What to replace a rusted rooftop TPRV? (stainless Temperature Pressure Relief Valve?)

[brycenesbitt]

What could I replace a rusted rooftop TPRV with?
The valve I have is a regular water heater TPR about 40 years old, that no longer reliably closes

Images are here:

https://imagebin.ca/v/6lVRyoqbMyWZ

https://ibin.co/6lVRyoqbMyWZ.jpg

As for some reason the form won't accept an image upload for this post.

[Mike 134]

Plumbing supply stores and big box home improvement stores will have replacements. BTW they are typically made of brass.

[brycenesbitt]

Plumbing supply stores and big box home improvement stores will have replacements. BTW they are typically made of brass.

The big box TPRV's have a steel center shaft. That's what rusted in the above photo.
I need something with a stainless steel center shaft, rated for outdoor exposure per the photo.

---
Does the forum support uploading images? I get "Error uploading image".

[Mike 134]

The big box TPRV's have a steel center shaft. That's what rusted in the above photo.
I need something with a stainless steel center shaft, rated for outdoor exposure per the photo.

---
Does the forum support uploading images? I get "Error uploading image".

that leaves a plumbing supply house

[sdold]

The big box TPRV's have a steel center shaft. That's what rusted in the above photo.
I need something with a stainless steel center shaft, rated for outdoor exposure per the photo.

---
Does the forum support uploading images? I get "Error uploading image".

It used to, but I can't upload even a small image now. We may need to look into this.

[J.P.M.]

[QUOTE=brycenesbitt;n434196]

The big box TPRV's have a steel center shaft. That's what rusted in the above photo.
I need something with a stainless steel center shaft, rated for outdoor exposure per the photo./QUOTE].

Given their function and service, relief valves are designed very robustly. However, they do need periodic inspection and usually replacement more often than 1X/40 years.
As I rule I check/inspect and manually lift the PT valve on my solar water heater every year and replace it every 5 provided the check lever doesn't fail (which happens sooner that later usually as f(water mineral content). I use standard PT valves from Watts. They're about 12 bucks or so.

I've specified st. stl. relief valves for specialized sanitary or steam service when I was a working engineer but I doubt you'll find a PT valve with a st. stl. stem only and brass/bronze body. So be prepared to either spend the BIG $$ for an all st. stl. valve and watch it crud up just like a PT valve with standard materials if you have hard water.
Or get a big box valve for a whole lot less and change it out every few years as it cruds up.

BTW, if you have a check valve in that system and it gets the same service/inspection schedule as that PT valve, there's a high probability it's not functioning, meaning your system is not doing much of what it was designed for.

[brycenesbitt]

BTW, if you have a check valve in in that system and it gets the same service/inspection as that PT valve, there's a high probability it's not functioning, meaning your system is not doing much of what it was designed for.

This system is NOT in a hard water area, so the first thing to go was the shaft rust. Our water does not do much of anything to metal.
The check valves have never been serviced or tested.

[J.P.M.]

This system is NOT in a hard water area, so the first thing to go was the shaft rust. Our water does not do much of anything to metal.
The check valves have never been serviced or tested.

Well, the spring was probably right behind it on rust progress and if so, the spring constant changed. I'd bet that valve you have doesn't have an adjustable lift pressure, all of which means the valve won't go off at its design set pressure, if at all.

Just replace the damn valve with one that has the same pressure and temp. rating as the old one and service/replace the new valve every few years. Better, cheaper, faster, solution as well as a whole lot safer than never inspecting a safety valve.

I'd also take the check valve out and see if it's still doing what it was designed to do.

[brycenesbitt]

I found a model with a brass center shaft, and have just replaced it.
The check valves operate fine (they are soldered in, so not easy to "remove").

[J.P.M.]

I found a model with a brass center shaft, and have just replaced it.
The check valves operate fine (they are soldered in, so not easy to "remove").

Given the quality of most check valves for residential service I've seen, if check valves are still functional after a few years, either your water is pretty mineral free or you're lucky, or both. How do you know they are still tight ?

[brycenesbitt]

Given the quality of most check valves for residential service I've seen, if check valves are still functional after a few years, either your water is pretty mineral free or you're lucky, or both. How do you know they are still tight ?

I think it helps that the check valves are 40 years old, and stuff was both just made better then, and it was good kit then.
There was a filter basket... all it had was some gunk that looked like flux. Very clean water from the Mokelumne river north of Yosemite National Park.

----
I tested by opening a drainback valve, and leaving pressure on the wrong side of the check valve.

----

But the new TPR failed today.
It was a hot day, the controller shut due to overtemp (as it should),
the TPR activated after the array reached 150F or so (as it should),
but it never really closed again, even after cooler water was pumped to the roof.

[peakbagger]

Do you have an expansion tank on the loop and have you confirmed that the bladder is intact and set to the correct pressure? If the tank is not working, and the fluid expands on a hot day, glycol is going to find a place to leak and the TPR may be the weak point.

[J.P.M.]

I think it helps that the check valves are 40 years old, and stuff was both just made better then, and it was good kit then.
There was a filter basket... all it had was some gunk that looked like flux. Very clean water from the Mokelumne river north of Yosemite National Park.

----
I tested by opening a drainback valve, and leaving pressure on the wrong side of the check valve.

