Author Topic: Skytop: Installing DCC and accurate tail light function Hiawatha Kato N scale  (Read 6918 times)

0 Members and 1 Guest are viewing this topic.

u18b

  • Crew
  • *
  • Posts: 4052
  • Respect: +2374
    • My website
We've talked a bit about it here and there.  But let's start a new thread.

I want to install an accurate DCC rear light in my Hiawatha Skytop sleeper.

Now, first of all..... fans out there....

What is it we need?

Running FORWARD, the tail light should be red and the function should be Mars or gyralight?

Then, when reversing, the light should switch to white?

Is that correct?

Ron Bearden
CSX N scale Archivist
http://u18b.com

"All get what they want-- not all like what they get."  Aslan the Lion in the Chronicles of Narnia by C.S.Lewis.

nkalanaga

  • Crew
  • *
  • Posts: 10577
  • Respect: +1755
0
If I remember right, and it was a long time ago, so I may not, the GN's standard for observation car lights was to use the small marker lights on the corners for normal operation.  The big rear light, Mars or Gyralight, came on any time they had an emergency brake application.

On the other hand, I have seen photos of trains, supposedly running, with the large light on, but the markers dark;  with both the large light and the markers lit; as well as the markers on and the large light off, so either the standard changed over the years, or it was never "standard" to start with.
N Kalanaga
Be well

milw156

  • Crew
  • *
  • Posts: 668
  • Gender: Male
  • Respect: +357
    • Modutrak
0
According to "Milwaukee Road Color Guide to Freight and Passenger Equipment" ..."the car's rear oscillating mars light was reversible, displaying red for normal operation, and clear for reverse movements". When the Olympian arrived in Seattle, it was after a backup move from Tacoma, so the rear car became the lead car so to speak. I also remember the (Morning) Hiawatha backing in to the Minneapolis Depot, after making the St. Paul stop, so both trains had some significant backing moves. Most of the pictures of the Skytops show them in active train consists, which could easily lead one to think the lens should be red, rather than clear. There were also windshield wiper s on the back window and those on each side of it for an added detail.

u18b

  • Crew
  • *
  • Posts: 4052
  • Respect: +2374
    • My website
+2
What I've been thinking about is replacing the red light pipe with a clear one.

Then install 2 micro LEDs side by side.  One red and one white.

Red would be forward on the white decoder wire.

The white would be for reverse on the yellow wire.
Ron Bearden
CSX N scale Archivist
http://u18b.com

"All get what they want-- not all like what they get."  Aslan the Lion in the Chronicles of Narnia by C.S.Lewis.

peteski

  • Crew
  • *
  • Posts: 35273
  • Gender: Male
  • Honorary Resident Curmudgeon
  • Respect: +6642
    • Coming (not so) soon...
-2
So that light can be either white or red?  That will make for an interesting installation. I like it!
. . . 42 . . .

Lemosteam

  • Crew
  • *
  • Posts: 6404
  • Gender: Male
  • PRR, The Standard Railroad of my World
  • Respect: +4817
0
What I've been thinking about is replacing the red light pipe with a clear one.

Then install 2 micro LEDs side by side.  One red and one white.

Red would be forward on the white decoder wire.

The white would be for reverse on the yellow wire.

Cool make sure you videotape that.  But how will the car know if it is going in reverse or not?  Would the decoder have to be matched to the locomotive decoder (forgive my ignorance if the answer is obvious)?
John "Lemosteam" LeMerise

u18b

  • Crew
  • *
  • Posts: 4052
  • Respect: +2374
    • My website
+1
Cool make sure you videotape that.  But how will the car know if it is going in reverse or not?  Would the decoder have to be matched to the locomotive decoder (forgive my ignorance if the answer is obvious)?

It will be just like a locomotive.
The car will not "know" if it is going forward or reverse.

I'll probably solve this by simply programming the skytop to the same decoder address as the locomotive.   For me (pulling with an EP-2, it is not a problem (one loco).  But for those of you using F units in lashups, you may need to consist.
Ron Bearden
CSX N scale Archivist
http://u18b.com

"All get what they want-- not all like what they get."  Aslan the Lion in the Chronicles of Narnia by C.S.Lewis.

u18b

  • Crew
  • *
  • Posts: 4052
  • Respect: +2374
    • My website
0
As I have done some preliminary reconnaissance on this project, here is what I see to be some of the issues involved....

1.  The stock board has a white LED.  This lights the upper side markers (which have red lenses) and light the lamps in the interior.  I want to LEAVE this stock.  It will always be on since DCC power is always on.  I don't feel a need to turn it off.

