[goedel]

Hello,

I currently have a DC N scale layout but will be upgrading to DCC in the near future.

The issue is that I'm about to install a pair of red SMD LEDs (or a single red LED and two fibre-optic cables) in the end coach of a rake to provide a pair of tail-lights. They will be connected to a resistor and a pair of diodes so that they will work on DC in both directions. I realise that their brightness will depend on the DC voltage but as I'm moving to DCC with a constant track voltage it doesn't matter.

More importantly I'm worried about possible flickering because of the DCC square wave signal or is the freqency too high for the human eye to notice? If the flicker is an issue then I shall add a capacitor to allow power to be provided continuously, but then I wonder if the capacitor will interfere with DCC?

Here is a proposed circuit diagram, the red and green lines represent opposite rail polarities:

The reason I ask is that because space is tight and I will have to modify the interior of the coach at one end then I want my design to be DCC compatible so that I don't have to make further modifications in future. The other option is to fit interior lighting to the coach (not a priority), and use this plastic light transmitter from Minitrix (66657), anyone have thoughts on this as an alternative?

(Obviously this only fits certain Minitrix coaches and I have several rakes of Roco ones so the SMD LED solution is still needed...)

Thanks,

Goedel

 

[Brian Considine]

Hi Goedel,

This is exactly the circuit we use for DCC tailights. You will not notice any flickering at all due to the DCC signal. The capacitor will eleiminate flicker due to losing contact over points etc. The diodes, in effect stop the DCC signal "seeing" the capacitor.
Hope this helps.

QUOTE (goedel @ 8 May 2007, 16:52) <{POST_SNAPBACK}>Hello,

I currently have a DC N scale layout but will be upgrading to DCC in the near future.

The issue is that I'm about to install a pair of red SMD LEDs (or a single red LED and two fibre-optic cables) in the end coach of a rake to provide a pair of tail-lights. They will be connected to a resistor and a pair of diodes so that they will work on DC in both directions. I realise that their brightness will depend on the DC voltage but as I'm moving to DCC with a constant track voltage it doesn't matter.

More importantly I'm worried about possible flickering because of the DCC square wave signal or is the freqency too high for the human eye to notice? If the flicker is an issue then I shall add a capacitor to allow power to be provided continuously, but then I wonder if the capacitor will interfere with DCC?

Thanks,

Goedel

 

[goedel]

QUOTE (dbclass50 @ 8 May 2007, 21:02) <{POST_SNAPBACK}>This is exactly the circuit we use for DCC tailights.
That's reassuring to know! I was thinking about how to do it and it just seemed the logical solution. GCSE Electronics turns out to have had a use after all these years!

Thanks for the capacitor 'DCC visibility' info.

Goedel

 

[SPROGman]

QUOTE (goedel @ 8 May 2007, 21:57) <{POST_SNAPBACK}>That's reassuring to know! I was thinking about how to do it and it just seemed the logical solution. GCSE Electronics turns out to have had a use after all these years!

Thanks for the capacitor 'DCC visibility' info.

Goedel
If you propose to add this circuit to many coaches, you should use ultrafast diodes such as UF4001 rather than bog standard 1N4001.

Any diode used in a rectifier circuit has a "reverse recovery time" during which it will conduct in the opposite direction when the input polarity reverses. In your all 4 diodes will conduct for a short period, each time the DCC polarity reverses, putting a short across the track. Using ultrafast diodes minimises the problem. In a lot of cases you'll get away with it, but...

Andrew

 

[Gordon H]

QUOTE (SPROGman @ 9 May 2007, 12:48) <{POST_SNAPBACK}>If you propose to add this circuit to many coaches, you should use ultrafast diodes such as UF4001 rather than bog standard 1N4001.

Any diode used in a rectifier circuit has a "reverse recovery time" during which it will conduct in the opposite direction when the input polarity reverses. In your all 4 diodes will conduct for a short period, each time the DCC polarity reverses, putting a short across the track. Using ultrafast diodes minimises the problem. In a lot of cases you'll get away with it, but...
I would also add that you should choose the size of your capacitor carefully, i.e. keep the value down to a minimum, otherwise you might find your command station objecting to the sudden load of a whole lot of discharged capacitors appearing across its output at switch on. I believe this situation cropped up with the new Hornby Pullmans a while ago.

 

[Gofer]

QUOTE (Gordon H @ 9 May 2007, 12:58) <{POST_SNAPBACK}>I would also add that you should choose the size of your capacitor carefully, i.e. keep the value down to a minimum, otherwise you might find your command station objecting to the sudden load of a whole lot of discharged capacitors appearing across its output at switch on. I believe this situation cropped up with the new Hornby Pullmans a while ago.

So is it a catch 22 situation inasmuch that the capacitor has to be small to avoid overloading the command station but large to be effective in reducing flicker on the LEDs? Do you (or anybody) have any idea what value the relative components should be. As I understand it the value of the resistor is fairly well fixed by the current required by the LED so presumably the only variable is the value of the capacitor.

Steve

 

[Brian Considine]

QUOTE (Gofer @ 13 May 2007, 16:28) <{POST_SNAPBACK}>So is it a catch 22 situation inasmuch that the capacitor has to be small to avoid overloading the command station but large to be effective in reducing flicker on the LEDs? Do you (or anybody) have any idea what value the relative components should be. As I understand it the value of the resistor is fairly well fixed by the current required by the LED so presumably the only variable is the value of the capacitor.

