If you get them in the right

If you get them in the right place resistors run about a penny each. In the wrong place (Radio Shack) they're a dime or two each.

Charlie

In the long run....

Resistors - and LED's.

Can you be more specific

Can you be more specific Jeff, the search engine here isn't exactly stellar. I know what LEDs are but don't have any experience with them. I'm curious as to why the manufacturer says replace the bulbs and the decoder manufacturer says to install resistors. ??

If you get the right TCS decoder

If you get the right TCS decoder, it comes with 1.5v bulb resistors right on the decoder (no extra resistors needed).

Check out the TCS A6X ...

I talked to TCS...great guys.

I talked to TCS...great guys. They recommended the T1-P2K for my Proto 2000 GP7. There are sooooo many decoders out there my head is swimming.

my head is swimming.....

Hi Allen,

well, just wait untill you get into sound decoders, then your head will be swimming in a perfect storm between the Tsunami's and the ESU's...  I am still scratching my head on days, tho I am not a newbie on sound decoders. Only thing that helps is read, read and then read some more before you decide. And when you get the chosen decoder....start reading again. Luckely there is JMRI with decoder pro which makes the programming of decoders ALOT easier. Thing is, once you are bitten by DCC, there is no going back. It's addictive, you'll see.

Wonders of lighting....

The manufacturer probably put in low voltage bulbs for a constant lighting circuit - so they believe that if you hook it up to your average DCC board, they're gonna get fried by 12 - 16v.

The DCC board manufacturer on the other hand doesn't think you should need to replace the bulbs, since resistors are probably cheaper and will cut the power off the board down to something the bulbs can take. Helps to know what that is...

OTOH, my suggestion is, ditch the bulbs entirely.  Go with LED's that are going to have a lifespan longer than you and me put together probably. If the board has resistors like Joe mentioned, you don't need to add them. Otherwise, you probably will, since I think most LED's run about 3v. If I read a box correctly today that means that about a 680 Ohm resistor is the proper size - I also ran the numbers through the LED in serial calculator at http://metku.net/index.html?sect=view&n=1&path=mods/ledcalc/index_eng

Which I found from this page: http://www.theledlight.com/ledcircuits.html

I'm not a genius, especially with electronics, so I like to reference my sources.

Sources

Good links, thanks!

Jeff I love to see people get

Jeff I love to see people get the FACTS about a subject and then post those facts with references. If more of us did that rather than give just our opinions we would get alot more work done and do alot less arguing. Great links and I've been looking up the same stuff for all the DCC sound cards I bought Friday that have the lighting functions as well.

Dan

Digitrax LO165

I have a lot of PK2 locos and I have tried installing various decoders, but always had to replace the lights.  The new Digitrax LO165 decoder is a plug and play without changing out the 1.5v LED or Lights.

A few comments about LEDs

An interesting thread - and interesting comments.

I'm an electrical engineer, so I look at this stuff from a slightly different angle to some people.

I can't remember the last time I used a grain-of-wheat bulb in a model. Given the choice, I'd always use LEDs (if properly connected, they don't break) - however, the position is a bit more complicated than most people think.

One of the links given above suggests parallel connection of LEDs - I know some torches and camping lights use it, but it's very bad practice - please don't try it.

If you check LED data - from manufacturers or suppliers' catalogues - there are 3 figures that really matter:

