Sanding contacts is a No-No.
The contacts are plated with a very fine layer of material. Once it's gone, the relay is garbage.

From the PDF


The life of the relay is 100,000 cycles @ rated load. Life decreases with temperature.
Over the course of a year, assuming a 2 minute cycle interval, the relay should die in about 4½ hours of use per day.

4.5hours x 30cycles/hr x 2on-off x 365 = 98,550

If the contacts open wide enough, a pencil eraser is acceptable for cleaning relays, assuming things haven't changed radically is the last 25 years.

Seems underengineered for a component to have an expected life cycle that short for what is reasonable usage. I keep my Silvia on 4-6 hrs a day for 7 years now.

I've come to the same conclusion as the last two posters -- that power relay is drastically underspecified for the 1-2 min cycle that it's being used for. An SSR retrofit should be trivial & can be done with a standard (SPST) device; the DPST configuration isn't needed, as the two poles are being used to switch both line and neutral.

I think that the Alex Duetto also uses the same relay in its steam boiler -- I hope we won't be hearing more stories like this in a year or two.

Note: Some of this thread has spilled over to the "other website" http://coffeegeek.com/forums/espresso/machines/397431. I posted there in the thread where George Gillespie (who was, AFAIK, instrumental in solving the 20A->15A problem for the Alex Duetto) had posted an excellent electrical schematic of the machine (page 3 of the thread).

One consideration when doing an SSR conversion is that the minimum contact load requirement is met.

From the PDF:

cafeIKE wrote:One consideration when doing an SSR conversion is that the minimum contact load requirement is met.

I assume this is because of leakage current in the off state? It's definitely not a concern for switching the heating element, but I suppose it was a consideration in the Duetto where the SSR is used to switch the relay coil.

I assumed from reply that SSR conversion mant that the relay would be used to activate the SSR.

In spite of their apparently mundane appearance, relay contacts are engineered. They have a maximum and minimum load. The maximum is intuitive. Low load currents are in the microamp or low milliamp range. Organic vapors [eg : smog] can be a problem. The sliding of the contact surfaces causes polymerization of the organic compounds with the result that deposits with high, unstable resistance are left on the contacts. Low load relays are often gas tight, with non-reactive gases sealed inside. Some specialty relays have mercury wetted contacts.

If replacing the relay with an SSR, then the minimum load requirement is replaced by the need to match of the mechanical relay drive circuit to the SSR drive circuit, along with the correct rating and heat sink for the SSR. Stainless steel is not a particularly good heat conductor. If the SSR is over stressed, it could have as short a life as the mechanical relay.

Like so many things, the devil is in the details.