Steve

To a certain extent the answer is almost "how long is a piece of string" - the ones in our units (which power 2 x LED's) are 100 mF & they are fine. I would not go much higher. We have some factory fitted lighting in our Trix Rhinegold set which I'm certain have higher rating caps - I it's any help let me know & I'll dismantle one & check.

 

[Edwin]

QUOTE (dbclass50 @ 14 May 2007, 09:03) <{POST_SNAPBACK}>the ones in our units (which power 2 x LED's) are 100 mF & they are fine.

For the avoidance of doubt, is 'mF' intended to refer to micro-farads? If so I would suggest using 'uF' to avoid possible confusion with milli-farads.

 

[goedel]

QUOTE (Gofer @ 13 May 2007, 16:28) <{POST_SNAPBACK}>As I understand it the value of the resistor is fairly well fixed by the current required by the LED so presumably the only variable is the value of the capacitor.
I believe that you can play around with Q = Q'exp(-t/RC) and as Q = CV then V=V'exp(-t/RC) (where Q' and V' are the initial charge and voltage respectively). This assumes exponential decay and so an infinite discharge time which is only an approximation to the real world.

The half-life t' = RC/ln2 is the time for the voltage to drop by a factor of 1/2. As you say R is fixed by the LED requirements for normal operation and so one can pick a suitable half-life and then get C.

I would just use 100mF (or 100muF?) as dbclass50 suggests since the wheels are only likely to lose contact instantaneously, (and anything larger is a tight fit in N scale (unless it's a tank wagen!!)). You could try both and if either makes your DCC system explode then use the other!

Goedel

 

[goedel]

I've just remembered an excellent website featuring lots of useful info on how to fit LED tail-lighting to all manner of wagons and coaches in N scale, and because it is N scale suitable it is automatically perfect for H0 or OO etc:

Looks good for N!

Modulbau Team Köln Bohn
http://www.mtkb.de/
(it is in German but the pictures transcend language - click on the bottom right hand yellow link called 'Schlusslicht & Verbesserungen')

Goedel

 

[SPROGman]

QUOTE (Gofer @ 13 May 2007, 16:28) <{POST_SNAPBACK}>So is it a catch 22 situation inasmuch that the capacitor has to be small to avoid overloading the command station but large to be effective in reducing flicker on the LEDs? Do you (or anybody) have any idea what value the relative components should be. As I understand it the value of the resistor is fairly well fixed by the current required by the LED so presumably the only variable is the value of the capacitor.

Steve
You can use another small R in series with the C to limit the charging current.

Andrew

 

[Brian Considine]

QUOTE (Edwin @ 14 May 2007, 09:18) <{POST_SNAPBACK}>For the avoidance of doubt, is 'mF' intended to refer to micro-farads? If so I would suggest using 'uF' to avoid possible confusion with milli-farads.

Well spotted (just checking to see if anyone's awake).

Yes, it should be uF !

 

[Gordon H]

QUOTE (SPROGman @ 14 May 2007, 12:02) <{POST_SNAPBACK}>You can use another small R in series with the C to limit the charging current.
I was just about to say the same thing!
I would suggest something between about 10 and 100 ohms.
47 ohms might be a good starting point.

 

[Gofer]

Hmmm...... I feel some experimentation coming on

. Thanks all and thanks for that link Goedel. I must admit I didn't understand a word of it but as you say, I did get the gist from the pictures. I did also discover that if you Google the rectifier (B40C400) it actually comes up with some of the articles from that site and allows the choice of translating the text to English (albeit literally). Interesting, some of the capacitors being used are as high a value as 220µf.

Steve

 

[zmil]

Hi All

Thanks Nick I had not seen this thread before an interesting read
I had a long while ago bought some wagons (from feeBay) that had been modified with Taillights etc
I have pulled them out recently and tried them with DCC and they are working fine
So it is an excuse to modify some more

Regards Zmil

 

[upnick]

QUOTE (zmil @ 14 Nov 2008, 00:45) <{POST_SNAPBACK}>Hi All

Thanks Nick I had not seen this thread before an interesting read
I had a long while ago bought some wagons (from feeBay) that had been modified with Taillights etc
I have pulled them out recently and tried them with DCC and they are working fine
So it is an excuse to modify some more

Regards Zmil

Glad you find the thread useful Zmil

i have the excuse to convert several wagons including some intermodal as in the link here and with the size so small with the SMD'S vechiles in N gauge are a possiblity though some on feebay are awful the light escaping badly needs a bit of blocking in with plasticard and black paint

 

[zmil]

QUOTE (upnick @ 14 Nov 2008, 19:24) <{POST_SNAPBACK}>Glad you find the thread useful Zmil

i have the excuse to convert several wagons including some intermodal as in the link here and with the size so small with the SMD'S vechiles in N gauge are a possiblity though some on feebay are awful the light escaping badly needs a bit of blocking in with plasticard and black paint

I have been using aluminum foil (from the top seal of Nescafe coffee jars) to block the light. Some HO loco's let a lot of light out at the coupler pocket as the light source is directly over the top of it . The foil is thin enough not to interfere with the body to chassis fit . I even glue it to the insides of the front and rear of the body if the LED's are aimed at it

Regards Zmil

 

[upnick]

Hi Zmil,
Great tip

in confined spaces a couple of coats of matt black paint can cover the inside panel lines where the plastic is thinner letting light show through.