With a quizzical look, JF Mezei observed:
On 2020-11-07 08:28, David Spain wrote:

Pretty difficult if there isn't someone on board with a match.

Question remains: how difficult is it to put an igniter in an engine?
Since turbopump no not near the engine bells, igniting that "engine"
requires a spark generator that is already in there, doesn't it?

Do these rockets really requite sparks that are created by GSE? If so,
what it is mean when T reaches 0? The rocket sends command to ground
equipment to generate the sparks? What is so difficult about igniting
the engine from within?

Nope. Since there's no propellant for it to combust.

Surely, they can do like SpaceX, and return launchers where there is
spare fuel, and expand the ones that push the limits and have no fuel left?

What is so different with Falcon9 that allows it to ingnite its own
engines that nobody else can do?

Designed for restarts from the beginning. Most second stage engines
are also so designed, plus the shuttle main engines. I don't know if
SLS has retained that capability, but Wikipedia suggests that the RS-25
has it still ... dual-redundant spark igniters in the injectors of the
preburners.

Merlins start with a canister of ugly goops (TEA and TEB) that go
bezerk when mixed. Raptors use spark plugs to light torches that then
light the preburner and the main chamber.

Vulcan will use BE-4 engines, which are supposed to be restartable
during flight. Head pressure restarts the turbopumps, but WP doesn't
explain what igniters are used.

Blue Origin's demonstration of powered landings with the BE-3 engine
may have been too late in the Vulcan design cycle to give ULA
confidence in the approach. The BE-3U variant is going into the
Vulcan, though, as the upper stage. New Glenn will use the same
combination as Vulcan, it seems.

/dps

--
"That's a good sort of hectic, innit?"

" Very much so, and I'd recommend the haggis wontons."
-njm