Train offloading

[Linosaurus]

I am tempted to say it seems to me that splitter logic must be seriously flawed for this to ever occur. However, I suspect there is something I am overlooking. If someone can express what it is, I would be very interested.

Splitters reliably divide inputs into outputs. They do not reliably pull equally from the inputs (as Hannu mentioned above). That's... just how it is, I guess. Of course if the output belts run at full speed it won't happen because all input is accepted.

Train stations are the only place I can think of where this matters. In other situations you either 1) are using more than you produce of a product, in which case you only care about splitters dividing the items equally among consumers. 2) Are using less than you produce, in which case the belts are going to get backed up all the way to the previous assembler/smelter anyway so it doesn't matter.

[Mr. Tact]

Since I start the testing with the factory at zero output and with the belts, chests, and wagons all being completely full, the extra length should make no difference. The issue is seemingly the uneven pulling of load by the inserters down the line. However, just to prove it to myself, I'll try moving the load balancers to the left...

[Mr. Tact]

In the image you provided a link for:

http://imgur.com/jrXTGio

The 2-8 ore train, how are you getting the six inserters to load the three full belts like that? I am not able to reproduce that...

[DaveMcW]

The six inserters are only loading 0.75 full red belts. There is a lot of redundancy in that design.

[impetus maximus]

i used splitters in the past for offloading, until DRY411S showed me this. [edit] credit to MadZuri [/edit] this setup will ensure that the crates are unloaded evenly so no chest is favored or ignored.

small.even.offload.jpg

wire all the inserters that are dropping to the belts together, with the condition *(everything) = 0 and Read hand contents enabled.
they will all grab together, then won't grab again utill all the inserters are empty handed.

since there is a gap after the last inserter in line, it will not keep the belts fully stacked. it is intended to be combined with other belts down the line.

i love how small a footprint, and simple this solution is.

apologies if my settings aren't the orginal design.

[Deadly-Bagel]

I agree with Mr Tact, the chests seem to be unloading at roughly the same rate but the further left (away from the merger), the more distance it has to travel and the more ore can fit on the belt, so at the tail end of unloading you see your leftmost chests run out first, then the next set, etc.

To get around this you need to have all your chests unload onto belts that are of equal distance to the first splitter. I would achieve this by having a splitter between carriages 1 and 2, another between 3 and 4, which are equally spaced from their two respective carriages in terms of belt length. These would output to 2 belts which then mirror each other to your 4-4 merger. Note that the two sides of a splitter don't need to be of equal length, but both sides need to match the length of their counterpart for the other two carriages.

To test, ensure that all your belts are empty then pull a train in to the station. You should see all four lanes hit the first splitters at the same time, then when they reach the 4-4 merger it should be symmetrical.

[Hannu]

The 2-8 ore train, how are you getting the six inserters to load the three full belts like that? I am not able to reproduce that...

These belts are not full except they are backlogged. There are more parallel parts in that part than needed, but it does not take more than couple of belts. The essential thing is that four blue lines on the right are always full. It is very easily fulfilled in this setup and 8 wagon trains.