Hi all,
I have been actively rewiring my Locos with LED’s for headlights and the only aspect of them I find difficult is that when the loco is going backwards, the LED’s flicker on. This is whether I have Diode protected the LED’s in my locos or used one LED as reverse protection for the other. It is a bit annoying!!!
Can anyone offer a suggestion as to how this might be overcome?
Thanks in Anticipation from Down Under
Trevor www.xdford.digitalzones.com FYI
not
I don’t know the cause either but my engines, especially Atlas engines, will have a flicker in the rear headlight when running forward. Some times this will happen when running over a turnout but other times it seems to happen for no reason at all. I don’t have an answer either but I’d sure be interested if someone has one.
Is this DCC? Some decoders, especially sound decoders, have a problem with small pulses that occur on the rear light function output when the front headlight is on. The pulses are small so lamps don’t respond to them, but LEDs require very little voltage to turn on and will show these pulses as flickers. There are two options to fix the problem: 1) replace the rear LED with a lamp, or; 2) put a 0.1uf capacitor and a 1K resistor in parallel with the LED. The capacitor will absorb the pulses and the resistor will discharge the capacitor - and the LED will stay off. This will not affect the normal operation of the rear light.
Chris
Good theory, Chris, but I’m running straight DC at the moment and still see the flicker. Understand, this is a flicker in the rear headlight when the train running forward - the headlight should be dark, so this may be different than the OP’s problem. I first noticed it when I was running the unit without the shell trying to diagnose a derailing problem anf saw the rear LED flicker. It’s bright enough that, in a dark room, I can see the flicker in the rear headlight on the boxcar coupled to the engine. This is a P2K SD-24.
I know exactly what causes this - I just hope I can explain it properly.
The problem occurs during intermitent power loss at the pick-up point, either from dirty track, dead frogs or what-ever. When electricity is applied to the motor it acts as a load creating rotational force. When the electrical input to the motor is interrupted - even for a split second - the motor is still turning. This rotating motor now becomes a generator producing a reverse polarity to what is being supplied. This momentary loss of pick-up will cause the generator which is now producing the opposite polarity to light the opposing LED.
A good cleaning of the rails and wheels will drastically reduce this effect. Leds are super sensitive to even the smallest burst of current. The burst is so small that standard bulbs don’t react to it when wired through reversing diodes. It can be minimized to some extent by increasing the value of the resistor you are using with the LED.If it does not, then a small capacitor across the motor in the locomotive will provide a permanent fix. The capacitor does not have to be big, say a 0.047uF to 0.1 uF polyester capacitor (often called a “Greencap”). These are small and should be able to be tucked against the motor somewhere.
To demonstrate the effects of a motor / generator set, the late John Allen had a unique throttle on his layout - two motors connected to each other by a common shaft with a large six inch flywheel between the two. Variable voltage was applied to the first motor and the second motor was attached to the layout. When the power to the first motor was shut off, the large flywheel slowed very gradually which in turn reduced the voltage on the second motor (generator) proportionately … kind of a poor man’s momentum throttle !!!
Mark.
Makes sense to me. Almost all of my Katos do this same thing. I never thought about that.
Jeremy
Well said, Mark.
Very good explanation, Mark, and one I suspected to be true. I gave the locomotive wheels and the track a really good cleaning and, although it reduced the flickering, it did not eliminate it. I will still get a flicker at some joints and turnouts and I suspect some momentary loss of power is almost inevitable. Like I wrote, I don’t think I ever would have noticed it if I wasn’t running the SD-24 without a shell and can tell that the dirtier the track, the more flickering. I think the only way to stop it entirely is a greencap type capacitor as you wrote. I’m a little suprised that the manufactures don’t do this since it must occur in almost all LED-equipped engines. Maybe we are the only ones to ever notice it. [:)]
Hi all,
Thankyou for your inputs although I have followed the suggestions, I still have not eliminated the flashing. I know my track situation is not the best (read not perfectly clean… but is any track perfectly clean) but it is very serviceable and I have very little trouble with pickup etc. I tried the 1K resistor and the Capacitor but is there a threshold where the capacitor and resistor take over from the diode? That is, is there an optimum combination which would stop the flicker?
Most of my locos are “ini-directional” in as much as they face one way for the most part A-B units, Low nose GPs etc but I have switchers that run both ways on transfer runs and I would like to run my A-A units with LED’s similarly instead of turning them to take advantage of lights I would put into the dummy etc. but the flicker is annoying given the inputs of you guys.
