Hacker News Reader

Comment by amluto | Hacker News Reader

I’m less convinced. Copper is susceptible to its own failure modes. Water moving too fast in a copper pipe can erode the pipe, and “too fast” is quite slow, especially for hot water. If you want hot water quickly, you want the fastest practical speed, which means the smallest practical pipe, and you can easily run into the limitations of copper. PEX and CPVC tolerate considerable higher water velocity. (I think this is because copper’s passive layer can be scoured away, whereas plastic pipes don’t need a passive layer.)

If you have well water or water from anywhere other than a utility that controls the chemistry correctly, the water can dissolve your pipes. The rainwater from my roof makes copper instantly shiny, which is a bad sign for the copper.

View on HN

That’s pretty uncommon, generally the joints fail before pipes do.

PEX with compression fittings is the closest thing to copper in terms of durability, but exists because of labor savings. Appropriately installed copper is the most durable interior water service and will comfortably outlast the life of the structure.

They say all it takes is not de-burring the pipe to create the turbulence.

We have a lot of problems in my county with copper pipes leaking. There are entire plumbing companies dedicated to finding the leaks and repairing them. There are still more companies dedicated to repiping.

I know of developments where they epoxy coated all the piping because they were getting so many leaks.

You can read about the huge class action lawsuits going on involving the local water utility (who is being blamed for water chemistry causing the issues. They point fingers elsewhere, obviously).

I’ve seen a copper pipe where in the middle of the pipe there is a pinhole leak on the top side of the pipe (too far for turbulence, can’t be cause by sediment).

If you want the most durable, I would think it would be either pex-a or perhaps 316 stainless steel (which is a valid code compliant option). But no one really knows, and it’s probably water chemistry dependent. It’s being argued in the courts now (for whatever that’s worth).

> perhaps 316 stainless steel

Stainless steel is awesome! Unfortunately the connections are not so easy. Stainless steel-to-brass threaded connections are probably safe under most circumstances but are dubious per the code. Push-to-connect fittings often use little stainless teeth to hold the tubing in, and this may work less well when the pipe is nearly as hard as the teeth. (John Guest makes special fittings for stainless steel tubing.) You can’t sweat stainless steel pipes.

Obviously everyone should plumb their houses with swagelok fittings! :)

The other option would be brazing with jewelry silver at like $60/oz. Yikes.

Probably a better compromise is stainless press-style fittings. The tools are going to be much too expensive for the average homeowner plumbing needs, but they work out in the commercial sector or for professional plumbers.

MegaPress and ProPress are a couple of popular brands.

The also have an advantage that they avoid "hot work" in places where it leads to greater regulation / stringent safety precautions.

37deg Flare fittings or bust!!

Washington Suburban Sanitary Commission (WSSC) changed its chemicals at some point in the late 1990s, and a lot of customers found pinhole leaks starting in their copper pipes. Considering the relative density of lawyers in suburban Maryland, I suppose there must have been a lot of litigation, though I don't remember hearing.

On the other hand, I lived in a rowhouse built during a brief window in which it was OK with the code to run PVC from the main into the house. Whatever PVC is good for, it doesn't handle shearing force well. Essentially all of the townhouses needed repairs (a copper "pigtail" through the wall).

Three pinhole failures in 30 yr old copper pipe in ten years in this house says otherwise. We've (hopefully) torn all that out - tired of replacing 1/4 of the house every time there's another pipe failure.

One possible contributing factor: flux must be removed from copper pipes (the inside!) for good performance. Plumbing codes require that flux meets ASTM B813, which means that the flux will flush out with cold water.

Sadly, even today, most plumbers will show up to a job site with petroleum-based flux that most certainly does not meet B813. It can stay in pipes for years, slowly leaching chemicals that do their job: weakening the protective oxide layer in the pipe. Also it tastes and smells terrible.

Solder doesn't last forever right? When did we start using it over iron?

I had iron pipe in my first house. Water was rusty when we got home after a weekend away. Got tired of rust stained sinks and laundry.

Ripped it all out and replaced it with CPVC (this was before PEX was really widely available). Never had trouble after that.

