Kato M250 10-565 DCC Digitrax DZ125

[CaptOblivious]

Not late at all! I generally just stick with the stock resistors in these cases, but you do make an interesting argument. Another option is to simply replace the existing resistor with one the same value, but with a larger power rating—20mA @ 12V is right at about a 1/4W, so putting a 1/2W resistor in place would ensure nothing overheated or caught on fire!

 

And, you can empirically estimate the forward voltage drop of an LED, although it is tedious. Wire an LED and pot to a 12V source, with a ammetter in series. Start with the pot at maximum resistance, and adjust the pot until you get a 20mA reading. Finally, read the resistance on the pot. Use a resistor of no smaller value in your installation.

[rankodd]

Aaaaand this is why I decided to forgo DCC on my layout. As Jeremy Clarkson puts it "Hands of ham and fingers of butter" = LOTS of blue smoke out of expensive and difficult to repair parts :(

[gavino200]

On 2/19/2010 at 11:00 PM, inobu said:

I have edited the previous entries in order to provide a better flow of this install. I had a lot of issues and some rework that may leave the reader confused so I am updating this entry to give a clear picture.

 

Inobu

 

Kato M250 10-565 Super Rail, Digitrax DZ125 Decoder

 

Break down the 250 as seen here.

 

[smg id=830]

 

The light conversion: based on this http://www.katomodels.com/hobby/dcc/dcc_tips/light_kato.shtml

 

The hardest part of this mod is the de-soldering of the LED components. The PCB is thin and the leads will lift off. My first one took a beating and now is heat up because of poor connections. I'm going to have to rebuild it. So read up on unsoldering surface mount components in order to avoid this problem. Make sure you test the LED for heat. High heat could mean a poor connections somewhere.  

 

Added note: Have been monitoring the LED and they seem to run hot. I'm not sure why but I want to check into it. I have to re-evaluate the circuit components. I'm not sure if the 271 resistor is correct based on a input voltage of 13.9. I wonder if this setup (271ohm) is based on another input voltage. Ran temperature check and saw 104 degrees on the board. The original board runs at 80 to 90 degrees. I believe that resistor value is incorrect and may be causing the higher current draw.  .  

 

Lets start,

 

Remove the cab from the lead cars. You will see the light board 6018. It slides out with the brass pick ups. The first step is to remove them. You need a larger iron because the smaller tip cannot heat the solder fast enough. It just causes massive heat soak and will lift the copper traces off the PCB.  recommend to use a larger iron for the removal of the components but Not On Install.

 

[smg id=813]

 

Once you remove the pickups you need to lift the LED up so you can access the 561 resistor.

 

[smg id=814]

 

[smg id=815]

 

Heat up the solder and flake the resistor off. be quick about it or the heat soak will get you.

 

[smg id=816]

 

Now that the 561 is off you have to install the 271 resistor in its place.

 

[smg id=817]

 

In this step we are going to separate the LEDs into two circuits. Take an Exacto knife or something break or scratch a small section of the copper trace. This will divide the leads in two giving you connection points for the decoder. This is a little different from the instructions given but it seems a little easier this way.

 

[smg id=819]

 

The last step is flipping the LED so the anode or + side is correct. The little side as you look into the LED is the anode this is the blue wire side. You have to flip the LED so it is in this configuration. Test it by connection the decoders red and black wire to the track and connect the blue, white and yellow to their location on the board and it should work.

 

Repeat these steps on the rear cab.

 

Inobu  

 

Inobu, any chance you still have the pictures from this thread. I'm thinking of taking the plunge with an M250. If you were to do it again would you do the conversion the same way? Same decoder?

[inobu]

I've started to place the images back into the thread. I believe the links were lost during the server migration.

 

Inobu

[gavino200]

39 minutes ago, inobu said:

I've started to place the images back into the thread. I believe the links were lost during the server migration.

 

Inobu

 

Thanks Inobu. Much appreciated. Anything you'd do different if you were to do it again?

 

I'm thinking of experimenting with the drill/tap/screw method of connecting to an engine block.

