Colour light polarity

[DCB]

I suspect you mean at the same 'If', not 'Vf', as 'Vf' is determined by the LED chemistry to produce the various colours. From a 12V supply, the 'If' difference between a LED with 2.0V 'Vf' and 2.2V 'Vf' is virtually irrelevant if they are fed independently via their own identical resistors. Their perceived brightness may well be different for the same 'If', not only because one type might be more efficient than the other, but the eye's perception of brightness varies with colour too. As has been said many times before, with LEDs you need to be considering the current they pass, not the voltage across them.

Sorry Gordon but you mis understand if you think the difference between 2.0 vf  and 2.2 vf is negligible, it is actually very significant and the Led will be much more than twice as bright on 2.2 volts as on 2 many greens wont actually light at all on 2 volts.

 

The problem is you don't understand that you use resistors to control the vf  the voltage, the LEDs ability to pass current is entirely dependent on voltage.  Connect an LED across a big 1.5 volt battery and wait for the bang, you will have a very long wait.  Connect it across a small 9 volt battery and it pops instantly.    If it was dependent on resistors it would pop on 1.5 volts.

 

I know you guys don't have controlled voltage power supplies but I do.  

 

Crossland says about variations in Coloured LEDs but the problem is the spec sheets tell a different story. Reds are  brighter than Greens, massively so in some ranges, it is due btothe different chemical composition between red 

 

I checked three resistance  power units, 

Hornby Dublo 23 volts on 12 Volt DC controlled 19 on 16 volt AC uncontrolled.

Triang 21 volts  on 12 volt controlled

Hornby 21 volts on 12 volts controlled, all on the minimum setting but no difference between minimum and maximum

 

And my On track  variable voltage 0.3 volts minimum 16 volts max.   The Ontrack happily lights Maplin 5mm LEDs on a fraction of a turn, however the Red lights first and is too bright before the Green glows.

[kevinlms]

 

I checked three resistance  power units, 

Hornby Dublo 23 volts on 12 Volt DC controlled 19 on 16 volt AC uncontrolled.

Triang 21 volts  on 12 volt controlled

Hornby 21 volts on 12 volts controlled, all on the minimum setting but no difference between minimum and maximum

 

And my On track  variable voltage 0.3 volts minimum 16 volts max.   The Ontrack happily lights Maplin 5mm LEDs on a fraction of a turn, however the Red lights first and is too bright before the Green glows.

 

Well there is your problem, an unregulated power supply, means that you have no known starting point. Everything after that is guesswork. Not only can't you be sure of the 'no load' voltage, but also the voltage at various part controller settings or loadings. That's why is called 'unregulated'.

 

Much better to start from a regulated power supply & if need be, adjust resistors to suit each colour independently.

[DCB]

Well there is your problem, an unregulated power supply, means that you have no known starting point. Everything after that is guesswork. Not only can't you be sure of the 'no load' voltage, but also the voltage at various part controller settings or loadings. That's why is called 'unregulated'.

 

Much better to start from a regulated power supply & if need be, adjust resistors to suit each colour independently.

Er that is what I am suggesting. 90% of readers are using unregulated power supplies which they erroneously believe to be regulated 12 volt.  All the transformers I checked were marked 12 Volt yet delivered around 20 volts, except my OnTrack (which is like a Morley) and various Hammant and Morgan Safety Minors which are variable transformers.  The problem with powering LEDs in parallel from a "12volt" supply is even in the best case scenario the resistor dissipates 80% of the power,   See screenshot

 

Connecting in series is better but hardly practical for signals so I use low voltage supplies and drive most of my LEDs without resistors, and lets face it a 1 Ohm resistor is not really worth bothering with. See Pic,   I drive the signal LEDs from a 2.5 volt supply using resistors to balance the colours, very different ones for green than for red.

[Gordon H]

Sorry Gordon but you mis understand if you think the difference between 2.0 vf  and 2.2 vf is negligible, it is actually very significant and the Led will be much more than twice as bright on 2.2 volts as on 2 many greens wont actually light at all on 2 volts.

 

The problem is you don't understand that you use resistors to control the vf  the voltage, the LEDs ability to pass current is entirely dependent on voltage.  Connect an LED across a big 1.5 volt battery and wait for the bang, you will have a very long wait.  Connect it across a small 9 volt battery and it pops instantly.    If it was dependent on resistors it would pop on 1.5 volts.

 

I know you guys don't have controlled voltage power supplies but I do.  

 

Oh dear, here we go again.

I'm afraid it is you that misunderstands the operation of LEDs, not the numerous people here who have to spend so much time and effort putting you right on a regular basis.

You really should stop promoting your fundamentally flawed concept of controlling LEDs by applying an exact voltage to them rather than controlling their current in a proper manner. You seem to be the only one wanting to do it that way - does that not tell you something?

As has been said to you ad nauseum, LEDs are current controlled devices. That they drop a particular and narrow range of voltage when passing current is a side effect of them being a semiconductor diode as well as a light source.

You DO NOT use resistors to control the Vf - you use them to control the If. Please stop saying otherwise.

Crosland is absolutely right - anyone reading anything you write regarding LEDs would be well advised to ignore it.

[dhjgreen]

David, Can you give us the URL for the guru website you have shown above. The selection of an applied voltage just 0.1v above the typical vf of an LED leaves very little scope for tolerance and regulation variation. There is possibly some merit in reducing the total power dissipated in the circuit as they seem to be recommending, but a constant current device would achieve the same outcome without the tight tolerance issues. There are a number of quite wierd on-line so-called electronics expert sites that propose circuits and concepts that are outlandish and often simply do not work.

