A different way to side load

[Trebor]

SideLoad.gif (34.16 MiB) Viewed 604 times

[computeraddict]

Neat sushi loading. Have you done any testing on how it behaves in different consumption and filling regimes? I.e. does it self correct if input is resumed or can it deadlock if it's starved of one input for too long? It looks like it would probably be fine if the input is restricted to a half yellow belt of each.

[Trebor]

This is not so much about the sushi, but how the upstream lanes are both used.

[Trebor]

Note traditional side loading messes up even upstream of a lane balancer.

Upstream.gif (74.74 MiB) Viewed 532 times

[computeraddict]

Note traditional side loading messes up even upstream of a lane balancer.
Upstream.gif

It messes up upstream of a lane balancer that isn't input balanced. That balancer is only output balanced. It won't draw evenly from its input. If you use one like this it won't matter what happens to the output:

[Tertius]

As far as I see it, the examples simply show regular sideloading. Rather, the examples visualize (and prove) how splitters work with two fully compressed inputs.
However, this perfect alternating between left and right input stops working as soon as gaps appear.

The topic name suggests some not widely known or interesting sideload technique is shown.

This is an example of what I mean. It shows how to use a splitter for sideloading to fully consume a blue belt in a confined space. Usually, the belt-to-chest speed of a stack inserter suffers greatly as soon as one or even two stack inserters already made gaps in the input belt. But this is a method to avoid such gaps, avoid the inserter having to change lanes (which kills their pickup performance) and still feed the whole 45/s.

2024-02-28 13-37-36 start=1 length=2.50 fps=60.gif (3.11 MiB) Viewed 472 times