• Maximum safe forward current -  this is the current at which the LED can be safely operated for extended periods. For most LEDs, this is about 20 mA - but some low current types might only be able to take a few mA. Generally, you can go a bit above this figure for a couple of seconds - but not for much longer, as the LED might overheat and burn out quickly. If the LED is in an enclosed space, heat might not be able to escape - so the current should be dropped. A normal rule of thumb is about half the quoted current - most of the time, this means 10 mA.
• Forward voltage - this is the voltage at which the LED starts to conduct. Below this level, the LED effectively acts as an open circuit and no current flows. Above this level, it conducts - even slightly above and the current will skyrocket, the LED will glow like the sun and get very hot - it'll probably give off smoke - it might even catch fire. (When I was a student, I actually saw it happen. Not again - once was enough.) This is the real reason why resistors are needed in series with LEDs - not to limit the voltage, but to limit the current.
  This sudden change in conduction status has a useful by-product - (like a lot of semiconductor devices) LEDs can be used to limit the voltage across them to the forward voltage. The exact value of this forward voltage varies - less for red (even less for infrared) - more for orange or yellow - more still for green - a lot more for blue (ditto violet and ultraviolet).
  The forward voltage even varies between LEDs of the same colour. A few years back, I tried connecting a load of supposedly identical LEDs in parallel, through one resistor - some lit and some didn't. I then tested the ones that didn't light - nothing wrong with them. This is the real reason why parallel connection of LEDs is such a bad idea.
  Ever wondered why white LEDs have a bluish or violet light and a very high forward voltage? They're actually blue (maybe violet or UV) devices in drag, with a coating or phosphor somewhere between the component chip and the casing.
• Maximum reverse voltage (sometimes called peak inverse voltage) - when an LED is connected the wrong way round, it doesn't take a very high voltage to wreck it - best to check the data but, normally, a good rule of thumb is to stay below about 5V. If you're using direct current (and know that the polarity might get swapped), or AC, there's an even easier method - LEDs connected in inverse parallel (+ to -; - to +) - most LEDs have a forward voltage a lot smaller than the maximum reverse voltage.

From these comments, you might think that LEDs are difficult to use - please think again. They're like a lot of things in electronics - once you've used them a few times, it actually becomes second nature. If you don't believe me, why don't you give them a try?

Regards,

Huw.

LED's in parallel

You mention later that...

Is this the only reason it is a bad idea, because if the LED's are not truly identical, some may not illuminate?  If that's it it doesn't seem like a horrible idea, just something one needs to be aware of, or is there a more sinister danger to wiring LED's in parallel?

LEDs in parallel

The real issue is that all the current goes through less LEDs than you were expecting - so the LEDs that are conducting (or lighting) are taking more current than they should.

This isn't so much of a problem if, like me, you aim for half the rated current per LED.

However, if you aim for the full rated current through each one, you end up with some taking more than they're rated for - so there's a risk of some of them burning out.

I don't know if this makes things any clearer - I hope it does, but (for reasons unconnected with this site) I'm not thinking as clearly as I normally do. If I'm just confusing people, I can only apologise.

Regards,

Huw.

The real issue is that all

Crystal clear now, thanks!  So if one insists on wiring LED's in parallel it would be wise to aim for current levels of no more than half the maximum.  That's good to know.  Thanks for taking the time to clarify that for me.

LEDs in parallel

"So if one insists on wiring LED's in parallel it would be wise to aim for current levels of no more than half the maximum."

Yes and no - the real issue is that, in parallel, you could get more current in individual LEDs than you bargained for - maybe even all of it.

As I only had 4 LEDs in parallel when I tried this, the total current was only twice the official rated current - so, even if all of it had gone through 1 LED, it would only have been taking twice the rated current. Most LEDs can take this for a few seconds - which was more than long enough for me to see I'd made a mistake.

If I'd been a bit less cautious with my resistor value, the current would have been higher, so there would have been more chance of an LED "taking up smoking".

After seeing that parallel connection didn't work as well as I'd hoped, I learnt my lesson - I've never tried it again. Since then, every time I've connected up LEDs, I've either gone for series connection or given each LED its own resistor.

The reason I went for half the rated current was actually to reduce the chance of LEDs overheating (something I still do - and not just with LEDs). I aim for a reduced current however I connect them - I want them to have a long life. However, it also bought me some time to see I'd made a mistake.

I said I'm not thinking clearly at the moment (a cold - which will clear in time). It seems I wasn't joking.

