Very neat idea!

I'd love to borrow the idea to demonstrate Proto:87 Wheels running reliably!

Is your "cycle" the 40 seconds it take for the box car to return to the start point and begin again?

Rock & Roll is right!

Very groovy display idea and well executed!

I like how the grasses near the track are similar to the brushes that stick up to stop the cars.

I did get a little seasick watching things roll baaaaack and forth, baaaaack and forth.
Very much pity the dude up in the tower!

Very Cool!

That's a great demo table Tim, and what a testament to the reliability of Fast Tracks track work!

I can imagine it was quite a feat to build and synchronize everything.

At first I was wondering how the heck the car was coming to a nice controlled stop.  Then I saw the brushes.  You guys are the bomb!

Thanks for sharing the video, very neat!

Man you gotta love Fast Tracks!

I remember ...

I remember seeing this quite a while back. I cannot recall when or where, but it was a video and I watched at home online. Maybe while viewing a review of the 2006 or 2007 shows ??

Fixing the braking brushes to the base frame was brilliant

No moving parts or mechanism to worry about controlling. Love it!

Ingenious, clever

Ingenious, clever

That's five 9's

50000 rolls through the trackwork without a single derailment is 99.999% reliability - five 9s. That's considered absolutely excellent. Amazon, for example, only guarantees an uptime of 99.8% - which means over 100 hours of downtime in a year.

It would be interesting to do the same test on some commercial turnouts to see how long they would go without a derailment. If you got 10 derailments in the same period of time, that would be 99.99% reliability.

Wouldn't that be 100%?

50,000 cycles with out any derailments would be 100% wouldn't it?....

While the cars and turnouts operated flawlessly, we did have some issues with the micro switch settings.  They were quite fiddly to get set correct, and sometimes the table would go a bit hysterical.  After the Detroit show we decided to retire it instead of re-working it as it would have been another bit project to get it fixed.

Very Neat.

I love the automated table.  What I find even better though is the amount of detail that went into finishing the module.  You could have stopped at just the track work and the technical bits, but instead went on to create a nice scene complete with buildings, vehicles and a crossing tower.  Outstanding work as usual.

Everything works 100% until it breaks. . .

It's  that "lies, damned lies and statistics"  thing.. . . . .so you have to be careful to double check your math and be very clear and sure about what you say you claim.

"During the three days of the train show it ran non stop, flawlessly.  For 50,000 cycles that boxcar rolled through the double crossover without a single derailment.  The four turnouts in the double crossover were all switched back and forth 50,000 times without any issues, pretty much proving that a properly soldered switch point on a PC board is very reliable".

For instance, the cycle of the box car going through all four turnouts, flipping each turnout just once, takes 40 seconds on the video..

To do that 50,000 times, actually would take nearly 3 weeks, running 24 hours a day, not three days. So presumably the test was run that long before you took it to the show.

The counter under the table is stopped at 50,278

Prior to the show we ran it non stop for a very long time to be certain it would perform as we had hoped.  It was rocking away in my shop for days.  To be sure it was a fair test, the car used was just one of the pieces of rolling stock I used on my layout, the only thing done to it was to confirm the wheels were in gauge.  We brought along a few cars to use on the display, but as I recall we only ended up using one.

I'm talking mathematically speaking

Actually...

Actually 500 of 501 is 99.8% not 99.9%.

50,000 reps is impressive. But I find that free rolling cars -- neither pushed nor pulled -- are easier to keep on the tracks.

A more rigorous test, at least in my opinion, is backing a 30-car train though a section of track at 60 scale miles per hour. Then repeat for the other direction. If that stays on the rails, then I'd call that piece of track bullet proof. However, be sure to use equipment with properly gauged (and I mean in the *center* of the wheel gauge cut outs on the NMRA gauge, not just that sort of fit in) wheels, nicely swiveling trucks and properly installed couplers (no dangling "glad hands" to catch on turnout closure rails.

And it's cheating to pick 30 cars that are all the same length and weight. Use a representitive mix of the cars operated on your layout.

But 50,000 is pretty darned impressive.

Charlie

Not sure about that 5 9's...

Joe-

If you assume 50,000 good cycles, then 1 derailment, then 50,000 good cycles, then 1 derailment, etc. that's 5 9's.  But if you're going to be assuming things about cycles that haven't happened yet, what's to say that there won't be 50,000 bad cycles in a row, making for 0 9's average?

I think the only reasonable claim to make about 50,000 cycles with no derailments is "wow".

- Ken

(Why do I have this strange urge to put my layout on gimballs now?)

No gimbals needed, Just use your hands

I don't have the stamina to hold it up for 50,000 cycles , but OTOH, no critical micro-switches needed!

Given the absolutely crazy 8" radius, combined with scale near 75 mph speeds, even though using real proto87 wheels, made me feel quite satisfied with the reliability of the glued technology. 

And I hate to be thought "pushy" , but we just "glued" this rail down without taking any special precautions in just two evenings in 2004 and shot the video.

Of course the video has been up on the Stores web site ever since. (Does that count? ). But the board is still sitting in good shape on the top our "museum" cabinet, so I could probably repeat and re-shoot the video in 1080 HD, even though it's now 8 years later.

Full details of the "construction" and standard box cars if anyone is interested is at:

http://www.proto87.com/HO_model_railroad_operations.html

Hmm, watching that animated

Hmm, watching that animated picture above from Andy....

4x8 layout...4 heavy duty springs on each corner...I could run my trains without using any power by simply leaning on the table! 

Tim has set me thinking about adding 4 servos. . . . .

. . . . And I have a spare AVR controller board looking for something to operate. Might be a better application than using them for switch machines. And my board is lightweight enough!

. . . So many whirls, so little time.