Quarry Track Project

Quarry Track Project

Postby Phauss » Sat Jun 11, 2011 12:59 am

Here is the discussion thread for one of my projects, the Quarry Rail System. Cody has been helping me out with some wiring ideas, and maybe some of you have a few thoughts that might help.

Image
The main shaft. (thumbnails until I fix the image resizing code)

Image
A standard station.

Here's the project:
A rail system inside a vertical shaft mine that has a cart on every station, that when a station button is pressed, all carts advance from their station to the station below, with the bottom level cart returning to the top.

Currently:
There is a vertical wire running down the shaft that connects to all stations, so that when a station button is pressed, it sends a signal to itself and all stations below it to send their minecart to the next level down. Testing with minecarts at all stations has been successful when the button is pressed at the top station.
Image

What needs to happen:
Mine cart stations need to send a cart down regardless of the level where the initial station button is pressed. This means that any station other than the top one needs to send a signal to stations above to get them to send their carts down a level and keep the system fluid.

Attempts (displayed via simplified circuit diagrams):

Image
At the bottom station, a cart goes over a detector rail after leaving the station, sending a signal to the top of the down wire, which sends a signal back down again to all stations to release their cart.

There are two reasons why this won't work:
  • The stations are actually told to release their cart twice. In a perfect world, the second time the signal is sent through the down wire, the carts that were released on the first signal would not have reached their stations yet. However, stations are too close together and carts arrive by the time the second signal comes through.
  • There is a .1 second delay every time a signal travels through a redstone torch or repeater. With one torch or repeater, the latency within the circuit is minimal, but because of the nature of the up and down wires, both the up and down wires have easily 20-30 torches each. What this means is that instead of each cart leaving the station simultaneously, there is a wave of carts that progresses down the shaft as the signal travels down. It takes just as long for the signal to then come back through the up wire after the bottom cart leaves the station and start going through the down wire again. by the time it makes the full trip, carts are already in their station and they are told to move down to the next level, starting the whole process again.

If there was no delay within the circuit, this system would theoretically work just fine, although it would be a bit clunky and inefficient.

Image
This is essentially the same as before, except that a detector rail has been placed before the entrance to the top station to try to reduce latency within the circuit.

Why didn't this work?
  • Although it reduced ciruit latency, now resending the signal had to wait on the bottom cart to climb the incline to the top station. A holding station was added at the top of the incline with a detector rail at its exit to combat this problem, but...
  • ... the fact remains that using a detector rail to send a second signal down is too inaccurate and carts will leave their station when they aren't supposed to.

There were a couple variations of this, mostly changing the postition of the detector rail, but nothing really changed in the end.

Proposed Ideas:

Image
In this diagram, I plan on using repeaters to act as diodes at every station, so that when a button is pressed, it sends signals through the up and down wires concurrently, eliminating any problems coming from circuit lag. This is also a way of getting each station to recieve a signal only once per button press since the detector rail system has been removed.

Things to think about:
  • The wires themselves act as diodes, so it will be some trial and error as to where repeaters would be placed (around buttons, boosters, etc.)
  • Since the lag from so many torches and repeaters will still be present, I would llike to see how effective this system is at reducing false triggers. Theoretically, only the button press can send a signal in this sytem. There is the possibility of shorting and loopbacks if it isn't put together correctly.
We are each a beautiful and unique snowflake that will melt in hell.

I got the words "jacuzzi" and "yakuza" confused.
Now I'm in hot water with the Japanese mafia.
User avatar
Phauss
Site Admin
 
Posts: 1095
Joined: Thu Oct 01, 2009 10:30 am
Location: Here and There

Re: Quarry Track Project

Postby Phauss » Sat Jun 11, 2011 1:01 am

More on this later. I've got another failed attempt and a few interesting theories to add to the above post.
We are each a beautiful and unique snowflake that will melt in hell.

I got the words "jacuzzi" and "yakuza" confused.
Now I'm in hot water with the Japanese mafia.
User avatar
Phauss
Site Admin
 
Posts: 1095
Joined: Thu Oct 01, 2009 10:30 am
Location: Here and There

Re: Quarry Track Project

Postby Phauss » Wed Jun 15, 2011 1:36 am

I've been working on that schematic I posted at the end of the top post. Here's what I started with, then here's where I ended with it:
Image

The schematic on the right has been given to every station, including the top and bottom stations, and the holding stations have been removed to simplify the circuit. It works magnificently and all minecarts and stations do what they're supposed to. The only thing left to do with this project is mark signs with station levels.
We are each a beautiful and unique snowflake that will melt in hell.

I got the words "jacuzzi" and "yakuza" confused.
Now I'm in hot water with the Japanese mafia.
User avatar
Phauss
Site Admin
 
Posts: 1095
Joined: Thu Oct 01, 2009 10:30 am
Location: Here and There


Return to Minecraft

Who is online

Users browsing this forum: No registered users and 1 guest

cron