Railpower 1300 modified for Z scale by Micro Trains

[mmagliaro]

Amazing!  I thought those circuits that put high frequency AC on the track for constant lighting had long since gone the way of the Dodo bird!

[peteski]

Amazing!  I thought those circuits that put high frequency AC on the track for constant lighting had long since gone the way of the Dodo bird!

I was just speculating, but what else could keep the lights lit while the DC engine is not running?

[brill27mcb]

Actually, constant lighting ("CL") power packs like the Tomix and newer Kato ones simply let one adjust the minimum width of the PWM pulses. You set it to a point where the LEDs come on, but the motor does not yet rotate. (Better yet, by turning the dial down to where the the motor stops turning but the LEDs stay on.) It's that simple, and not an injected or overlaid pulse like you may be used to with Model Railroader's old TAT or the Model Rectifier designs with "spike" pulses. Your train power is PWM pulses that are basically 0% wide (of the frequency cycle) when the throttle dial is "off", and pulses of 100% of the frequency cycle when the throttle dial is at maximum. The Tomix and Kato power pack PWM pulse amplitude is a safe 12 to 15 volts.

Rich K.

[mmagliaro]

I was just speculating, but what else could keep the lights lit while the DC engine is not running?

I found a post and discussion on the azl forum.  I think all that little tab and stop does is keep a low DC voltage on the rails so the
DC LED headlights stay on.  The post says they had some engines that unfortunately creep on that low "constant lighting" setting,
so it's not high frequency AC.  It's just "really low speed to light up the lights and hope the motor doesn't move".

[peteski]

Actually, constant lighting ("CL") power packs like the Tomix and newer Kato ones simply let one adjust the minimum width of the PWM pulses. You set it to a point where the LEDs come on, but the motor does not yet rotate. (Better yet, by turning the dial down to where the the motor stops turning but the LEDs stay on.) It's that simple, and not an injected or overlaid pulse like you may be used to with Model Railroader's old TAT or the Model Rectifier designs with "spike" pulses. Your train power is PWM pulses that are basically 0% wide (of the frequency cycle) when the throttle dial is "off", and pulses of 100% of the frequency cycle when the throttle dial is at maximum. The Tomix and Kato power pack PWM pulse amplitude is a safe 12 to 15 volts.

Rich K.

Similar principle but different method.  Most LED dimmer circuits use PWM to control their brightens and I'm surprised that even when the pulses are narrow enough not to move the motor armature, they still allow the LEDs to glow brightly. I would think powered with such short pulses the LEDs would not be glowing anywhere near their full brightness.  This would not work with incandescent bulbs since the short pulses would not heat up the filament enough. That however is a moot point, since most model lighting is now done with LEDs.

[peteski]

I found a post and discussion on the azl forum.  I think all that little tab and stop does is keep a low DC voltage on the rails so the
DC LED headlights stay on.  The post says they had some engines that unfortunately creep on that low "constant lighting" setting,
so it's not high frequency AC.  It's just "really low speed to light up the lights and hope the motor doesn't move".

Not sure if that's it. White LEDs need around 3V to glow.  I'm fairly certain that any Z scale model's motor would run steadily at 3VDC.

[mmagliaro]

Not sure if that's it. White LEDs need around 3V to glow.  I'm fairly certain that any Z scale model's motor would run steadily at 3VDC.

Good point.  Whatever it is, there are reports on Z forums (like AZL's and Marklin) about engines that either won't light from the constant lighting feature, or whose motors move from the lighting feature.    I let my curiosity get the better of me on this, as I don't even have Z equipment, nor do I plan to, so I'll just shut up now.  (I just want to put one of those Rokuhan controllers on a scope to see what the heck that lighting feature does!)

[peteski]

Good point.  Whatever it is, there are reports on Z forums (like AZL's and Marklin) about engines that either won't light from the constant lighting feature, or whose motors move from the lighting feature.    I let my curiosity get the better of me on this, as I don't even have Z equipment, nor do I plan to, so I'll just shut up now.  (I just want to put one of those Rokuhan controllers on a scope to see what the heck that lighting feature does!)