 

I've also considered using a tiny neodynium magnet to make contact with engine blocks. Solder wire to magnet. Magnet sticks to block magnetically. See any reason that wouldn't work?

[kvp]

One small remark about the leds: The light boards are made for 12V, but Kato models are also 16V tolerant. DCC track voltage should be 14V for N, but decoders have bridges in it, so this drops the voltage down to about 12.4-12.8V. Leaving the stock resistor in place should work just fine. For wires, i would say a stock 7 wire heatshrinked digitrax decoder (dz1xx series) should be ok as 3 wires go to the lights, 2 to the pickups and 2 to the motor block.

 

ps: A simple trick i found useful for brass pickups on Tomytec models is to isolate the strips with tape, and add two small ones on top of the isolated part with either double sided tape or another tape wrap. The ends of these strips could be turned up and soldered to the decoder motor leads and the strips will contact the motor pickups instead of the original strips under them without modification to the motor blocks. The chassis frame will hold the whole thing together.

[inobu]

it was easier to use the clip method than to do the drill method but either was is fine. Today I would try using the

ESU Lok Sound micro.

 

The magnet will heat up not a good idea. 

 

Inobu

 

 

[gavino200]

25 minutes ago, inobu said:

The magnet will heat up not a good idea. 

 

 

 

 

Jeff uses this technique for connecting LEDs for building lights. He says that it's tricky, but if you do it fast the neodynium doesn't lose it's magnetic field. I haven't tried it yet.

[inobu]

7 minutes ago, gavino200 said:

 

Jeff uses this technique for connecting LEDs for building lights. He says that it's tricky, but if you do it fast the neodynium doesn't lose it's magnetic field. I haven't tried it yet.

The LED use 20 mA a motor can draw 1000mA.

 

Inobu

[gavino200]

8 minutes ago, inobu said:

The LED use 20 mA a motor can draw 1000mA.

 

Inobu

 

Ah, I get it. You mean the magnet will heat up during use. I thought you were talking about during soldering (thereby losing field). 

 

You're tempting me with the Loksound micro :) I still haven't invested in a Lok programmer. 

Edited December 31, 2017 by gavino200

[inobu]

OK, I got most of the images restored and in order for the most part. 

 

You should get it now as you will be using it shortly. In the interim get the 4.0 sound decoders for you favorite engines. 

 

Inobu

 

 

[gavino200]

On 2/19/2010 at 11:00 PM, inobu said:

 

\

 

[smg id=819]

 

The last step is flipping the LED so the anode or + side is correct. The little side as you look into the LED is the anode this is the blue wire side. You have to flip the LED so it is in this configuration. Test it by connection the decoders red and black wire to the track and connect the blue, white and yellow to their location on the board and it should work.

 

Repeat these steps on the rear cab.

 

Inobu  

 

It looks like LED 1 has a small capacitor soldered in parallel with it, while LED 2 doesn't. I curious why that would be. Any ideas. On mine it's soldered piggy-back onto the leads of the LED itself. 

 

I'm hoping it's LED 2 that needs to be flipped. LED 1 might be a bit trickier because of the cap.

 

[inobu]

Its a diode that make the directional feature work. Used on DC.

 

When you are traveling in one direction LED #1 operates but LED #2 cannot because of the polarity of the track. When you reverse the track voltage LED #2 will operate because of the tracks change in polarity. The diode blocks the voltage in one direction. When I say operate its the correct color matching the direction of travel.

 

Because the decoder is providing the power you have to insure that the anode and cathode are wired correctly and you are creates two independent LED circuits.

 

Inobu

 

Edited January 29, 2018 by inobu

[gavino200]

5 minutes ago, inobu said:

Its a diode that make the directional feature work. Used on DC.

 

When you are traveling in one direction LED #1 operates but LED #2 cannot because of the polarity of the track. When you reverse the track voltage LED #2 will operate because of the tracks change in polarity. The diode blocks the voltage in one direction. When I say operate its the correct color matching the direction of travel.

 

Because the decoder is providing the power you have to insure that the anode and cathode are wired correctly and you are creates two independent LED circuits.

 

Inobu

 

 

Interesting. I thought the directionality of the LEDs themselves made the DC directional feature work. They are themselves diodes after all. 