Not to mention water itself is horribly corrosive. Pure water will strip copper pipes. That’s the real reason fluoride was added to tap water: 1) it was cheap 2) it was safe to drink 3) it kept the water from eating the city’s pipes.

We had a city in Illinois do whole city RO and advertise the cleanest water in the country. Well it was the cleanest water going into the pipes, but if you own a house in that city the water coming out of your tap is full of old pipe!

> water itself is horribly corrosive

Funnily enough, this is technically true, but would only manifest in a warped reality.

In the acidic/caustic scale, regular water is just slightly on the caustic side of neutral. (pH 5.5 or thereabouts?) - It won't strip the pipes it runs in. But that's not the kind of water we're talking about here. That would be pure distilled water.

It's the near-universal cleaning agent. Because of the complete lack of impurities, distilled water has the ability to absorb fair amounts of almost anything it gets in contact with. It also tastes really bad. (The taste of "nothing" for regular water is our calibrated baseline with various minerals and other impurities. If you take all of that out, the perceptual difference between expected and experienced is quite stark.)

For what it's worth, cavitation and turbulence are much more likely culprits to eat the plumbing material. Also: shoddy quality will always cause problems, regardless of your choice of materials.

That's complete bullshit.

Only acidic water (and strongly so) will damage copper pipe.

Most copper pipe will easily last 50-100 years, if not more.

Flouride isn't added in any sort of concentration that would measurably change water Ph.

And more.

Your statement is at odds with what I know about water.

Your saturation index is what drives water’s interaction with the surfaces it touches, not just pH. If the index is too high it will deposit what it’s carrying (scale) and if it is too low it will pick up material from the pipe it’s flowing through (corrosive).

These are used in swimming pools, but originally the saturation index was developed by Langelier to protect city pipes from corrosion: https://en.m.wikipedia.org/wiki/Hard_water#Langelier_saturat...

IIRC most cities try to run a high saturation index to intentionally lay down scale that further protects the pipes from corrosion.

It may be true that fluoride isn’t in sufficient quantity to impact the saturation index, but that is distinct from pH.

1) That isn’t distinct from PH, it’s a refinement of it?

2) There is theory, and there is practice in the field. What you’re referring to is the equivalent of ‘wet bulb temperature’ for temps. Which is a thing, but water conditions for a given area rarely vary like temperature will - and are in control of the local water treatment plant. What you’re referring to was specifically made to measure and help control these conditions.

If copper piping is allowed by local code, they’ll know - and keep it within these ranges. Areas with water that is not easy to keep within these ranges will also corrode galvanized pipe (commonly used for water in most areas), and will likely also corrode brass fittings and valves (including back flow valves) commonly used in many areas, leaching lead, copper, etc. among other things. Those areas prohibit them by code for that reason.

Copper piping will generally not corrode unless PH is < 6.5 or > 8.5, which is significantly outside what most would consider reasonable water quality.

I personally have a well in an off grid location with 6.2 PH, and use only CPVC for that reason. Where I live, I’ve done some retrofits and replaced 75 year old copper pipe that was good as new inside.

> If copper piping is allowed by local code, they’ll know - and keep it within these ranges.

I think I’m misunderstanding you. Your original comment was “that’s bullshit” but this seems like it’s in rough agreement.

IIUC copper pipe will last for 50-100 years IFF you pass impure water through it. Pure water will eat the pipe. Same with galvanized steel and brass like you pointed out. So the city treated it with _something_ to keep the water from eating pipe (knowing copper pipe is code) and I’ve anecdotally heard that part of that “something” is fluoride.

No, that’s pure bullshit, for exactly the reasons I explained. What PH do you think pure water has, exactly?

