Bachmann Spectrum USRA 4-8-2 light Mountain

[mmagliaro]

I get 10 mm on my light mountain (measuring the driver face and excluding the flanges).

Cheers,
-Mark

That's .393",  which is very close to what I got with a caliper. 

[nickelplate759]

I know this is an old thread, but I'm just getting around to adding DCC to a 1st-run Light 4-8-2.   I can't for the life of me get the boiler off.  I can get the cab-end free, and I understand that I need to somehow slide the boiler shell forward, but I there seems to be some kind of interlock up front that I can't see and is keeping the boiler from sliding.  Any tips would be appreciated.

[Mike C]

 Have you taken the steps off the front of the running boards ?  And maybe the pilot deck braces , Its been many years since I did DCC on one of these .  Yes it's still a PITA to get off , and even harder to get back on .....Mike

[nickelplate759]

Have you taken the steps off the front of the running boards ?  And maybe the pilot deck braces , Its been many years since I did DCC on one of these .  Yes it's still a PITA to get off , and even harder to get back on .....Mike

I tried to get the pilot & steps (they seem to be one piece, but maybe not?), but that's tricky too - I don't want to break them, and I can't seem to free the little bits of running board that interlock with the cylinder exhaust pipes.  I did loosen the pilot deck braces.

[Mike C]

 IIRC the steps/short run boards are seperate castings . I think they have tabs that fit in the pilot deck . The short running boards are connected to the steps , and there is a slot in the exhaust pipe that it sits in . 

  OK hold on I just looked at Spookshows site and remember a bit better now .  Try lifting up a bit as you push the boiler forward . I think the problem is that the exhaust pipes do not connect to the boiler , the boiler slides past them , and its a tight fit .  The pics here should help http://spookshow.net/loco/bachmann482.html ....Mike

[nickelplate759]

looking at Spookshow's pictures, I wonder if my primary problem is the removable zinc(?) weight.  Either it's stuck in the boiler, or I haven't lifted the boiler high enough to clear the part of the weight that sticks up into the sand dome.

The front running board steps are indeed separate pieces but appear to be glued into the pilot deck.


<edit> Success!  I was correct - the removable weight was causing the interlock (it was up in the shell).  I did need to loosen (but not remove) the pilot deck in order to get the deck braces out.
</edit>

[nickelplate759]

I've got the decoder in, but decided to test it before putting the shell back (hey, I'm slowly learning from experience).
It runs (Huzzah!) but the headlight doesn't light.  I can test the LED itself (in place) and it's fine (lights up). 

I note that the headlight board has the LED, what appears to be  capacitor, and 3 "222" SMD resistors (2.2kOhm).

I poked around with a multimeter -  I think the white wire goes to all three 2.2kOhm resistors in parallel, then to the + of the LED.  The blue wire goes to both the - of the LED and to the (apparent) capacitor, the other end of which is connected to the + of the LED.

From the white wire to the + of the LED I measure 728 Ohms - which suggests that all three 2.2kOhm resistors really are in parallel.

Oh - in typical Bachmann fashion, the light board can short out against the frame.  I hope that didn't fry the decoder's light outputs.

So - is the fact that the LED won't light on DCC control due to a fried controller (Digitrax DZ126), or should I try and remove what I think is a capacitor and see if that helps?

[nickelplate759]

Here's my crude attempt at a diagram.  The ? indicates the "apparent capacitor".
[ Guests cannot view attachments ]

[peteski]

George, that capacitor in parallel with the LED should not have detrimental effect while powering the LED from a function output of a decoder. It is installed to prevent the LED from flickering when the loco runs on DC and the motor generates BEMF. Having three 2.2k in parallel is unusual, but as you noted not wrong.

But the colors are reversed.  The positive side of the LED PC board should have the blue (not white) wire attached. Blue is common positive on DCC decoders. And by the same token, decoder's white wire should go to the negative (cathode) side of the LED circuit.

