Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[flameaway]

Since the TinyLoader failed miserably when operating at speed, I was forced back to the drawing board.  I'd like to report some limited success with a new loader design. (It's new to me, if someone has already done this, just think of Leibniz and you'll feel my pain.) I've been significantly more careful to do appropriate testing this time around. The first optimistic TinyLoader fiasco taught me that much at least. 

Pity the amateur virtual scientist.

Here is the intial trial data on a new inline loader:



This shows data from a wooden cart being accelerated by a 10 track ramp strip into a one z-level hole that is the first tile of the loader.

The numbers in the first three postions from the left are the time elapsed from when the cart enters the loading level (from above in all cases) to various landmarks in it's journey. To the middle right are the various water levels in each position within the loader.  The last two columns give acceleration and z-levels dropped during the trial run.

Same machine using a lead mine cart.



Same machine dropped from two z levels instead of accelerated.



Here I succesfully load two carts inline. (I've modified the original machine to increase flow inline with the pump. The data I've supplied indicated that water supply was a choke point in the original three tile aquifer access design used in the first machine.)

http://mkv25.net/dfma/movie-2520-twocartsloadtrial1

Same machine different result:

http://mkv25.net/dfma/movie-2522-twocartsloadingtrial2

You'll notice, from this movie, that the flow from the pump pushes the first cart out of the loading position too fast for it to load.  The second cart loads correctly.

Here is a new machine, that is exactly like the first except that the aquifer access has been increased from the side (now four tiles of flow) instead of directly inline with the pump.  I'm hoping to avoid that early push this way. (using wooden carts)

http://mkv25.net/dfma/movie-2523-twocartloadingtrial1

Another trial with a step by step identical result: (Strangely enough, during these two trials the two mining carts occupy the same postion in the loader for about seven ticks.  Anyone ever see more than one mine cart in a tile before?)

http://mkv25.net/dfma/movie-2521-twocartsloadingtrial2

In the final two trials, the last ones I've done to date, both carts loaded within 13 ticks -- that's total elapsed time. 

This completes the intial testing for a new loader that appears to work very quickly (three times faster than a traditional trench loader using rollers) and does so succesfully inline with the gun barrel. 

Next up:

I will be building a four level gun that will accelerate mining carts to the max allowable speed, fire and recycle the cart/s into the flow loader.

I'll post an update here in the next couple of days.  Succeed or fail.

Flow Loader, quick explanation:

This loader uses flow (in this case provided by a pump) to:

1) load a cart
2) push the cart from the loading track ramp (direction ramp) to the next ramp (acceleration ramp)

The cart drops into the loader, loads, then moves south past an acceleration ramp, across a floor grate and into the gun barrel (very short and oddly configured in this test model.  This configuration was chosen because my primary concern for the gun barrel was to provide enough momentum to unload the cart (for the next trial) when striking the blocking cart to the south.

This Flow Loader uses 15 power in this configuration (the only one tested to date), one pump and one gear assembly hooked to a lever which is used to turn the loader on and off. (On/off is not strictly necessary if you are careful about drainage)

For the Future:

I'm hoping, since this loader is flow fed, to eventually setup and run a loader directly from the flow provided by a brook.  This is in pursuit of my stated goal of developing a power free Infinity Gun.  It is an interesting possibility that I might be able to add another ramp (if necessary) just north of the floor grate in the loader. This will give me a three tile long trench (not counting the floor grate) in the loader.  I'm hoping to be able to place a water wheel in that spot, (the mining carts will be loading and running right through it.)   

Initially, I'll have the flow loader hooked to an independant power supply in order to start the pump. Then, once flow is established, I'm hoping the integrated waterwheel will provide steady power to the flow pump, enabling me to disconnect the loader from the independent power supply.  This will mean a gun that is a net power producer. *EDIT I tested the waterwheel idea.  It did not work.  It appeared that the water wheel simply recycled the flow directly back to the aquifer somehow.  The grate to the south never activated and all the aquifer holes stayed locked at 7/7.  The wheel continued to show as active. END*

Getting two carts to load effectively was a big initial step.  Now, I need to move on to a full size gun and do some more testing.  Including finding out how many carts I can load into one gun, and if my strategy of putting the loading process inline at just this step in the firing process will obviate any jamming/unloading problems, because carts are gravity fed and simply stack above the loader and wait their turn.

Once I get a working model, I also want to fiddle around with gun barrel designs.  I want to see if I can effect how far a thrown cart wanders off a straight course along it's journey to the edge of the map, just by manipulating the gun barrel variables.

Finally I want to figure out a good way to do away with the blocking minecart and the fortified wall.  It's not a completely reliable solution(pass throughs and teleportations), and I think the fortification causes the succeeding shots to spray out rather randomly in a fan shaped pattern.  Might be nice to be able to throw a block of water just as straight as a cart...


Flow Loader Detail (small gun barrel also shown - last three southern eastern track ramps):

Code: [Select]

AQ AQ AQ
AQ PU **
## PU ##
## SE ##
## SE ##
## GR ##
## SE ##
## SE ##
## SE ##/code]

AQ = Aquifer access
PU = One half of pump (pumps from north)
SE = SE track ramp
GR = Floor Grate
** = Gear Assembly
## = Wall