Jump to content
 

3v transformer


MOH
 Share

Recommended Posts

16 hours ago, MOH said:

Thank you,

 

I quoted a 3v transformer as the leds were advertised as only to be used with 3v, I have test wired three leds to a mobile phone charger and those are working with no resistors added and no blow outs so far.

 

Really no idea what voltage that mobile charger is rated as it's quite old and the details are blurred on the casing.

 

I will plough on so and order that speaker cable and see how that goes.

I do not know why people insist on trying to run LEDs without resistors, after all they are not the most expensive component neither do you need a vast collection of them. DC voltages associated with model railways tend to be 12V in the main. Other power sources often used such as ex-phone chargers are mostly 5V since all phone manufacturers and tablet makers have standardised on USB charging. Even the forward voltages of LEDs range from just below 2V to around 3.5V and since most modellers choose not to run their LEDs at full power the variety of resistors required is limited as long as the resistance exceeds the minimum calculated value.

 

If you are not sure of the resistance required and cannot do Ohm’s Law calculations there are plenty of free calculators out there. One of the simplest to use is Electronics2000 it can work out values from the barest of information.

 

 www.electronics2000.co.uk

 

Richard

 

 

  • Like 2
  • Agree 1
Link to post
Share on other sites

  • 3 weeks later...
  • 2 weeks later...

Hello, I am new to this forum and I would like to thank Forum Members for a great meeting place.

I am in the stages of constructing a simple L shaped layout in 4mm scale ( somewhere in the English Sth Coast area where I can run green painted locos) and as my supply of Grain of Wheat bulbs has been exhausted on previous layouts, I am playing around with LEDs for structure illumination. Great fun and some surprises for me - Yellow, Water Clear LEDs make great interior light sources giving a real 1950's low wattage glow through windows. I am playing around with resistor values to find the differing colours /intensities they provide. I am using 12v wall transformer for power source and my local supplier provides a handy R value suggestion at various input V for their extensive range of LEDs in their huge catalogue. Makes it easy !

My query is, can I set up a central supply point with say, a 5w wire wound resistor to the +ve output terminal , a common -ve terminal and then distribute power to each LED ? So, I need a 560 Ohm resistor to power the  LEDs off 12v DC  and calculating the total wattage limit - can I power a number of LEDs  to the limit off the 5w resistor ?

I am used to providing a separate LED to each resistor.

Your advice will be appreciated.

    

Link to post
Share on other sites

3 hours ago, lyctus said:

My query is, can I set up a central supply point with say, a 5w wire wound resistor to the +ve output terminal , a common -ve terminal and then distribute power to each LED ? So, I need a 560 Ohm resistor to power the  LEDs off 12v DC  and calculating the total wattage limit - can I power a number of LEDs  to the limit off the 5w resistor ?

If the following diagram is the circuit you are describing then my answer is no

778781878_Leds1resistor.png.3a8c0997442cb9c9b1653d1be9bcaefe.png

 

 

 

The following 2 diagrams should be ok for building lighting 

By adjusting the resistor values in the top diagram each LED's brightness can be adjusted by varying the value of its resistor. 

I would start with 1k resistors, if too bright then try 2k2, 4k7 or even 10k

 

In the lower drawing the 3 LEDs in series should be identical(from the same batch) for best results

I would start with a 1k & adjust the value up or down to get the required brightness

 

 

Leds.png.f469f79103af1b4cdfaab6fce1616345.png

Hope this helps 

John

  • Thanks 1
Link to post
Share on other sites

Using John KS's first diagram it is possible to have up to half an Amp's worth of LEDs, so at 20mA per LED you could have 25 identical LEDs in parallel fed with a 20R 5W resistor - but you will have to be very careful that none of the LEDs become disconnected or there will be a very bright display.

 

Much better to use a lower voltage power supply (5V is ideal and very, very cheap) and individual resistors for each LED.

  • Agree 1
Link to post
Share on other sites

2 hours ago, Suzie said:

Using John KS's first diagram it is possible to have up to half an Amp's worth of LEDs, so at 20mA per LED you could have 25 identical LEDs in parallel fed with a 20R 5W resistor - but you will have to be very careful that none of the LEDs become disconnected or there will be a very bright display.

 

 

Even if there are no disconnects it's not a good idea. If all the LED's are the same colour they will not have the same forward voltage. Some will "steal" more current and be brighter than others. In the most extreme case one or two might steal most of the current and become Flame Emitting Diodes (although that's fairly unlikely).

 

Link to post
Share on other sites

On 19/09/2020 at 21:31, AndyID said:

 

Even if there are no disconnects it's not a good idea. If all the LED's are the same colour they will not have the same forward voltage. Some will "steal" more current and be brighter than others. In the most extreme case one or two might steal most of the current and become Flame Emitting Diodes (although that's fairly unlikely).