 

I'm puzzled as to why only one of the LEDs has the component in question and not the other. Also, I'm not sure why it would be in parallel if it were a diode. 

[inobu]

Looking at the train traveling west.           (Front train )LED1 "White" ON LED 2 is "RED" OFF ------------ --------------- ---------- (Rear Train) LED 1 "White" OFF  LED2 "RED" ON

 

The diode does not allow for the reverse voltage to hit the LED when ever the track switches polarity.

 

So you will always see the combination of LED 1 - LED 2 on.

 

Never LED 2 LED 2 or LED 1 LED 1. 

 

Inobu

 

 

 

 

[kvp]

On 3/23/2011 at 10:18 PM, inobu said:

that appears that the 561 was replaced by a 271 so I followed it.

 

I found that wiring and reusing the LED are a crap shoot because you cannot identify the specs for the LED and the output voltage can be a number of values. Without the exact information it is hard to find the optimum resistor value to incorporate in the board. A built-in LED has 3 volts in the leads where as the function leads have 13 volts and is depending on the manufacture. I think the board was heating up because the resistor value was not the correct one for my Digitrax decoder.

 

The forward voltage of a LED is mostly dependent on the material composition, so basically its color. The forward current is what varies. Considering these leds were wired for 12V DC and N scale DCC voltage is 14V, the decoder's diode bridge leaves around 12.4V on the blue (light power) output. This means the same resistors used for 12V should be used. Also, if you change to a different resistor because of higher voltage, always use a bigger one and never a smaller one, like above. In this case, just leave the original resistor there.

 

For leds, the back to back configuration of DC mode doesn't need a protection diode as it uses a trick that if one led is always lit up, then it will protect the other one from excessive reverse voltage. LEDs have very low reverse voltage tolerancies (usually at most 5V) and comparable forward voltages (2-4V), unlike normal diodes that have very low forward voltages (0.3-0.7V) and very high reverse voltages (several hundred volts). This means if a lone LED is used in a light board, then you have to add a parallel protection diode in the reverse direction (some people would use a series diode, but that increases the turn on voltage, which is a bad idea in DC mode).

[inobu]

It's a problem when posting MOP's (Method Of Procedures) as the users can have varying setups which can have different outcomes. Layouts can run anywhere from 14 to 16 volts and the issue starts when you run your train on someones layout.

I started using 1k to be on the safe side.

 

I think the diode is there because of the red/white LED configuration. The white LED was getting spiked by the RED I think. 

 

Inobu

 

 

[gavino200]

I just finished the installation. I took me about 7 hours (front cab only), and was painful. The train runs great, so all's well that ends well. But literally every single step of this installation is hard. 

 

Regarding, the lights, I left the stock resistor and I didn't bother replacing the smd component that's parallel with the white LED. It came off when I flipped over the LED. The lights look and work great. 

 

Thank you so much Inobu, for making this amazing guide!

Edited January 30, 2018 by gavino200

[kvp]

11 hours ago, inobu said:

I think the diode is there because of the red/white LED configuration. The white LED was getting spiked by the RED I think. 

A parallel diode there would open first and would cut power from the white or the red led, depending on the orientation. A series diode would have to be added to the red led to bring the forward voltages in line. (doesn't really matter though as long as the reverse brakedown of one is higher than the fowrad opening of the other in both combinations) Adding a very small smd cap on the other hand could help. Anyway, neither is needed for DCC as the decoders have filtering and low active outputs.

[The Next Station Is...]

I've just completed a DCC conversion of the M250 myself. After seeing someone in Japan having done it, I've modified Kato FL12 decoders to drop in the cabs where the DC lightboards normally live. I also used a Digitrax DZ126T for the motor block, following Inobu's method with the motor block cap but running the copper strip to the pickups at the end of the motor block. My ethos is to try and make DCC installs as unintrusive as possible, so I'm grateful for the previous guidance in this thread.

 

I've put a couple of photos of my conversion on my blog at http://tsugi-wa.com/post/175349709464/in-my-last-post-i-shared-a-video-of-part-of-my-dcc