2.  Assuming "clear" means white as far as LEDs go, then we will need two LEDs.  One red and one white.

3.  The end light pipe is 60 mil and red.  Will need to be replaced.

4.  Space it tight.    LED options I have looked at are
       A.  cut the long light pipe after the room "lamps".  Put a nano LED on each end of light pipe where it splits.
see photo.



      B.  Cut the stock long pipe in same place but discard.  Mount the LEDs in the rear tail section, right up next to the rear pipe.
Space will be tight this way too.

Either way, I think I'm looking at 0402 nano LEDs.  Which also means micro-surgery as far as soldering goes.

5.  Decoder choice.  Size is an issue.  I think the standard DZ126 (which is moderately cheap) is too big.   A TCS Z2 might work.  Like them but expensive.  I'll probably go with the DZ126T (I think it is called) with the white wrapper.

6.  This is a loco decoder.  Which means based on my experience, you can't program it unless there is resistance across the orange and gray wires (where the motor would go).  Some folks on net suggest wiring a 12v light in that spot.  Will need to be small.  Also, I'll set Vmax very low to keep heat down in the bulb.   This also adds to the room situation.

7.  Decoder.  I know Kato makes a decoder for the CZ series.  The Skytop appears to be designed the same.  But I don't know anything about it and how the outputs work.

8.  Two nano LEDs will need resistors.  This means a resistor board.  So space will have to be found for it too since there will not be enough room in the end.

There are probably other issues, but these come up first.



Ron Bearden
CSX N scale Archivist
http://u18b.com

"All get what they want-- not all like what they get."  Aslan the Lion in the Chronicles of Narnia by C.S.Lewis.

nickelplate759

  • Crew
  • *
  • Posts: 3814
  • Respect: +1553
0
There are several small function-only decoders that might work.  TCS FL4  is one example (oddly, the FL2 is bigger...)
George
NKPH&TS #3628

I'm sorry Dave, I'm afraid I can't do that.

u18b

  • Crew
  • *
  • Posts: 4052
  • Respect: +2374
    • My website
0
There are several small function-only decoders that might work.  TCS FL4  is one example (oddly, the FL2 is bigger...)

George,

Thanks for the suggestion.  I have not used function only decoders before.

I looked up your suggestion.

The FL4 measures  9.12mm 14.1mm 2.92mm    (The 14 must be length)


The Z2 (more in line with the size we need) is 12.95mm. X 6.65mm. X 2.79mm    The 12 must be length.

Those measures are off their webpage.  Notice I see an inconsistency in the measures.   One of the first two numbers is reversed?

At any rate, the FL4 is 1 mm longer and almost 3mm wider.   Actually, I'm thinking the wider number is more important since I have considered mounting the decoder in the tank mounted sideways.

As a comparison.....

The DZ126 which is too big) measures:  11.56mm x 9.37mm x 3.2mm 

The DZ126T which will work measures:   14 x 7.13 x 3.25 mm

So based on my preliminary measures and experimentation, a width of 9 mm is too wide.

(By the way, all these measures are with the wrapper off I assume).

So unless there is a function only decoder that is close to the TCS Z2 or the Digitrax DZ126T, I'll stay with the DZ126T so far.
Ron Bearden
CSX N scale Archivist
http://u18b.com

"All get what they want-- not all like what they get."  Aslan the Lion in the Chronicles of Narnia by C.S.Lewis.

mmagliaro

  • Crew
  • *
  • Posts: 6489
  • Gender: Male
  • Respect: +2011
    • Maxcow Online
0
Do really need to go down to 0402 LEDs?  It looks like there is a lot more room than that near the split light pipes.
Is height the problem?  The difference in "Trickyness" between soldering an 0603 and an 0402 is huge.  That's why I ask.
Maybe solder magnet wires to the LED and then drill a hole into the plastic below the light pipe, and sink it in there?

Point353

  • Crew
  • *
  • Posts: 3722
  • Respect: +915
0
An alternate solution for the rear mars light would be to use a single R(ed)-G(reen)-B(lue) LED.
Light up just the red segment when you want a red light and all three segments when you want a white light.