----

But the new TPR failed today.
It was a hot day, the controller shut due to overtemp (as it should),
the TPR activated after the array reached 150F or so (as it should),
but it never really closed again, even after cooler water was pumped to the roof.

What type of check valve is it ? Of all the types of check valves and because of the small sizes usually required for domestic service they are usually spring loaded but sometimes gravity operated checks are used . Gravity checks are probably better but only if they can be isolated for regular inspection/cleaning, and they must be vertically oriented.

Spring loaded valves for domestic water service fail sooner with the common mode of failure usually being the seal - particularly in the cheap, residential checks because that seal is most often something called Buna-n (or just buna- synthetic and cheap rubber). Buna has an upper service temp. of ~ 60 C. and also doesn't like regular temp. excursions which increases its stiffness, causing it to not seat properly. Other seat materials like Viton have higher temp. limits but different hardness requiring stiffer spring constants and no one wants to spend an extra few $$'s for material and a redesign of an entire valve, which design has probaly been around for the better part of a century.

As for older being better, that may be, but check valves in potable water systems haven't changed much in my working lifetime of well over 50 years. They almost never fail due to materials of construction. When they do fail, it's mostly due to water chemistry and/or suspended solids in the water that collect on surfaces where materials don't meet as good as they used to and or usually at places here material change directions and or at surface discontinuities (think of corners or sharp edges) that, not coincidentally and sort of off topic but maybe insightful, are also sites for stress concentrations. Anyway, the crud loosens things up, much like the difference between a newly weatherstripped door with magnetic seals and one with old weatherstripping that blows open in the wind.

Your test wasn't a good one for a determining if a check valve is functioning to prevent thermosiphoning in a solar domestis hot water system.
good (leak tight) seat on a spring loaded check from high pressure might seem the right test, but the idea is for a check valve for the prevention of thermosiphoning needs to operate at a small differential pressure and keep a tight seal while doing so.
Keeping a tight seal at a relatively higher differential pressure is not usually a good indicator that a check valve is doing its job under operating conditions.
Think about it. DHW check valves need to operate at some relatively small cracking (opening) differential pressure. In check valves, to prevent thermosiphoning, that differential pressure is caused by warm(er) water being less dense than cooler water. For most residential systems that differential pressure amounts to maybe a few inches of water column or some fraction of a PSI. A crudded up seal and/or a brittle o-ring may well need more differential pressure to close than may exerted by temp. differential induced pressure differences alone, so the may leak. On the other hand, or for your test, at a differential pressure of maybe 3 or 4 bar (roughly 45 or 60 PSI), such as you used, the chance of a forced and so a tight seat on a crudded up pair of surfaces is a lot greater and in any case such a test means little with respect to the cracking or lift pressure of a check valve.

On the pressure/temp. relief valve, technically, it reads to me like it didn't fail. It prevented an upset condition from getting worse. That is its job.
That it didn't re-seat properly is a common occurrence with cheap relief valves. Also, as Peakbagger writes, without an expansion tank, system pressures from thermal expansion can cause a relief valve to lift. All ASME pressure and temp. relief valves require a means to test for continuing operation. Cheap domestic valves meet that criterion with a test lever. A great idea but most experienced plumbers and service technicians I know will never use it to test a relief valve. Reason ? Once the lever is activated, there's a good chance the valve will leak. Now the test lever opens the valve just like an overpressure or overtemp. event does.

As an aside, once a valve lifts for whatever reason it's good practice and sometimes (depending on service) a requirement to replace the valve.
For lots of reasons, a lot of folks question the wisdom, logic and safety of putting test levers on relief valves. Go figure.

[LucMan]

But the new TPR failed today.
It was a hot day, the controller shut due to overtemp (as it should),
the TPR activated after the array reached 150F or so (as it should),
but it never really closed again, even after cooler water was pumped to the roof.

The most widely used TPRV's are rated 210 degrees F and 150 PSI if it's releasing you have another problem. Most likely the expansion tank / steamback bladder is defective allowing the pressure to exceed 150PSI that's why it wouldn't reseat.

[J.P.M.]

The most widely used TPRV's are rated 210 degrees F and 150 PSI if it's releasing you have another problem. Most likely the expansion tank / steamback bladder is defective allowing the pressure to exceed 150PSI that's why it wouldn't reseat.

I'd first check to see if the system has an expansion tank. It should, but I've seen many that don't, particularly if it's a direct system tied to the rest of the DHW system that may and should already have an expansion tank even though that tank may be undersized if the solar water heater was add to the system.
Still, I'd agree that if there is an expansion tank, the tank's bladder may have failed. They don't have a long life expectancy.

Q for the OP: Is the PRV "weeping" more/less continuously or lifting intermittantly and releasing a fair amount of liquid and then reclosing tightly ? If the first, I'd suggest it's the valve that's the problem. If the second, it's likely the expansion tank. If both, then both are bad.

A typical domestic PRV has positive action in that it will actuate in a very definite and abrupt manner and release a fairly large amount of fluid in doing so, and then close just as abruptly. However, if the seal is compromised by faulty reseating caused by seat fouling or misalignment for whatever reason, the valve will still operate as it should but also weep. Or, the weeping may relieve the overpressure by leaking just enough to keep the valve from actuating until the increased pressure from a relatively large temp. increase from relatively rapid thermal expansion causes pressure increase than the weep rate can accomodate so the valve operates. as it should.