[Gordon H]

The selection of an applied voltage just 0.1v above the typical vf of an LED leaves very little scope for tolerance and regulation variation. There is possibly some merit in reducing the total power dissipated in the circuit as they seem to be recommending, but a constant current device would achieve the same outcome without the tight tolerance issues.

 

David has been told this kind of thing umpteen times, but seems unwilling to accept it. The concept of constant current would appear completely alien to him in his world where voltage is the only thing to consider.

[Grovenor]

David, Can you give us the URL for the guru website you have shown above. The selection of an applied voltage just 0.1v above the typical vf of an LED leaves very little scope for tolerance and regulation variation. There is possibly some merit in reducing the total power dissipated in the circuit as they seem to be recommending, but a constant current device would achieve the same outcome without the tight tolerance issues. There are a number of quite wierd on-line so-called electronics expert sites that propose circuits and concepts that are outlandish and often simply do not work.

I think this is it, http://ledcalc.com/ It looks to be just a calculator, not proposing any solutions, it is David who is doing that.

Regards

[DCB]

No once again, the resistor is to limit the forward voltage.  Not the current.   The current is always proportional to the voltage.  Reduce the voltage and the current reduces.  If you feed an LED with the voltage equal to its rated vf it will pass the rated current without a resistor, add a resistor and it will pass less than its rated current.    That is exactly backwards to what some people are suggesting.  

 

The theory propounded by various people would see LEDs passing huge currents at their rated forward voltage if resistors were not provided, This simply is not the case.

 

Going back to the very first post the guy is using a Hornby controller he believes is 12 volt whereas it is probably nearer 20 volts.

[dhjgreen]

No once again, the resistor is to limit the forward voltage.  Not the current.   The current is always proportional to the voltage.  Reduce the voltage and the current reduces.  If you feed an LED with the voltage equal to its rated vf it will pass the rated current without a resistor, add a resistor and it will pass less than its rated current.    That is exactly backwards to what some people are suggesting.  

 

The theory propounded by various people would see LEDs passing huge currents at their rated forward voltage if resistors were not provided, This simply is not the case.

 

Going back to the very first post the guy is using a Hornby controller he believes is 12 volt whereas it is probably nearer 20 volts.

This is, as others have said, quite wrong. I will now take my Diploma in Engineering (Electrical and Electronic) and bow out, sayonara

[olivegreen]

This is, as others have said, quite wrong. I will now take my Diploma in Engineering (Electrical and Electronic) and bow out, sayonara

 

Having just read all of this, I am tempted to join you.

Methinks someone here might benefit from going back to school...

[Gordon H]

No once again, the resistor is to limit the forward voltage.  Not the current.  

 

Who was it that actually told you this?

Or is it something you have dreamt up yourself based on a misunderstanding?

Please try to take it from the numerous people on this forum who are all agreed that your concept is deeply flawed, and you are ploughing a lonely furrow in pursuing it - if you're not careful to the point of becoming a laughing stock.

[dhjgreen]

I think this is it, http://ledcalc.com/ It looks to be just a calculator, not proposing any solutions, it is David who is doing that.

Regards

I am not coming back. Just to point out this joker is offering an option of calculations using parallel LEDs, so is in my category of falsely "self-styled" expert.

 

Edit: not you Keith, the led calc website.

[Grovenor]

No once again, the resistor is to limit the forward voltage.  Not the current.

Wrong as others have said above.

 

 

The current is always proportional to the voltage.

True for resistors, not for LEDs, try looking at the data sheet for an LED.

 

Reduce the voltage and the current reduces.  If you feed an LED with the voltage equal to its rated vf it will pass the rated current without a resistor, add a resistor and it will pass less than its rated current.

True, more or less, but not relevant to normal use of LEDs.

 

That is exactly backwards to what some people are suggesting.[/quopte]

No-one has suggested anything that is exactly backwards to this.

The theory propounded by various people would see LEDs passing huge currents at their rated forward voltage if resistors were not provided, This simply is not the case.

No it would not, try reading more carefully, and in relation to the data sheets.

Regards

 

[Crosland]

Er that is what I am suggesting. 90% of readers are using unregulated power supplies which they erroneously believe to be regulated 12 volt.  All the transformers I checked were marked 12 Volt yet delivered around 20 volts,

A transformer is not a power supply.

 

Connecting in series is better but hardly practical for signals so I use low voltage supplies and drive most of my LEDs without resistors,

Which is all the proof we need that you do not understand.

[Crosland]

If you feed an LED with the voltage equal to its rated vf it will pass the rated current without a resistor, add a resistor and it will pass less than its rated current.

The Vf of a LED is not a fixed parameter. It varies with the current that you allow to flow through it, just like any diode.

 

The theory propounded by various people would see LEDs passing huge currents at their rated forward voltage if resistors were not provided, This simply is not the case.

The reason you don't see LEDs blowing up, and you have been told this before, is because you are (unwittingly or ignorantly) relying on the internal resistance of your low voltage source to limit the current.

[nigelb]

there is a simple answer to this  use grain of wheat bulbs and paint them green red or yellow still need 3 wires and a Change over switch

[reevesthecat]

Thanks for the advice everyone I have managed to get more than 1 light working at last by standardising on make of light.

Cheers

Mark