 

If you look at the characteristic curves of leds the VF rises with current , Hence your sceanario will not occur 

Edited by Junctionmad
Link to post
Share on other sites

2 hours ago, Junctionmad said:

If you look at the characteristic curves of leds the VF rises with current , Hence your sceanario will not occur 

 

If you look at the data for LEDs you will observe that there is no minimum forward voltage. There are only typical and maximum values. Therefore, although it's unlikely, there is nothing in the data that says it cannot happen.

 

And they definitely will not all have the same intensity which was my main point.

Link to post
Share on other sites

6 hours ago, AndyID said:

 

If you look at the data for LEDs you will observe that there is no minimum forward voltage. There are only typical and maximum values. Therefore, although it's unlikely, there is nothing in the data that says it cannot happen.

 

And they definitely will not all have the same intensity which was my main point.

 

On 19/09/2020 at 21:31, AndyID said:

 

Even if there are no disconnects it's not a good idea. If all the LED's are the same colour they will not have the same forward voltage. Some will "steal" more current and be brighter than others. In the most extreme case one or two might steal most of the current and become Flame Emitting Diodes (although that's fairly unlikely).

 

 

Hardly flame emitting, just a brief super bright flash and then nothing, though I have seen them smoking so I suppose they could catch fire.    Worst case scenario with LEDs in parallel is a chain reaction as all the LEDs fail in sequence.

  • Agree 1
Link to post
Share on other sites

45 minutes ago, DavidCBroad said:

 

 

Hardly flame emitting, just a brief super bright flash and then nothing, though I have seen them smoking so I suppose they could catch fire.    Worst case scenario with LEDs in parallel is a chain reaction as all the LEDs fail in sequence.

 

Yes, most likely they'll just flash then pop. The FED (Flame Emitting Diode) is an old joke. It was featured on the front cover of EDN as an April Fool in 1983.

Link to post
Share on other sites

17 hours ago, Junctionmad said:

If you look at the characteristic curves of leds the VF rises with current , Hence your sceanario will not occur 

 

The problem is that the converse is also true. Applying a higher voltage will cause a higher current to flow.

 

If one LED is mismatched and generates a higher Vf, at a safe working current for that particular LED, then that Vf is also applied to the other LEDs which may then be operating at an unsafe current. Another LED pops, the Vf gets even higher, ...

 

Yes, it's a corner case, but it can happen, and is one of the reasons why parallel connected LEDs are always used with individual resistors by anyone doing it for a living.

 

Edited by Crosland
  • Agree 2
Link to post
Share on other sites

11 hours ago, Crosland said:

 

The problem is that the converse is also true. Applying a higher voltage will cause a higher current to flow.

 

If one LED is mismatched and generates a higher Vf, at a safe working current for that particular LED, then that Vf is also applied to the other LEDs which may then be operating at an unsafe current. Another LED pops, the Vf gets even higher, ...

 

Yes, it's a corner case, but it can happen, and is one of the reasons why parallel connected LEDs are always used with individual resistors by anyone doing it for a living.

 

It’s such a corner case it’s around the corner

 

in real life electronics with similar leds , one resistor will “ work”   Yes it has certain specific drawbacks but in practice it actually works. Yes it’s not a Recommended approach but for leds  from the same family operating well into their safe current zone  it will work  

 

and no I’m not recommending  it :D 

Edited by Junctionmad
  • Agree 2
Link to post
Share on other sites

8 hours ago, Junctionmad said:

 for leds  from the same family operating well into their safe current zone  it will work 

 

The band gap determines Vf. Band gap is a function of the doping chemistry and that's why there is a wide range of Vf in the specs. If all the LEDs happen to be from the same manufacturing batch you'll most likely get away with it.

 

Of course if you really want to avoid the enormous expense of all those irritating resistors and you want all the LEDs to have the same intensity just string a lot of LEDs in series and feed them from something like 20 volts through a simple constant current regulator.

  • Agree 1
Link to post
Share on other sites

On 23/09/2020 at 05:40, AndyID said:

Of course if you really want to avoid the enormous expense of all those irritating resistors and you want all the LEDs to have the same intensity just string a lot of LEDs in series and feed them from something like 20 volts through a simple constant current regulator.

That's the problem, if they aren't the same spec they won't give the same intensity, 

Also fault finding can be a nightmare, remember the old screw in bulb Christmas lights, shorting the faulty bulb with silver paper when you couldn't get a spare on Xmas eve.

