Kato JR 500 10-510 DCC Digitrax DN163

[inobu]

Inobu, does it do metric, or only mph?

I'd seen this before, and I'd been thinking about it. Right now I use a stopwatch. But there's very little about it online.

It only does MPH but it is only a perceptional number. Meaning that 60 MPH is around 100 KPH so it is just about realizing the translation.

Dialing the speed tables are so much easier.

It uses 2 sets of IR LED beam one on each end of the tunnel. The train breaks the beam which starts the counter. When the nose of the train breaks the 2nd beam the counter stops and calculates the speed. You can enter the tunnel from either direction and it is wide enough for a double N Scale track.

I like it because you can dial in and operate at true proto speeds. You will be surprise at how fast we run our units.

I did have a problem with mine. One transmit LED died. Instead of sending it back I went to Radio Shack and installed a new one. I contacted the company and told the guy about what had happened. He stated that he uses a different manufacture now.                                                                                                                                                                               

Inobu


     

[KenS]

I like it because you can dial in and operate at true proto speeds. You will be surprise at how fast we run our units.

I wouldn't be. Now.  The first time I measured mine with a stopwatch I was though.

The reason I asked if it did metric was that there's this one that does do metric.  It's intended for panel mounting, and apparently uses a photocell and ambient light.  They also make a four-channel one.

I've been thinking of getting some kind of speedometer (preferably not a panel-mount), but if I'm going to do that, I'd prefer one that read in kilometers.

[Webskipper]

http://www.ulrichmodels.biz/servlet/the-490/Speedometer-for-HO-and/Detail


Inobu

This is good stuff for a new thread.

Is that 62MPH already adjusted for the scale? Otherwise the loco may be showing 62mph but, experiencing conditions that of 9920MPH. Like boating up river.

Still far off from the cruising speed of 186,282 MPH, of light.

[Webskipper]

So which DCC decoders are the favorites for the JR500?

DZ125 and FL12?

I have a JR500 and 800 Tsubame to DCC so there are lots of car lighting to do as well.

[quinntopia]

Inobu - I want to thank you for sharing this really easy technique...I just added a DZ125 (crossing fingers I don't have the kind of luck that CaptO has had with these, although I did just have a DZ125IN 'die' on me for no apparent reason) in about 10 minutes using this approach on my new Kato Sunrise Express (nearly identical chassis).  Love the technique with the heat shrink tube to isolate the motor contacts!   Brilliant!

[inobu]

Quinntopia,

That's the best thing about a good forum. Everyone achieves success.

INobu

[Martijn Meerts]

Coming back to the speedometer, I've been thinking...

Several of us have an iPhone/iPod touch with camera. With iOS 4, we now have access to the raw camera data while it's being captured. With that, one could use various ways of detecting things.

The idea I had was that you place 2 colored strips on any location in the layout or test track, a certain distance apart. Using the iPhone/iPod Touch camera, you detect those colored strips. If strip 1 gets obscured by a train, the camera will loose track of it, signaling the app to start the timer. Once strip 2 gets obscured by the train, the counter stops. From there you can basically do any calculations on the data you want.

Taking it a step further, one could use multiple strips to calculate acceleration and deceleration as well.

The disadvantage is the need to specify the colors to detect, and they need to be pretty precise. Of course, what would be possible is that you just tap on the "start color and "end" color to register those colors in the app. I actually have code examples of detecting and tracking colors in a video feed, so it might be worth experimenting with.

The data would be quite accurate actually considering everything can be done hardware accelerated. With only simple detection on a lower resolution camera stream, you can easily do 60 fps, which should be enough for accurate speed detection. It would require a more recent device. Not even sure an iPhone 3Gs would work, depends a bit on the camera hardware...


Could be a fun little experiment though ;)

[drobnak]

inobu, could you take a foto how you modified the circuit board for th front and taillights, I am in the process of doing the same thing and honestly I am lost on how to modify the board in order to isolate the red and white LED's from each other.

[inobu]

I cant post pictures because I took the light off.
I did not like the job I did on them so I pulled them out. I wanted a setup up that did not produce too much heat and have not got around to replacing.


This is what I used.

Go over this process until you really understand it and then try.

First we will go through the basic LED wiring flow.

Starting with the
Common + side from the power supply it is connects to one side of a resistor (which reduces the voltage and current flow for the LED)
The output of the resistor is connected to the Anode + side of the LED.
The Cathode - side of the LED is connected to the return side of the power supply.

Now lets look at the decoder's wiring and match them up.

The common side is the blue wire from the decoder.
The return wire is the Yellow or White wire to the decoder.  

Because this was a DC power board the LED's cathode opposed each other. Remember the polarity changed on the track which switches the current flow and only one LED will work based on the position of the Anode.

In DCC the voltage polarity is not being switched. Power is being supplied by two different power sources (White and Yellow leads). Therefore we need two separate return leads that are isolated from each other.



Looking at the diagram you can see the removal or scratching of the traces (Red X )isolates the track power (Red and Black) from the LED's.

Remember the flow?

Blue wire, resistor, Anode, cathode, return.

You can see that the Red LED is fine as it is. The clear LED needs to be modified.

The clear LED need to be flipped so that the Anode is attached to the resistor/blue side of the board and the white wire is tapped onto the cathode side of the LED on the bottom of the board.

This is where you can run into trouble. You need to get the white wire to the cathode on the bottom of the board. This is added work. I will show you a better way to
scratch the traces.

More details...

This is really harder but yields a better finish.
On the modified drawing I split the LED runs horizontally so we can use the existing surface material. Desolder the LED first so you can carve the runs.  

Use a sharp exacto and just carve a thin but clean path to isolate the runs. This way you can keep everything on top. Don't forget to flip the clear LED when you solder it back.

The key to success if in your soldering technique. You need to have clean tip and solder fast. The traces are thin and can take only one or two attempts. Anything more then the traces will pull off the boards and you will have a mess.

Inobu