Hard water to the fridge will intimately kill the ice maker.

RO to the ice maker is the way to go.

KCl (potassium chloride) is the salt substitute for NaCl (sodium chloride) softener regenerant and will work just fine. Depending on your water conditions, brand of softener, and how the softener is set up you MAY need to change the salt dose settings for KCl.

If you change to KCl you can simply add it on top of the NaCl when the level goes down.

Some fridges are very cranky about the water pressure.

If it were me I'd buy the length of PE tubing you would require and try it with the tubing running across the kitchen floor. If the icemaker refills and makes cubes then you'd know it's OK. If the icemaker won't fill then you need to rethink the project.

Now that is a great idea. Why didn't i think of that. I will do just that. Thanks.

FWIW....

At two different locations I have RO supply to freezer ice-makers. Have had these since 1992. Air-tank RO at first; line-pressure tanks for the last year. Both supply-lines run along the floor; one about 20 feet connection-to-connection; the other about 35 feet. Both have worked fine.

Thanks Asolo. I am on a concrete foundation so I don't think I have that option. Unless I am not thinking about this correctly.

At one location on slab with tile on top, we cut into grout between tiles and down into the concrete just a little bit to accommodate the hose. After re-grout, looked like nothing happened. Other location had smaller kitchen and we just went around the room behind the cabinets -- fishing rather than cutting at that location.

Actually I mentioned these installs more to tell you the length of the hose than anything else. That seemed to be a concern of yours. Going up-and-over with similar length would add no new issue at all. Once the water is boosted up-and-over one time there's no other issue as long as the line stays full....which it will. It's not like the RO tank would be working against an imposed pressure-head every time it fills the ice-tray. That would only be true if your ice-tray was up in the ceiling.

Agree with previous poster about running a test-length and trial in order to become 100% confident.

Thanks again asolo. That is good to hear. At this point, I am looking for the path of least resistance in terms of connecting the RO system. It still feels like it would be the attic as regardless of which direction I go, the tube would have to travel across an open space. I probably still need to cut through cabinetry (as well as some kitchen ceramic tile.

In other words, this will be pain no matter what.

Depending where you live FREEZING may be a problem in the attic.

This kind of stuff is ALWAYS a pain unless you're really lucky...

In my experience, there isn't anything having to do with plumbing that isn't a pain.

I do think you have a "conceptually" simple task, though. And, assuming you're successful, I have no doubt you'll really like that ice. My guests and I certainly have these last 19 years.

Thanks again. And just to answer the last few questions:

I live in Southern CA where we consider 49 degrees as freezing. So no, actual freezing won't isn't an issue.

And I have re-thought the path of the hose and think it may be a bit easier. Rather than come up through the tile, I will go up the side wall. Not sure why I didn't think about this before.

Now, it seems that every wall in my house has fireblocks so even that will be require be poking a hole in the wall. But that should be easier than poking a hole in the tile.

Thanks again for the help.

After you run the test with the tubing running out in the open and it works then you sit down and think long and hard about the best route to take and do nothing.

Wait a few days and think on it some more so you give the better ideas time to come to you and then wait a few more days before you drill anything.

"Once the water is boosted up-and-over one time there's no other issue as long as the line stays full...."

That would be correct if the discharge end of the line was at an elevation that is lower than the feed point, inwhich case it would create a syphon effect.

Technically, whether the line is empty or full we must provide enough pressure to overcome vertical static head loss to raise the water up to the horizontal section, however, while we have vertical static head loss on the rise at the input end we also have vertical static head gain on the discharge end.

Example, if the water has an 8ft rise on the input end but only a 6ft drop on the discharge end the total differential is 2ft of vertical head at .439psi/ft-vertical. This means that the static head loss to the discharge end would be 0.439 x 2= 0.878psi.

We then have to compute the friction head loss for the tubing.

3/8" copper tubing with a flow rate of 1gpm has a friction head loss of 0.07psi/ft.

With a total run of 30ft the friction head loss is 2.91psi, however that is assuming a flow rate of 1gpm while your ice maker is in the order of 1/2gpm.

All things considered your vertical head loss and friction loss combined are not significant enough to even concern yourself.

"Once the water is boosted up-and-over one time there's no other issue as long as the line stays full...."

Actually, there is another issue... some later model and new fridges are picky about the pressure they see at the line connection and will not trip the solenoid to let the water in if the water pressure is too low.

That would be true with these fridges regardless of the RO location. Anything that remote location of the RO might or would do to lower that pressure might invite this problem. Best to KNOW before going to the work of running a long line.

To LazyP: Huh?

In a nutshell, LazyP was saying (with the math to back it up) that the pressure drop you will see from the tubing, including the length of tubing as well as the elevation difference from your RO tank to your fridge, is negligible compared to the pressure drop you will have already seen from the RO itself. So, as long as your fridge ice maker will operate at the outlet pressure from your RO, the tubing run is unlikely to cause a problem.

Any updates, Andresrr? What did you end up doing, and how did it work out for you?

I ask because at my wife's dental office, we have an RO system under the sink that is split 3 ways: cold faucet, Quick&Hot instant hot to hot faucet, and refrigerator. I'm pretty sure we're using a 1/4" plastic line to the refrigerator over about a 10' distance with a 4' rise, and it barely dribbled out of the water dispenser. Ice was working then stopped. The whole system seems to have crapped out, with now no cold or hot water flow and obviously no flow from ice dispenser. Grrrr.

I'm trying to do another remote RO to freezer at our new house. The line is about 18-20 feet + 9' rise + 4' down, this time from the beverage center (wet bar) RO system to the freezer. Haven't installed yet. Because of my last experience, I got a permeate pump to see if it will improve the pressures. I also tried copper pipe for rigidity but since reading that copper corrodes with RO water, I'm not having to replace that line with PVC or pex, or consider another RO system by the freezer.

Any great advice would be appreciated!