Burlingame Eichlers / Radiant heat
The system under the floor
Every Eichler here is copper. That is the single luckiest fact about this tract.
Eichler's early houses have steel radiant tubing in the slab. When steel corrodes through, the corrosion tends to spread, and repair stops being practical. He switched to copper, and copper systems are repairable one leak at a time. Mills Estate was built in 1964–65 — comfortably on the right side of that changeover.
Hold a magnet against the tubing at the manifold, where the loops come out of the slab near the boiler. Steel is magnetic. Copper is not. If the magnet does not stick, you have copper — which is what we would expect in every house on the eight streets.
How it works
The floor is the radiator.
A gas boiler heats water and circulates it through tubing cast into the concrete slab. The slab is both the structural floor and the heat emitter. There are no ducts, no registers and no air being blown around — which is exactly why the houses feel the way they do, and exactly why there is nowhere obvious to put a conventional furnace when the system fails.
Typically the whole house runs on one or two zones. Room-by-room control was not part of the idea. The slab takes hours to come up and hours to come down, so an Eichler is heated by setting it and leaving it, not by adjusting it.
What we cannot tell you is the tubing's diameter, wall type, or how the loops are laid out in a Mills Estate slab. We looked. No published source gives it for any Eichler tract, and we are not going to fill it in from general 1950s radiant practice and present the guess as a specification.
If you have an original plan sheet showing the loop layout, that would close one of the more annoying gaps on this site. The rest of the list →
Why they fail — and the sources disagree
Two accounts, and the argument between them is the useful part.
Account one: mechanical stress. The Peninsula's best-known Eichler radiant specialist argues that copper in a slab does not corrode at all, because the loop is closed and holds the same water year after year. What kills it is movement: the slab cracks over decades, tubing crossing a fracture flexes, and the copper work-hardens and splits.
Account two: pinhole corrosion. The more commonly repeated version — aggressive water chemistry, oxygen in the loop, dissimilar metals.
The bridge between them is the automatic fill valve, and this is the thing worth understanding as an owner. A sealed closed loop really does stop corroding. But a system with a slow leak and an auto-fill valve is no longer a closed loop — it is being continuously topped up with fresh, oxygenated water. At that point corrosion becomes available as a mechanism, and the leak is quietly accelerating its own cause.
So both accounts can be right in sequence. Movement opens the first leak; the fill valve hides it; the hidden leak changes the chemistry. Which is why a fill valve that runs is the alarm, not a nuisance.
We could find no building-science or peer-reviewed source that settles this, and the stress account comes from a firm that sells the repair. We are giving you both because the practical advice — watch the fill valve — follows from either one.
The number that actually predicts failure
It is the slab's behaviour, not the tubing's age.
| Claim | Where it comes from | How much weight to give it |
|---|---|---|
| An average of 1–5 leaks develop after 30–50 years of slab stress in houses built 1950–65. | A Peninsula radiant specialist, echoed independently on the Eichler Network owner forum. | Reasonable. Two roughly independent statements of the same range. |
| A 1958 copper system still sound at 50+ years, with the observation that a slab which has not shifted in fifty years should stay good indefinitely. | Owner report on the Eichler Network forum. | Anecdote — but it is the right frame. Ask about the slab, not the calendar. |
| Seven or eight leaks a year means stop repairing and convert. | Eichler Network reporting. | The cleanest decision rule anyone has published. Worth holding onto. |
Mills Estate houses are turning 61–62 years old as of 2026 — squarely inside the 30-to-50-year window where the first leaks are expected, and past it. A house here with no repair history is not necessarily a house with no leaks. It may be a house nobody has tested.
Before you buy one
The pressure test, done properly.
The specialist firm on the Peninsula publishes an actual protocol for real-estate transactions. It is specific enough to hand to an inspector, so here it is. We are quoting them, not certifying it.
If it does leak
Four routes, in the order most owners should consider them.
| Route | What it involves | What it costs you beyond money |
|---|---|---|
| 1. Spot repair the usual first answer |
Locate the leak, open roughly six inches of slab with a small electric jackhammer, cut out the damaged section, silver-braze in a slip joint, patch. A typical 2,000 sq ft house is described as one working day, and furniture and flooring generally do not have to come out to find the leak. | Almost nothing. This is the option that keeps the house as designed. |
| 2. Repipe in PEX | Either slot-cut grooves into the existing slab and lay new tubing, or pour an overlay on top and abandon the original loops in place. | Floor height. See below — this is the one that can quietly wreck the house. |
| 3. Convert to mini-splits | Ductless heads on the perimeter walls. The reason this is the consensus conversion in Eichlers is structural: exposed beams and no attic mean there is nowhere to hide ductwork. A head needs only a 3–4 inch hole through the wall. | The floors go cold, and you accept equipment on the walls. Owners who have done it are candid that they miss the warm slab. |
| 4. Small-duct high-velocity, hydronic baseboard | Both appear in the literature as options. | Baseboards cost you wall space, which an Eichler has very little of. We have found no Eichler-specific reporting on either — treat as less well-trodden. |
| Epoxy lining the existing loops | Marketed for potable pipe; one source flags that warranty ceilings sit around 140°F while radiant operating temperatures spike higher. | We would not. Single-sourced caution, but the temperature argument is easy to check with any manufacturer. |
The Eichler-specific trap
An overlay raises the floor. The floor meeting the glass is the whole house.
Every overlay approach — poured or panelised — adds height. Three quarters of an inch, an inch, more with finishes. In most houses that is a trim detail.
In an Eichler it collides with the one detail the architecture is built on: the flush threshold at the sliding glass doors, where the interior floor runs out level to the terrace and the inside and outside read as one room. Raise the floor and you get a step. Get a step and the house stops doing the thing it was designed to do.
It is solvable — door clearances can be reworked, thresholds redetailed — but it has to be designed for at the start, not discovered at the end. If a contractor quotes an overlay and does not raise the slider transition unprompted, that tells you something about how many of these they have done.
The same logic applies to flooring. An Eichler's floor thickness budget was spent in 1964. What else is structural here →
Money
One set of published figures we will repeat, and a pile we will not.
The Peninsula specialist publishes a rate card, which is unusual enough to be worth something. Their stated range for a typical Eichler slab repair is $1,250–$2,500, with the statement that it should never exceed $7,500 even in a difficult house. Detection and brazing line items are published alongside.
We are repeating that because it is on the record, it is attributable, and it has an upper bound attached — which is the part that actually helps you in a negotiation.
Everything else we found, we are not printing. The conversion and repipe figures circulating online come from real-estate blogs with no attribution, and one of them gives two different ranges for the same repair in the same document. That is not a price; that is a number somebody typed.
Get three bids. If a figure matters to a decision you are making, it is worth an afternoon of phone calls rather than a number lifted off a page that will not stand behind it.
Timing, if you are in contract
Two to four weeks to get a specialist out. Plan your contingency around that.
In-slab repairs are reported as taking one to two weeks to schedule even in an emergency, with routine service waits of two to three weeks on the Peninsula and South Bay. There are perhaps half a dozen firms in the Bay Area that do this work properly and they are busy.
The practical consequence: if radiant matters to your offer — and on a sixty-year-old slab it should — book the test the day you go into contract, not in week two. Remember the 24-hour cool-down sits inside that window too, which in a house being shown is not always easy to arrange.
Talk to us
Want a second opinion on a quote?
Send us what you were told and what it costs. We have watched a lot of these systems get diagnosed, well and badly, and we will tell you which one this looks like. No charge and no obligation to list with us.