New shower: leaking silicone joints -> water damage! New tiles, or what to do?

hello,
our bathroom was renovated a few years ago by an acquaintance. We have a brick shower enclosure with a plastic shower tray.
Now we had water ingress at the silicone joint between the pumice stone and the shower tray (see sketch). The pumice stone was previously also tiled, the sketch shows the status quo.
The probable cause of the leak is that the joint width was sometimes only 2mm.
In the meantime, I have scraped out all the joints, removed the tiles and the rubber coating from the horizontal pumice stone so that it can dry again.
How can I now make the shower permanently watertight?
My idea was to double the pumice stone, apply new rubber coating, tile again, so that the joint from pumice stone to shower tray forms a 90-degree corner, which is probably easier to seal than when the surfaces are at 180 degrees to each other as before, is that correct?
But if I double the pumice stone, I will also get more joints all around where the pumice ends against the walls, all of which I have to seal.
Or should I just completely tear out the old tiles, double the pumice stone, apply new rubber coating completely, and tile again?
I am totally annoyed but want to make it watertight this time for good, so I am grateful for any tips and hope what I wrote is understandable!
Cheers :)

Yes, this detail was executed poorly. In the area of this pumice stone, a fabric sealing tape should have been glued in and painted over – then the silicone joint applied afterward. These silicone joints are not actual sealing joints but serve more as a visual finish of a tub/wall connection – the seal should, as mentioned, lie underneath.

What could be done:
Tear off the tiles in the lower area, about 20 cm above the shower, remove the shower, glue in a fabric sealing tape into the corner where the pumice stone wall rises, over the pumice stone edge, and into the corner where the tub base meets the pumice stone; this is available as a system accessory for liquid sealants (rubber coating). Then paint the entire area from the tile demolition edge up to and over the corner of the pumice stone-tub base again with liquid sealant.
Then reinstall the tub and put a neat silicone joint over it.

This is how it should have been done – then the whole thing would still be tight. Experienced tilers also “gladly” skip the fabric sealing tape... “I’ve spread it well, it seals just the same.” But these fabric sealing tapes are absolutely necessary, as the liquid sealant (rubber coating) cannot form an adequate film thickness on sharp edges.

Correct execution?

Hello Hertweck,
thanks first of all for the quick response. A fabric membrane was indeed still under the tiles on the pumice, but it ended at the front edge of the pumice and was not connected to the shower tray, so water could easily get through.
I have one question about the connection to the shower tray: Was it meant as shown in the sketch?
That is, the pumice shaped with tile adhesive, tiles removed from the wall, then fabric sealing tape in the corner or connection of the pumice with the shower tray, then rubber coating over it.
Then a block of pumice placed on top, sealed exactly the same way, made and newly tiled on top.
The new pumice block basically only so that a 90-degree corner is created, which I think is easier to seal? Or is this unnecessary?
Thanks in advance, your answer definitely brought back my motivation.
Regards, Phlipzner

As shown in the drawing, it could be executed like this, but it would then need to be waterproof. The pumice stone (the upper block) would actually not be necessary if the shower tray could rest on the lower block and be tiled; with this upper stone, you get an additional corner that needs to be sealed – if you execute it as drawn, however, the pumice stone is not problematic, it could also be done differently (as with any connection detail).