Link to post
Share on other sites

18 hours ago, DavidCBroad said:

That's the problem, if they aren't the same spec they won't give the same intensity, 

Also fault finding can be a nightmare, remember the old screw in bulb Christmas lights, shorting the faulty bulb with silver paper when you couldn't get a spare on Xmas eve.

 

So don't use LEDs with different specs. The one thing you can guarantee is that they will all pass the same current and it's the current that determines the light output, not the voltage.

Link to post
Share on other sites

1 hour ago, AndyID said:

 

So don't use LEDs with different specs. The one thing you can guarantee is that they will all pass the same current and it's the current that determines the light output, not the voltage.

While there is definitely some difference in illumination , in many cases it’s irelevant ( mimics etc ) for example for blue , red, green and yellow I use 1k8 on 5V. Works fine 

  • Agree 1
Link to post
Share on other sites

On 19/08/2020 at 21:31, MOH said:

Thank you,

 

I quoted a 3v transformer as the leds were advertised as only to be used with 3v, I have test wired three leds to a mobile phone charger and those are working with no resistors added and no blow outs so far.

 

Really no idea what voltage that mobile charger is rated as it's quite old and the details are blurred on the casing.

 

I will plough on so and order that speaker cable and see how that goes.

Watch out for the charger running hot, If the LEDs are taking excess current they will glow very brightly, too many on the charger and it will get hot. Charger is probably 2.5 or 3 volt as a 5volt  will usually incinerate a Yellow LED pretty instantly (if you don't use a resistor.) You may need resistors in the 10K or 100K range if the LEDs are too bright.  My 3 volt wall warts have run strings of LEDs in parallel as building lights for 10 years or more with the transformers running warm after two hours my only concern,  (but no warmer than my laptop chargers which run 24/7/365)

I woudn't like to forget and leave them on for a week.

Bell wire has too much thick insulation for use in buildings, OK for an LED Bus but it is pretty feeble load wise . I use phone wire actually inside buildings and street lamps etc. 

see also   https://www.waveformlighting.com/pcb-designs/when-and-why-do-leds-need-current-limiting-resistors

Edited by DavidCBroad
Crossland
Link to post
Share on other sites

On 30/09/2020 at 13:58, Crosland said:

 

Unless you abuse them a LED is unlikely to fail in any modellers lifetime.

 The OP is using  " single leds plus street lamps and platform lamps to try and illuminate areas of my layout," so it would be the connections which would give trouble not the LEDs.  I use LEDs the same way and some give trouble when buildings are disturbed, but mine are in parallel which simplifies trouble shooting.

Link to post
Share on other sites

On 23/09/2020 at 03:55, Junctionmad said:

It’s such a corner case it’s around the corner

 

in real life electronics with similar leds , one resistor will “ work”   Yes it has certain specific drawbacks but in practice it actually works. Yes it’s not a Recommended approach but for leds  from the same family operating well into their safe current zone  it will work  

 

and no I’m not recommending  it :D 

As in so many technical areas, there is considerable wriggle room between best/recommended practice and what you can get away with. Even as an engineer, I've always found the latter area far more interesting than the former. Largely because I'm a lazy cheapskate who enjoys annoying people who tell me things won't work :D

 

I do, however, subscribe to the belief that it's important to know what the rules are, and why, before you break them, so you can do an informed mental cost-benefit analysis. I don't support bodgery for its own sake. There must be some benefit, however marginal. 

  • Agree 3
Link to post
Share on other sites

On 03/10/2020 at 04:31, PatB said:

As in so many technical areas, there is considerable wriggle room between best/recommended practice and what you can get away with. Even as an engineer, I've always found the latter area far more interesting than the former. Largely because I'm a lazy cheapskate who enjoys annoying people who tell me things won't work :D

 

I do, however, subscribe to the belief that it's important to know what the rules are, and why, before you break them, so you can do an informed mental cost-benefit analysis. I don't support bodgery for its own sake. There must be some benefit, however marginal. 

 

Everyone should print this out and stick above their workbench :)

 

Whether it's LEDs, frog polarity switching, or, ...

 

 

Link to post
Share on other sites

On 03/10/2020 at 03:58, DavidCBroad said:

Watch out for the charger running hot, If the LEDs are taking excess current they will glow very brightly, too many on the charger and it will get hot.

 

LEDs do not take excess current unless you abuse them. Use a resistor.

 

Quote

Charger is probably 2.5 or 3 volt as a 5volt  will usually incinerate a Yellow LED pretty instantly.

 

Only if you abuse the LED. Use a resistor.

 

Quote

You may need resistors in the 10K or 100K range if the LEDs are too bright. 

 

You ALWAYS need resistors with LEDs.

 

Please, for everone's sake, stop making these posts about driving LEDs without resistors.

 

  • Like 1
  • Agree 6
Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...