Bristles
  • ***
  • Posts: 416
  • Location: MT
Re: Hydronic heating update and request for sanity check
« Reply #5 on: October 12, 2018, 01:38:03 PM »
Well, you will get air in the system. If you check somewhat regularly and have a way to manually vent it, you can get by without it.

To your second point, sure I suppose, but it's overkill. Airtight > water tight. And it doesn't work if you have a scoop in your circuit. You could accomplish the same thing using water. Small amounts of water are not a problem. It's a lot of water, or water over a long period of time that's problematic(drywall being the exception maybe).

a_scanner_brightly

  • 5 O'Clock Shadow
  • *
  • Posts: 49
  • Location: Eugene, OR
Re: Hydronic heating update and request for sanity check
« Reply #6 on: October 12, 2018, 02:07:46 PM »
Quote
Well, you will get air in the system. If you check somewhat regularly and have a way to manually vent it, you can get by without it.

Oh.  Hmm.  I guess I'll up the urgency on planning to splice one in. Thanks for the encouragement. :)

For posterity, what's a way of manually venting?  I was planning to manually connect water from the mains into it, opening a drain and collecting that into a bucket until the drain stops shooting out air.

BDWW

  • Bristles
  • ***
  • Posts: 416
  • Location: MT
Re: Hydronic heating update and request for sanity check
« Reply #7 on: October 12, 2018, 02:57:19 PM »
That works, but it probably won't get all the bubbles. Air is obviously less dense, so a valve at a high point in the circuit would work better.

A small amount of air in the system won't cause too many problems, but it reduces the efficiency of the system, and a larger amount of air in the wrong place (pump) can cause the circuit to stall.

An air scoop does a much better job of removing air. It's essentially an air trap with a float (not unlike a toilet float). The water flows over some turbulent material, and air is broken up and directed to the top of the scoop. Once enough air is in it, the float sinks and opens a valve allowing the air to escape, and water is allowed back in and closes the float.

If you keep your circuit connected to the mains - with a check-valve - the process is automatic, and you generally don't have to worry about it. Air goes out, and the check valve puts a bit of water back in to keep the system going.

a_scanner_brightly

  • 5 O'Clock Shadow
  • *
  • Posts: 49
  • Location: Eugene, OR
Re: Hydronic heating update and request for sanity check
« Reply #8 on: October 12, 2018, 03:02:23 PM »
Awesome.  I actually have the mains inlet on a check valve just in case I learned that I needed to leave it connected full time, say if there was an ever so miniscule harmless leak that added up over time.  Good to know I have a solution if I add an air eliminator. :)

Hey, do all air eliminators/air scoops need to be mounted on a horizontal run of pipe? The entirety of the system before the pump is on vertical pipe. There's a horizontal leg of it downstream from the circulator pump that the eliminator can go on, but it seems like ideally it would be before the pump?  Will it be effective at all if it's a few inches downstream?
« Last Edit: October 12, 2018, 03:06:33 PM by a_scanner_brightly »

BDWW

  • Bristles
  • ***
  • Posts: 416
  • Location: MT
Re: Hydronic heating update and request for sanity check
« Reply #9 on: October 12, 2018, 03:11:15 PM »
It's a circuit, a little ways downstream is the same as a long ways upstream :) . That being said, before the pump is probably better, but it should work fine once the system is pressurized to keep ahead of any air intrusion.

Anyway, the general rule of thumb(might be code) is that it should be placed 16 inches from an elbow or other element (pump) causing turbulence. This allows the air to coalesce before it hits the scoop, and allows it to be more efficient.

I'm not aware of any that work on a vertical run.