Belt saturation detector/counter
Re: Belt saturation detector/counter
@disentius this is a nice balancer, also does lane in addition to belt balancing. I always overlooked this one and thought it was inferior because it seems bloated, but with this example I realized the lane exchange actually does something: mixed with the original lanes, it "computes" the mean value of both lanes, so they become balanced. It's of course better, because it balances the lanes of 4 belts with only 2 splitters instead of 8. Nice!
-
- Fast Inserter
- Posts: 192
- Joined: Mon Aug 31, 2020 10:09 pm
- Contact:
Re: Belt saturation detector/counter
THANK YOU SO MUCH, ANSWER AND FOR YOUR INTEREST @Tertius
There are 20 ovens:
45 i (ore) -- 54 i (plate)
In other words, 1 blue belt is insufficient, but 2 blue belts should carry 54 items easily. (However, the last three ovens give a yellow light, the output is full)
When the first 8 furnaces fill the near corner and the next 8 furnaces fill the far corner, approximately 1 (0.9...) arch is filled. For the last 4 ovens, I split the path with the distributor. There should have been a smooth exit with the second generation. But only 17-18 ovens can work.
If it is considered that there is not enough consumption: no it is not. I ship directly to steelmaking.
Steel production: 45i iron input, 10.8i steel output. In theory, 1 blue belt should be enough for entry. but 2 ovens red light.
In summary: why 2 straps are not enough for the iron? (54 i/s was enough for production though)
+steel production demands 45 i/s iron input, ie 1 full blue belt. but here, too, the iron input is insufficient.
theory practice mismatch.
There are 20 ovens:
45 i (ore) -- 54 i (plate)
In other words, 1 blue belt is insufficient, but 2 blue belts should carry 54 items easily. (However, the last three ovens give a yellow light, the output is full)
When the first 8 furnaces fill the near corner and the next 8 furnaces fill the far corner, approximately 1 (0.9...) arch is filled. For the last 4 ovens, I split the path with the distributor. There should have been a smooth exit with the second generation. But only 17-18 ovens can work.
If it is considered that there is not enough consumption: no it is not. I ship directly to steelmaking.
Steel production: 45i iron input, 10.8i steel output. In theory, 1 blue belt should be enough for entry. but 2 ovens red light.
In summary: why 2 straps are not enough for the iron? (54 i/s was enough for production though)
+steel production demands 45 i/s iron input, ie 1 full blue belt. but here, too, the iron input is insufficient.
theory practice mismatch.
viewtopic.php?f=33&t=105093
Beacon domain & Cable access area coloring (Color Tone variations)
Beacon domain & Cable access area coloring (Color Tone variations)
Re: Belt saturation detector/counter
It is not mine, it's from a balancer book by raiguard:
https://pastebin.com/igs2CvVd
( i am not smart enough to design these myself)
https://pastebin.com/igs2CvVd
( i am not smart enough to design these myself)
Re: Belt saturation detector/counter
In practice I saw 3 problems :
1) more output than input => require extra lane for output.
2) make sure steel receive enough iron => require balanced lane.
3) not having yellow furnace => require little buffer between furnace and output belt
In my example there are some furnaces that have 2 ouput inserter. They output in 2 different lane( the blue belt and the red belt). Because 1 lane is almost saturated. They have room to ouput in the other lane. no output blocked, this allow minimum number of furnace. If no yellow furnaces 3), then only using 16 furnace is enough for 1 blue belt of input.
The last furnace of the column has 2 output inserter but it's different. It side load on the belt at the end. This make sure that the last furnace is used to fill in the gaps. It has a mini buffer for output because inserter cannot fill in the gaps 100%. You need to sideload. The furnace 16 receive fewer ore because it's the last.
The output of furnace is balanced on the 2 side of the 2 lane because there is the same number of furnace used for each side.
Your method of doing the balance of the 2 side of the belt with the priority splitter is the same but it's not working for the last 4 furnace this cause all the little trouble.
You can add mechanism to balance the 2 side of the belt at the end of the column. But it will not solve all the problems, because the unbalance will cause some furnaces in the column to have output-blocked. This is why I made furnaces output in 2 different lane.
If you need 5.5 furnace to fill in 1 side of the belt, you can have 5 furnace output on 1 side. And the furnace 6 need to output on 2 side. This solve the yellow furnace problem. You need to have the 2 different output belt with the same number of plate on their 2 side. If not then 1 side will have little gaps and the other lane too many plate. In this case the furnaces that can output on 2 different side of lane will equalize.
-
- Fast Inserter
- Posts: 192
- Joined: Mon Aug 31, 2020 10:09 pm
- Contact:
Re: Belt saturation detector/counter
Is it really me or the game? I will be a 40-year prayer for anyone who has a logical explanation for that.
Even for a very short time, green card entry may be insufficient. (The red light is on!)
No road jam. The blue inserter is sufficient, the blue belt is sufficient, but there is still a problem! Could you please review the Blueprint and provide feedback?
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
viewtopic.php?f=33&t=105093
Beacon domain & Cable access area coloring (Color Tone variations)
Beacon domain & Cable access area coloring (Color Tone variations)
Re: Belt saturation detector/counter
This is true only when the blue inserter can work 100% of time.
