DCB Posted April 21, 2014 Share Posted April 21, 2014 Is it possible to obtain or make miniature voltage regulator small enough for the guards compartment of an OO coach which will deliver 2.7 ish volts DC output (or 4 or 5 volts) over a range of 6 to 20 volts DC input? Having run a Triang 2 coach suburban "B Set" and an Airfix Autocoach with LED lights for many years with no problem I tried them on a wet dark garden layout last Saturday and ran into problems, The poor old Hornby Dublo 0-6-0 with a 64XX Pannier body could hardly drag itself along with the 3 lighted coaches but was fine on 20 wagons, slipping furiously on the wet track. On investigation the power unit was only delivering 14 volts on full power instead of the 19 volts mine deliver. This set me thinking, first should i rewire the LEDs to series instead of Paralell to reduce the current, or should I try to find a current regulator which will deliver 2.7 ish volts output over a range of 6 to 20 volts DC input, as above?t Link to post Share on other sites More sharing options...
Suzie Posted April 21, 2014 Share Posted April 21, 2014 Adding a regulator will not reduce the load on the inadequate controller, but changing to series wiring (with a smaller value resistor) will. Realistically you might like to look at replacing the controller with something a bit more up to date than the very low power variable resistance unit you are using. It should provide full power when the knob is at maximum and not be fazed by a couple of hundred milliAmps of extra load that the 'B' set lighting will be drawing. For best results add a regulator to give you constant lighting brightness and get a new controller. A transitorised controller, either linear or PWM will be fine. A simple regulator made with a 78L05 using the manufacturers recommended components will be ideal. Use a bridge rectifier made from 1N4148 diodes to rectify the power from the track. Link to post Share on other sites More sharing options...
Gordon H Posted April 21, 2014 Share Posted April 21, 2014 To answer the original question, you don't want a voltage regulator for DC LED lighting at all - you want a current regulator instead. Link to post Share on other sites More sharing options...
Brian Posted April 21, 2014 Share Posted April 21, 2014 Hi This may help?.....http://www.pollensoftware.com/railroad/ Link to post Share on other sites More sharing options...
Suzie Posted April 21, 2014 Share Posted April 21, 2014 I would stick with a voltage regulator when running the LEDs in parallel, it is not always possible to guarantee that the current sharing will be correct on all the LEDs so best to use a voltage regulator a bit higher than the LED voltage such as 5V and use individual current limit resistors typically somewhere between 120R and 470R but you might need higher values if they are too bright. Link to post Share on other sites More sharing options...
DCB Posted April 22, 2014 Author Share Posted April 22, 2014 Sounds like the LM334Z is the way to go, thanks Brian. Link to post Share on other sites More sharing options...
Gordon H Posted April 23, 2014 Share Posted April 23, 2014 Note that the LM334Z is good for up to 10mA. If you need more than this you can use something like an LM317LZ in current regulator configuration (essentially the same as the LM334Z set up with a single resistor), which would give up to 200mA. Link to post Share on other sites More sharing options...
Crosland Posted April 23, 2014 Share Posted April 23, 2014 Sounds like the LM334Z is the way to go, thanks Brian. A poor choice for more than one LED. Series connection will require an increased minimum voltage, parallel connection wil overload the device. The author of the article referred to by Brian is another one who clearly does not understand the meaning of absolute maximum ratings on data sheets. Andrew Link to post Share on other sites More sharing options...
DCB Posted April 23, 2014 Author Share Posted April 23, 2014 Or maybe an LM317LZ in current regulator configuration! (Or maybe I should give up on the old pair after all they are over 50 years old) Link to post Share on other sites More sharing options...
jdp298 Posted April 27, 2014 Share Posted April 27, 2014 I'd push this a little further. Use an LDO, but make sure it's as ultra-low as you can get, and if necessary run 2 or 3 in parallel. With that you can then also put in a super-cap or 2, dependending on your space, and the lights will stay on when you stop, or even put the train away for a bit. It's something I've used for years on my bikes with hub dynamos, and to prove it's really not a large installation: Link to post Share on other sites More sharing options...
DCB Posted April 27, 2014 Author Share Posted April 27, 2014 Thanks everybody, it has given me some ideas around voltage and current regulation and not just for on board, the LM 317LZ, sounds good. Interestingly it highlighted the disagreement between Crossland and myself, The article referred to by Brian Lambert says "LEDs require a certain voltage to make them start working and past this will consume all available current" The voltage is not far wrong but they certainly do not consume "All the available current", my test LED consumed 16,6 ma at 2.04 volts and 101.7 ma at 2.8 volts almost 10X the current for a 33% increase in voltage. That power unit will deliver over 1 amp (1000ma) on that 2.8 volt setting. There is a very sharp bend in the current vs voltage curve for an LED around the 2 volt range which is irrelevant to 99% of users but not to me as I have a variety of power packs capable of delivering voltages in 0.1 volt increments in the 2 to 4 volt range. What will really upset the experts is what happened when I accidentally turned the control knob the wrong way and put 8.84 reverse voltage through the LED. It lit quite normally, and suffered no ill effects from the brief trial despite getting too distressed to test on 3.5 volts forward, now nothing I have ever read has suggested LEDs light normally when reverse voltages are applied,, everything suggests they fail instantly. Maybe I should experiment further...... Link to post Share on other sites More sharing options...
RMweb Gold SHMD Posted April 27, 2014 RMweb Gold Share Posted April 27, 2014 I give up... Link to post Share on other sites More sharing options...
Gordon H Posted April 28, 2014 Share Posted April 28, 2014 What will really upset the experts is what happened when I accidentally turned the control knob the wrong way and put 8.84 reverse voltage through the LED. This says it all about the fundamental lack of electronic engineering knowledge from this particular source. You don't put volts through anything - not even a dead parrot (in spite of what the script of the sketch might suggest). Link to post Share on other sites More sharing options...
Crosland Posted April 28, 2014 Share Posted April 28, 2014 What will really upset the experts is what happened when I accidentally turned the control knob the wrong way and put 8.84 reverse voltage through the LED. It lit quite normally, and suffered no ill effects from the brief trial despite getting too distressed to test on 3.5 volts forward, now nothing I have ever read has suggested LEDs light normally when reverse voltages are applied,, everything suggests they fail instantly. Maybe I should experiment further...... No matter what you think you did or saw, normal LEDs do not light normally when connected in reverse. Perhaps you have an AC LED. No reputable source would claim instant failure, without qualification. The end result depends on a number of factors. As with any diode, little or no current will flow and nothing bad will happen until the reverse breakdown voltage is exceeded, typically quoted as worst case (i.e. lowest) around 5V for an LED as opposed to 50V+ for common diodes. Andrew Link to post Share on other sites More sharing options...
DCB Posted April 29, 2014 Author Share Posted April 29, 2014 I will have to do some more experiments, meanwhile I have a picture of it lit on 8.84 volts connected backwards... Forwards it was distressed on around 3.5 volts so it is all very weird. It is probably to do with my voltage being controlled by the ratio of primary coils to secondary coils in my power unit (Hammant and Morgan Safety Minor Variable Transformer) rather than by resistors. Link to post Share on other sites More sharing options...
Gordon H Posted April 29, 2014 Share Posted April 29, 2014 More likely to be an effect of reverse leakage in the selenium rectifiers that H&M used in those days. Try doing the same with a pure DC supply and you will undoubtedly get a different result. Alternatively, look at the controller output with an oscilloscope and see what it is actually producing. Link to post Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.