So please, further suggestions if possible would be gratefully accepted and tested!
Regards from down under
Trevor
Have you considered converting to DCC ? [;)]
This is an inherent problem caused by marrying old world control (DC) and new world lighting (LEDs). As you’ve already discovered, there’s no simple cure. LEDs do not function well with a DC system regardless of what type of constant / directional circuitry you use. LEDs are extremely sensitive to current change, and with changes in voltage (how DC control works) there is going to be changes in current - there’s no way around it.
I’m not saying it can’t be done, but the line needs to be drawn somewhere. You could design a circuit using a voltage regulator that would control a small 3-5 volt mini relay that would physically dis-connect the opposing LED from receiving the BEMF voltage. You could even step that concept up and design a small TTL circuit to do the same as well. Both of these concepts would require 3-4 volts before they were to function however, so anything below 3 volts on the track would do one of two things - the flickering would return or the front light won’t light. Taking THAT issue a step further, you could in turn add an additional series of diode bridges to the motor that would prevent the motor from turning until the track voltage reached the required voltage of 3-4 volts to make the physical dis-connect functionable.
Thing is now - you have all those additional components you have to somehow cram in the shell. The price of the parts and the time involved to design and install these components is probably going to be more than the price of a low end decoder. I know that’s not what you want to hear. Unfortunately, there’s no easy way to change the rules of electron flow.
Even when the engine / motor is running forward on perfectly clean track, there is still an extremely small amount of reverse current existing. This occurs as the motor windings pass between the magnetic field of the internal magnets. This reversal is called back-electromotive-fiel
Hi there Mark, I have considered converting and in fact went to a talk about it put on by my club last Saturday. While I know it is an illusion, I like my throttles (www.xdford.digitalzones.com and follow the link) and how my engines perform with the throttle especially the braking aspect and I have yet to see the DCC equivalent of it in terms of the realism and control I want to achieve. However I may yet be dragged kicking and screaming to DCC as I see it.
I am an electrician by trade so yes I do know about back EMF although I am a little surprised that permag or can motors can generate enough to do this, flywheels not withstanding. You are right about the space with caps and resistor and protecting resistor and LED and connecting wire in place as well as the biggest Mashima I could fit to the switcher, it is getting crowded in there!
Thanks for your help with it anyway,
Regards from Down Under,
Trevor
I have never seen this
Now that may just be luck - or it could be the system that I am using for my lights.
I run DC with a pulse inserted to keep cars lights etc lit…
Constant lighting (for car lights and headlights) uses a bridge rectifier - with the AC connections to the pickups. A 100ufd 16v capacitor is connected across the DC leads of the bridge and then I have a current regulator to provide constant lighting level. The -lead of the led goes to the negative terminal on the bridge.
The backup light currently uses a single diode + capacitor + current regulator. The negative lead of the led goes to the negative lead on the above bridge… (although it could be wired to the appropiate pickup.
I haven’t seen pulses on the b/u light - maybe the capacitor filters them out…
Villy
There you go - Villy just proved my point.
It CAN be done, with the number and size (capacity) of the caps Villy is using, I’m certain it is more than adequate to filter out the minute BEMF that is created. Again, to my point, it’s not a plug-n-play solution. If you have cheap access to these components and the knowledge to piece it together, this is probably one of many solutions that will work. You will still have to find a place to install these components within your engine. There are a lot of folks who love doing this sort of thing (bonus for them), but on the flip-side, there are some folks who have trouble soldering two wires together.
All I’m saying is that there are always solutions - you have to decide for yourself what the options are, your capabilities and how much time you’re willing to spend on it. I used to do this sort of thing all the time, but having since converted to DCC years ago, I have drawers full of electronic components I don’t have to fidgit with any more. [;)]
Mark.
SIMPLEST SOLUTION:
Substitute 1.4 v 15ma. grain of rice bulbs (approx $1 ea.)
When run across a pair of cheap diodes one has a constant 1.4v drop when in series with motor (no resistors), regardless of voltage. A single diode provides directionality.
ADVANTAGES:
CONSTANT VOLTAGE lighting comes on after 1.5v is applid to track (and before the engine starts to move), and doesn’t dim or brighten with speed.
G.O.R. bulbs give prototypical headlight ‘white’.
Some white LEDs, and LED circuits, require 5 v. before activation.
I gotta agree with Don - for DC lighting, the simplest approach is to use bulbs. Here’s a couple ways to build exactly what Don was describing …
Mark.