This is one example of such a part: https://industrial.panasonic.com/content/data/SC/ds/ds4/LNJ757W86RA_E.pdf
It's available from Mouser: https://www.mouser.com/search/ProductDetail.aspx?R=0virtualkey0virtualkeyLNJ757W86RA
and DigiKey: https://www.digikey.com/product-detail/en/panasonic-electronic-components/LNJ757W86RA/P16825CT-ND

u18b

  • Crew
  • *
  • Posts: 4052
  • Respect: +2374
    • My website
0
Do really need to go down to 0402 LEDs?  It looks like there is a lot more room than that near the split light pipes.
Is height the problem?  The difference in "Trickyness" between soldering an 0603 and an 0402 is huge.  That's why I ask.
Maybe solder magnet wires to the LED and then drill a hole into the plastic below the light pipe, and sink it in there?

You are probably right.

At present, I went ahead and bought a variety of 0402.  I'm leaning toward installing the LEDs right in the tail.   Mounting them one on top of the other.

Bought from this guy.  In USA, so dodn't have to wait a month from Asia:
https://www.ebay.com/itm/332425926401?var=541457368097&ssPageName=STRK:MEWNX:IT

I've already experimented with soldering these to a board.   The price is cheap enough to lose a couple to the carpet gods.  I hate paying a couple of bucks PER LED.

Anyway, I'll have pics later, but I mounted two on a board such that the positive sides are soldered together (blue wire) and the negative sides are separate.  Each negative wire gets its own resistor.

Before I do ANYTHING for the car, I'm experimenting to make sure it all works.  Especially true since the LED guy gives no current draw info.
So I made a board and then wired it to an old FM chassis that has a TCS M4 decoder in it.  I leave the loco (non-running) and let the light work.

Here's what I have found so far.

The red LED appears to be working perfectly.  I have a 1k resistor on it.  No heat, looks good.  Programmed it for Mars light.  Looks great.

The white LEDs from this guy are a problem.  I keep burning them out.
1k
then 2k
then 3.3k

When 4.7k burned it out (though it lasted longer- about 5 or 10 minutes), I figure these are just too cheap quality to work.  I can't imagine an LED that takes a 6k resistor.  It was already a little dim at 4.7k.  I may still try a higher value just to see.  Gives me good practice soldering 0402s.

By the way, I bought some really expensive 0402s from N-gineering.  They are better quality and only take a resistor of something like 563 ohm (I can't remember- but close enough for comparison).

I have other white LEDs that I'm using on the EP-2 project.  Bought from Asia.  I've never burned one out- so quality appears to be better.  I think they are 0603.  I use 4.7k resistor on the EP-2 -- which is overkill.  I want them to last forever.  And at 4.7k, they are bright enough.



I'm thinking of everybody else and trying to do something that everybody else can replicate just like I did.   That's why I bought the mixed LED set.  It was easy.
But that may not work out.


« Last Edit: November 16, 2017, 11:58:59 AM by u18b »
Ron Bearden
CSX N scale Archivist
http://u18b.com

"All get what they want-- not all like what they get."  Aslan the Lion in the Chronicles of Narnia by C.S.Lewis.

nickelplate759

  • Crew
  • *
  • Posts: 3814
  • Respect: +1553
0
Digitrax also has function decoders, very close to the DZ126T
TL1 and TF4 are both 11.7mm x 7.8mm x 4.1mm
George
NKPH&TS #3628

I'm sorry Dave, I'm afraid I can't do that.

u18b

  • Crew
  • *
  • Posts: 4052
  • Respect: +2374
    • My website
0
An alternate solution for the rear mars light would be to use a single R(ed)-G(reen)-B(lue) LED.
Light up just the red segment when you want a red light and all three segments when you want a white light.

This is one example of such a part: https://industrial.panasonic.com/content/data/SC/ds/ds4/LNJ757W86RA_E.pdf
It's available from Mouser: https://www.mouser.com/search/ProductDetail.aspx?R=0virtualkey0virtualkeyLNJ757W86RA
and DigiKey: https://www.digikey.com/product-detail/en/panasonic-electronic-components/LNJ757W86RA/P16825CT-ND

That's a good suggestion.  Might work.  Would require some creative wiring.

White wire lights only the red.

Yellow wire lights all three. 

I assume this is an LED package and that you would need increased resistance for lighting all three versus lighting only one.

Also, I wonder how white the white is when all three are lit.


Further, says common is Anode.  It that pos or neg?  I never remember.  That's why I just stick with pos and negative.  Easier for my simple brain.

We need a common POSITIVE for this to work.  Since the decoder function wires are the negative.  The blue wire is positive.


« Last Edit: November 16, 2017, 11:54:24 AM by u18b »
Ron Bearden
CSX N scale Archivist
http://u18b.com

"All get what they want-- not all like what they get."  Aslan the Lion in the Chronicles of Narnia by C.S.Lewis.