Sorry about not expressing myself clearly.

Regards,

Huw.

Be very careful for those

Be very careful for those that like to experiment with LED's as if you apply the 12-16 volts all at once I've seen them go BANG and send micro fragments in all directions thats when they didnt flash one time as a bright flash and a hot little flame. if you aim for .7 Volts as a happy medium as a starting voltage you'll be safe.

I have some strange tri-colored LED's and as you click the voltage from .7v to 1.5v to 2.5 v the same LED changes colors from green - yellow - Red  now I don't remember off hand which voltage give's you which color but there are more than 12 different Tri-colored LED's that I've seen and the red-yellow-green should work in a single bulb signals.

one other item I got 5 sets of christmas lights for the house with 100 LED's for $4.99 for each box at CVS pharmacy so that about .05 cents each LED and thats not a bad price.

Dan

more LED stuff

Multicolour LEDs can sometimes be very useful - for searchlight signals - and for head/tail lights on trains. If anyone's thinking of using them, I'd advise checking data - either from suppliers' catalogues or (better still) manufacturers' datasheets - a lot of this stuff is available free on the internet (and extremely useful). The same goes for data on any electronic stuff you use - components, modules, or complete gadgets - you never know when you might need it.

You make an excellent point about LED Christmas lights. I also keep my eyes open for them - however, I mainly go for the white and blue ones - in the UK, these colours are extremely expensive if you get them from component places.

There isn't so much of an issue with other colours (but the ones in the sets are still probably much brighter than the basic ones sold cheaply by electronics suppliers).

By the way, thanks for the warning about exploding LEDs.

Regards,

Huw.

It appears that

It appears that LEDS are the more popular method from these posts. However, in my limited exposure to both bulbs and LEDS, I prefer bulbs for the asthetics. They just look like more real light to me. Heck they are real light! They have that longer distance glow and twinkle, where the LED is a light dot, for lack of a better term. I have installed 1.5 V, 1 milliamp bulb with resistor, and don't qoute me on these values, with 480 ohm resitors. The light was bright. I lost 2 bulbs out of about 10, after several months running, then rewired with 520?? ohm resitors, still bright enough. Most of my freinds use this resistor and the biulbs have been in use for a long time. I experimented with the need for a brighter bulb.

Anyway, it is my opinion about liking bulbs better. That includes Chrsitmas lights!!

As a side joke, now, when my visiting crew brought their engines over to run on my new DCC set up, I commented to them about how nice and bright the bulbs looked on their engines. Until they started to burn out after 30 minutes to an hour. Yes, the Digitrax switch was set to O gauge or space shuttle power or something, and not HO.  Not funny at the time, but as they say, you learn from your mistakes. 

Steve

However, in my limited

Totally agree with that.  LED lights look way too artificial.  Real locomotives didn't have HID headlights, neither should models. 

Especially amusing is seeing an HOn3 C-19 with a bright WHITE led light.

Also amusing is the individuals (usually rivet-counters) that will point out their prototypical air brake rigging, correct fans and hatches, expertly rendered cab and yet have LEDs, ummmm k, where is the warm glow of a YELLOW light like the REAL thing?

Josh 

Josh you don't have to use

Josh you don't have to use bright white LED's as there are a number of different colors, shades and styles of LED's including soft yellow lights if that's what your looking for and soft enough to look like the old reflector oil lamps on D&RGW, RGS, C&S and other RR using C-16, C-19 or any other Narrow gauge engine.

Dan

Josh you don't have to use

Hmmm... that's good to know.  However, I am doing just fine using Miniatronics 1.5 volt bulbs.  My gripe was with rivet counters and magazines harping the use of bright white LED lights in applications that have no use for it.  Other than to say "Hey, I have LED lights in my SD38-2 with ditch lights, yay!" 

My HOn3 days are over as well.  I now (again) am working on modeling the Moffat and Tenneesee Pass lines.

Josh