Sorry for not getting back to yet! I’ve have been looking at this and analyzing it. I haven’t come up with a fix for it yet which is why I’ve been slow to report back.
I pretty much understand why it is happening and it is essentially related to the known issues around playing back from a buffer that is simultaneously being recorded. Most people have reported this issue as a sort of clicking sound which normally happens when the play head (from a sample player) collides with the record head (of a sample recorder) because the play head is either going in the reverse direction or it is faster/slower than the record head. If the play head and record head are going in the same direction with the same speed then typically you would not expect them to collide and hence there would be no problems.
However, in your case, static-like sound is coming from the fact that your punch-in timing is pretty good so that you have the record head right on top of the play head, continually clobbering it as they both move along together writing and reading the same spot in the buffer.
Based on the above, there is a workaround (more like a hack). The goal is to get the play head away from the record head even if just a little bit. Now, I assuming that you are using the same gate from the Digitakt to reset both the player and the looper. The temporary workaround is to insert a uDelay (Micro Delay) in the player’s trig sub-chain and set the delay to 2 or 3ms. This will cause the play head to lag behind the record head by 2 or 3ms. That means the play head will be behind the record head and thus include the newly dubbed material as you are playing it.
If you instead put the uDelay in the reset sub-chain of the looper then the record head will lag behind play head. This will also fix the static-like artifact but now the playback will not include your live dubbing but rather play the undubbed sound since the record head will be writing its new material behind the play head. You can choose where to put the uDelay based on which behavior you would like.
A proper fix for this type of shared buffer scenario is coming but it requires some fundamental changes in the internal signal routing and DSP scheduling which should happen sometime in v0.4.