LSI seems to trend towards negative infinity as TDS et. al. approach 0.

    pH = 7 (pure water)
    TDS = 0 mg/L
    Calcium = 0 mg/L (or ppm) as CaCO3
    Alkalinity = 0 mg/L (or ppm) as CaCO3
    Temperature = 21 celsius (room temp)

A = (Log10 [TDS] - 1) / 10 = (Log10 [0] - 1) / 10 = -Infinity

B = -13.12 x Log10 (Temp + 273) + 34.55 = -13.12 x Log10 (21 + 273) + 34.5 = 2.12

C = Log10 [Ca2+ as CaCO3] - 0.4 = Log10[0] - 0.4 = -Infinity

D = Log10[alkalinity as CaCO3] = Log10[0] = -Infinity

pHs = (9.3 + A + B) - (C + D) = 9.3 + A + B - C - D = 9.3 - Infinity + 2.12 + Infinity + Infinity = Infinity

LSI = pH - pHs = 7 - Infinity = -Infinity

https://www.wolframalpha.com/input?i=7+-+%28%289.3+%2B+%28%2...

From my understanding a negative LSI means the water will try to "pick up" atoms/molecules from the container it is in (corrosive). A positive LSI indicates it will lay down atoms/molecules (scale build-up).

A value of negative infinity tells me that pure water is going to eat the pipes.

That being said, I don't know if the LSI scale is well behaved as these values trend towards 0! The values of water purity in the equation seem to dominate both pH and temperature as those values trend towards 0 - no big difference in results if pH is 7 vs. pH is 3 and no big difference if water is frozen or boiling. Temperature can have a big impact for _very cold_ and _very hot_ but then you're dealing with solid water or steam and that doesn't really make sense to me for this model. That probably suggests the equation is only well behaved for a certain range for each value, and I'd suspect 0 is outside the range these equations can model for TDS, Calcium, and Alkalinity. But I can't find anything that defines the ranges the LSI model applies to.

And yet, distilled water tests a 7 with a PH strip, people use copper pipes with distilled water with no major ill effects, and distilled water won't eat it's way through a copper pot either.

Can’t tell if you’re trolling me at this point.

I wonder what you think the pH of steel wool is. I bet you could place steel in a copper pot and it won’t eat through it.

You're clearly trolling someone. I guess I'm done wasting my time on you?

Your assertion is that 'pure water' will corrode copper pipe, so it needs to be 'impure' to not do so (using flouride of all things as an example?)

Yet, distilled water - which is as pure water as anyone will ever see outside a lab - doesn't do so to any meaningful extent. And you keep using various clearly inapplicable theory to try to prove a point that is clearly false in the real world.

This has nothing to do with steel wool, after all.

Lol don’t you love it when you give a smack down with math?

If you replace fluoride with various mixes of phosphates, calcium, and carbonates, you might be correct.

Where did you get the info about fluoride?

From an ex-well driller - which I yield isn’t the most reliable source.

I haven’t been able to find a source since posting this. I may have my facts backwards! The closest I have is this CDC report saying it reduces corrosiveness: https://www.cdc.gov/fluoridation/engineering/corrosion.htm

My understanding is that, if you take fluoride out of the water, you’ll need to replace it with something or the water will replace it itself (aka eat the pipe).

That obviously doesn’t mean fluoride was originally added for that reason though.

I doubt most freshwater sources have enough minerals in them to not be corrosive enough to need to add anything to reduce corrosiveness (ground water wells, rivers, etc...). The only purpose for adding fluoride would be to add fluoride for dental purposes.

The only thing I could think of that might have corrosive issues would be desalination water that was overly purified like the sibling comment mentioned. But even then, they should be adding some kind of minerals back before distributing it to people's homes. Ultrapure water tastes horrible and likely isn't safe for drinking long term.

I recall reading somewhere the massive underground water container project in Qatar ran into issues with RO water eating pipes everywhere

So much wrong here I don’t know where to start.

> Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something.

I’m totally open to being wrong. I’m eager to learn - which is generally why I share information. If I don’t readily share what I know to be true, like in this case, nobody will ever correct me and I’ll spend the rest of my life being incorrect.

On HN you generally start by telling someone why they’re wrong and supporting that in some way.

I.E. if you know/believe fluoride was added for another reason - that’s a good place to start.

Or if you know/believe purified water is not corrosive, that would be a good place to start.

Reply on news.ycombinator.com