In your blueprint, sometimes the first few inserter will grab green circuit and leave a hole in the belt.
This creates some pause in other inserter work because there is a gap in the belt and they lose some working time.
Replacing blue input inserter with green input inserter solve the problem because green inserters are better than blue inserter.
They can fill in the assembly in less time and so the little gaps are not a problem for them.
-
- Fast Inserter
- Posts: 192
- Joined: Mon Aug 31, 2020 10:09 pm
- Contact:
Re: Belt saturation detector/counter
thank you, it works. but max rate calc. and rate cal. according to both, 1 blue inserter should be enough, but the game is always weird...
viewtopic.php?f=33&t=105093
Beacon domain & Cable access area coloring (Color Tone variations)
Beacon domain & Cable access area coloring (Color Tone variations)
Re: Belt saturation detector/counter
Try connecting all the blue inserters together and let the belt fill. Set them to enable when anything > 0 and then last set one to read hand content / hold. With them all working in sync none should try to pick items in the hole created by a previous inserter.
Or the timing is really bad and they now always hit a hole from a previous cycle.
Or the timing is really bad and they now always hit a hole from a previous cycle.
-
- Fast Inserter
- Posts: 192
- Joined: Mon Aug 31, 2020 10:09 pm
- Contact:
Re: Belt saturation detector/counter
mrvn wrote: ↑Thu Nov 11, 2021 7:45 pm Try connecting all the blue inserters together and let the belt fill. Set them to enable when anything > 0 and then last set one to read hand content / hold. With them all working in sync none should try to pick items in the hole created by a previous inserter.
Or the timing is really bad and they now always hit a hole from a previous cycle.
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
viewtopic.php?f=33&t=105093
Beacon domain & Cable access area coloring (Color Tone variations)
Beacon domain & Cable access area coloring (Color Tone variations)
Re: Belt saturation detector/counter
Looks bad. So the later inserters do hit gaps. You could add a clock and time each inserter swing to happen at an optiomal time. Would take some experimenting and observing pickups tick by tick. Using faster inserters seems easier then.
Another thing to try is moving the power poles and spreading the underground belt part so the inserters pick up from a plain belt. Iirc inserters have more problems picking up items coming out of underground belts because they see it later and it can move to the next belt before they can catch up to it. That wastes time and the inserter becomes slower. The plain belt gives them the maximum time to catch up to items. Having to race after items still slows down the inserter but not as much as failing to catch the item altogether.
And one last thing to try: replace the red circuit belt with a half red/green circuits belt. Then you have a blue and red inserter picking up green circuits. This would actually give you more capacity to insert green circuits than you consume so you aren't working on a knifes edge. (same as switching to green inserters)
Another thing to try is moving the power poles and spreading the underground belt part so the inserters pick up from a plain belt. Iirc inserters have more problems picking up items coming out of underground belts because they see it later and it can move to the next belt before they can catch up to it. That wastes time and the inserter becomes slower. The plain belt gives them the maximum time to catch up to items. Having to race after items still slows down the inserter but not as much as failing to catch the item altogether.
And one last thing to try: replace the red circuit belt with a half red/green circuits belt. Then you have a blue and red inserter picking up green circuits. This would actually give you more capacity to insert green circuits than you consume so you aren't working on a knifes edge. (same as switching to green inserters)
-
- Fast Inserter
- Posts: 192
- Joined: Mon Aug 31, 2020 10:09 pm
- Contact:
Re: Belt saturation detector/counter
thank you.mmmPI wrote: ↑Thu Nov 11, 2021 11:23 amIn practice I saw 3 problems :
1) more output than input => require extra lane for output.
2) make sure steel receive enough iron => require balanced lane.
3) not having yellow furnace => require little buffer between furnace and output belt
In my example there are some furnaces that have 2 ouput inserter. They output in 2 different lane( the blue belt and the red belt). Because 1 lane is almost saturated. They have room to ouput in the other lane. no output blocked, this allow minimum number of furnace. If no yellow furnaces 3), then only using 16 furnace is enough for 1 blue belt of input.
The last furnace of the column has 2 output inserter but it's different. It side load on the belt at the end. This make sure that the last furnace is used to fill in the gaps. It has a mini buffer for output because inserter cannot fill in the gaps 100%. You need to sideload. The furnace 16 receive fewer ore because it's the last.
The output of furnace is balanced on the 2 side of the 2 lane because there is the same number of furnace used for each side.
Your method of doing the balance of the 2 side of the belt with the priority splitter is the same but it's not working for the last 4 furnace this cause all the little trouble.
You can add mechanism to balance the 2 side of the belt at the end of the column. But it will not solve all the problems, because the unbalance will cause some furnaces in the column to have output-blocked. This is why I made furnaces output in 2 different lane.
If you need 5.5 furnace to fill in 1 side of the belt, you can have 5 furnace output on 1 side. And the furnace 6 need to output on 2 side. This solve the yellow furnace problem. You need to have the 2 different output belt with the same number of plate on their 2 side. If not then 1 side will have little gaps and the other lane too many plate. In this case the furnaces that can output on 2 different side of lane will equalize.
viewtopic.php?f=33&t=105093
Beacon domain & Cable access area coloring (Color Tone variations)
Beacon domain & Cable access area coloring (Color Tone variations)