My other recommendation is that if you have a spare LED and some resistors (like a 1K) hanging around, test the decoder's "white" output. If it doesnt' work (and if yu then test the "yellow" output and that works, you blew up the F0F output transistor on the decoder.

Some manufacturers (to reduce the number of wires) hook up the positive side of the headlight to one of the rail pickups, while the negative is connected to the headlight output of the decoder.  Since the DCC track voltage flips polarity many times a second, the headlight illuminates roughly 50% of the time (it actually flashes very quickly).

[nickelplate759]

Well, after googling around, I don't think it's a capacitor at all.  I think it's a Zener diode. It's a cylinder with a clear case and the leads go straight through. There's a dark band near the end connected to the + end of the LED (in my diagram).

Bachmann put the Blue wire to the - lead of the LED.  Silly me I just assumed that was done correctly.
Even so, I believe the LED should light up when I reverse direction - it doesn't.

<edit>  I took out the light board and wired in a spare LED and resistor to the DCC decoder - and it works.  So Huzzah! I didn't fry the lighting output.     I'm going to try taking out the (I think) diode. </edit>

<edit2>  Took out the (I think) diode, reversed the blue and white wires so they are in the right place, and everything works!</edit2>

[peteski]

Well, after googling around, I don't think it's a capacitor at all.  I think it's a Zener diode. It's a cylinder with a clear case and the leads go straight through. There's a dark band near the end connected to the + end of the LED (in my diagram).

Bachmann put the Blue wire to the - lead of the LED.  Silly me I just assumed that was done correctly.
Even so, I believe the LED should light up when I reverse direction - it doesn't.

<edit>  I took out the light board and wired in a spare LED and resistor to the DCC decoder - and it works.  So Huzzah! I didn't fry the lighting output.     I'm going to try taking out the (I think) Zener diode.
</edit>

It is likely just a standard small-signal silicon diode (like 1N914).  It is wired in reverse parallel with the white LED (to protect the LED from excessive reverse voltage (when running in DC).  White LEDs have low tolerance for reverse voltage (actually reverse current is the LED-killer), so that diode basically clamps the reverse voltage to its forward voltage (0.7V).  This type of circuit is utilized in some other DC models which use white LEDs for headlights.

You can take that diode out or leave it - should work either way, and in DCC the LED will never see any reverse voltage on the LED either way.

[nickelplate759]

It is likely just a standard small-signal silicon diode (like 1N914).  It is wired in reverse parallel with the white LED (to protect the LED from excessive reverse voltage (when running in DC).  White LEDs have low tolerance for reverse voltage (actually reverse current is the LED-killer), so that diode basically clamps the reverse voltage to its forward voltage (0.7V).  This type of circuit is utilized in some other DC models which use white LEDs for headlights.

You can take that diode out or leave it - should work either way, and in DCC the LED will never see any reverse voltage on the LED either way.

Well, that makes sense.  Thanks for the explanation.  In any event, I took it out, fixed the blue/white problem, and everything works fine now.  Now I just need to get this beast back together again.

[nickelplate759]

Before I put the boiler shell back (I'm still not sure how to do that - some of the piping detail keeps getting in the way), I need to correct the gauge on some of the wheels.  Pilot wheels are easy.  The trailing truck is already correct, and the drivers are close enough (just a bit tight) that I don't want to mess with them. The tender wheels are WAY too tight, but if I pull them out then they don't roll, because the wheel and half-axle are one piece and widening the gauge also lengthens the effective axle length.   An obvious solution would be to replace the wheelsets - suggestions?

I'm starting to come around to Spookshow's assessment of these as "total Hobby Shop Babies"  http://www.spookshow.net/loco/bachmann482.html

[nickelplate759]

In my parts box I found a package of Kato #932080  "33in Wheelset for models with Pickup Ability".  They are a perfect fit and perfectly in gauge.   

[Mike C]

 Good to see you got it figured out !  IIRC putting the boiler back on was a bit harder than getting it off . And while you have it apart you may want to shim the traction tire drivers with a sliver of Scotch tape between the frame and bearing block . Made quite a difference in pulling power for me .....Mike