2026-11-12 · 4 min read
Two boxes of E7018 sit on the same shelf. One was opened on Monday and taped shut again. The other has been open since last month. They carry the same stamp and the same certificate. Only one of them is still low hydrogen.
That is the awkward part of this grade. Low hydrogen is not a property the coating owns. It is a condition the coating is in on the day you strike the arc.
What the classification actually promises
Take E7018 H4R. Each part is a separate promise, and only some of them survive a month on a shelf.
- E7018 is the coating itself: a basic, lime-type flux. The calcium fluoride in it is what suppresses hydrogen pickup during welding. That part does not decay.
- H4 is a measured limit: 4 millilitres of diffusible hydrogen per 100 grams of deposited metal, measured on a sample made from electrodes in a known condition.
- R is a moisture-resistant coating. It resists taking water back. It does not refuse it forever.
The first survives storage. The other two are claims about a state, and a state changes.
The numbers worth asking for
The certificate for this grade lists a maximum coating moisture of 0.3 per cent, and reports 2 millilitres per 100 grams against the 4 that H4 allows. Both are measured values on the same document. That is what makes them worth reading: one says what the coating may contain, the other says what it actually did in a test.
There is a detail in that certificate most buyers never notice. AWS A5.1 tests a moisture-resistant coating over nine hours of humid exposure. If a supplier quotes twenty-four hours, that is its own test, run longer than the standard asks. It is not dishonest and it is not comparable. Nine hours is the number you can hold two suppliers to.
What a holding oven protects
An oven keeps the coating at the moisture level it left the factory with. That is the whole job. It does not dry an electrode that already took on water during the day, and it is not a substitute for leaving the pack sealed until you need it.
The detail that matters on site is the log. An oven at the right temperature with no record of what went in and when is a warm cupboard. The question an inspector asks is not whether there is an oven. It is what the electrodes were doing between the oven and the joint.
What storage cannot fix
Four things, and they come up often enough to list.
- A coating that has already absorbed moisture. Re-drying removes the water; whether it restores the coating is a question for the datasheet, and the honest answer for a badly stored batch is to stop treating it as low hydrogen.
- The core wire. Rust under the coating is a separate problem, and drying does nothing about it.
- Mechanical damage. A chipped coating at the arc end changes how the arc starts and where the hydrogen goes.
- The base metal. Hydrogen already in a thick, restrained section, or a joint that was never preheated, is not fixed by a dry electrode.
The failure you will not see
Hydrogen cracking happens after the weld has cooled. Sometimes hours. Sometimes days. It appears in the heat-affected zone of a thick section or a restrained joint, as a crack nobody was welding when it opened.
That is why the storage question is not housekeeping. A damp electrode produces a weld that passes visual inspection and fails next week.
Rules that survive a real workshop
- Keep the pack sealed until the moment of use, and reseal what you open.
- Holding oven for the working day, not the working month.
- Re-dry only to the temperature the datasheet for that grade names. It differs by coating, and guessing is how coatings get ruined.
- Write it down: grade, date opened, oven, who put it in.
- If the procedure names H4 or H8, the certificate has to name it too, and the storage has to match the claim.
What to put on the order
If your procedure names the classification, order it in full: E7018 H4R, not E7018. The two trailing characters are the part a certificate has to prove.
If you only need a basic electrode and nobody measures hydrogen, say that instead. It is a cheaper conversation and an honest one.