The li'l "peteski voice" in your head says "Do it Max!" 

Or maybe one of the Rokuhan throttle owners on TRW could let you borrow theirs to analyze the voltage.

But the more I think about this, going by what you have read about the behavior, it might be just a simple as you mentioned: basically low voltage is sent to the rails, sometimes enough to light the lights, or in other examples higher than the starting voltage of some specific motors.

[nscaler711]

Ok, Like I said in my last post I have no real way of testing the true output of my RC-02, But what I could initially read on a basic voltmeter on DC the throttle and secondary light dial, it reads 0.02, on AC it very quickly read 2.01 and then dropped to 0.51 but I would assume if it is pulsing the voltmeter is not fast enough to read it. If someone wants to send me a Keysight Oscilloscope (since I didn't win any from ElectroBOOM) that'd be great. 
I think the PWM is just fast enough to light the LEDs but not to have visible Flickering so possibly in the 60hz+ range.

In the last few days I have left a AZL GP38 on the track with the dial off, it has not moved since being placed on the tracks. nor has a MTL SD40-2.

I Chose the AZL for the basic lighting test as its LED has been modified/replaced, and is way brighter than the GP38s.
I also had an ancient Pro Z E231-500 Yamanote set on hand as well. The interior lighting is so perfect on these.   
This is throttle at "Off" 0.02VDC and briefly 2.01VAC





This is the minimum setting that my AZL F7A would not start, which according to the very brief reading was 2.49VAC or 0.50VDC
The Pro Z E231-500 was even higher before the motor started buzzing.... its old... like a decade prior to Rokuhan and yet some parts like the trucks are interchangeable...




Also, It's not that I wouldn't trust Max @mmagliaro to do a teardown on mine, I just have only one and running Z scale has been nice lately. That and I put fancy stickers on it. You know for +5HP and to protect me from Angels.... though Gaghiel would be cool to see.

[mmagliaro]

No worries.  I don't expect anyone to send me one to check out.  But I will say, I would not be doing any "tear down".
I just want to put the output on an oscilloscope to see what sort of voltage it is putting out that the LEDs
see, but motors do not.  It could be something as simple as high frequency AC, like the old-school "lighting generator" circuits
used to use.  But I was curious. 

[nscaler711]

No worries.  I don't expect anyone to send me one to check out.  But I will say, I would not be doing any "tear down".
I just want to put the output on an oscilloscope to see what sort of voltage it is putting out that the LEDs
see, but motors do not.  It could be something as simple as high frequency AC, like the old-school "lighting generator" circuits
used to use.  But I was curious.

That's fair, if I were to take mine apart, would that help decode whats going on at all?
I only have a general knowledge of electronics so I can not say for certain whats what on most circuit boards.
then again who knows I may have a need for a second one, and can always pick it up and ship one to you if you are still curious.

[mmagliaro]

I wouldn't bother taking it apart.  You might be able to tell if it has a hf AC oscillator in there, but then again these days,
that entire function could be done on a chip and it might be impossible to tell anything by looking at the board.
It would be a lot easier to just put a scope on the output and see what's coming out of there.
A multimeter isn't going to be able to tell.

[Chris333]

It runs off of 8 AAA batteries, can you get AC out of that?

[mmagliaro]

It runs off of 8 AAA batteries, can you get AC out of that?

Sure.  You put a little oscillator circuit in there.  The old Peter Thorne circuit that did this uses a two transistor circuit
that oscillates at 25 kHz or 30 kHz (he has two versions in my old book).  Because it was designed to handle
power-hogging incandescent bulbs in all the cars and engines, it was designed with big metal-cased power
transistors.  But it could be done with much smaller parts to provide current for LED lighting.  If the "AC" is really
just a simple square wave, it could be created by a 555 timer and a few external parts to control the frequency
and boost the current.

[Chris333]

Does this help at all?

The pins sticking up are the power and direction switches.