I was curious if anyone knew of a vendor who makes/sells high voltage LEDs. I have a 25 volt power supply, and for various reasons don't want to step down the voltage to ~12 volts with a voltage regulator.
If it helps, or makes a difference, I'm looking for an IR led with a wavelength of about 830 nanometers, but would like to hear of any suggestions that people have for LED vendors.
Seriously, even if such a high voltage LED existed, you would need to current regulate, since the forward drop will change with temperature. Ideally, you need to get an LED with drive electronics built in. 25v units are quite common, since trucks use nominally 24v electronics, and many now use LED rear lights, and instrument panel lights like this. Many are made by SLi
formatting link
and Agilent, but these companies rarely do IR LEDs, so it does make a large difference that this is the frequency band needed. You don't say what sort of brightness you require. Remember also, that you can series connect several LED's, and minimise the resistance that has to be used. So (for instance), you could series connect perhaps 9 IR LEDs, and then have perhaps a 25R resistor to limit the current (you would need to check the typical forward voltage at the rated current, but normally this is about 2.4v, for the AIGaAs/GaAS Technology normally used at this frequency). This would be for a 100mA rated LED, like the QLD-830-100S. Some of the LED manufacturers do offer small switch mode drive modules that will cover this voltage range.
The operating voltage of a single LED is typically around 1.6 Volts. This is determined by semiconductor physics.
You can run a single LED by adding a series resistor which sets the current. It wastes most of the power in the resistor though.
You could run, say, 12 LEDs in series ( operating voltage approx 19.2 V ) and lose less power in the current limiting resistor.
'LED lamps' use multiple LEDs like this. A.C. versions also require a rectifier to provide DC to the LEDs. LEDs only work with current flowing in one direction, so need D.C. current.
In short, there is no *high voltage LED* ( except several LEDs in series ) - you have to work within the constraints of science.
'Clever circuitry' can be used to optimise efficiency - but you're clearly not operating in this region of expertise.
Or a combination of series LEDs and a resistor, or (if efficiency is critical) a switched mode regulator. I've seen some app notes for LED drivers, but not having done any LED work recently, I can't lay my hands on the part numbers as quickly as a good search engine might.
--
Paul Hovnanian mailto:Paul@Hovnanian.com
note to spammers: a Washington State resident
I got some ultrabright whites the other day. Not much info on 'em, though. Typical values only given; no maximums. Vfwd was 3.4V; Ifwd was 30mA. Can anyone hazard a guess as to what the *do not exceed* figures might be? They tend to draw more current as they get hotter and when bunched close together and they do generate quite a bit of heat. I'm worried I might exceed the 30mA (typ.) figure by too much and blow 'em.
--
"What is now proved was once only imagin'd." - William Blake, 1793.
Not the voltage of the aircraft, and probably not the voltage going to the socket, but they certainly had to change the voltage going to the diode(s) proper (said circuitry would likely be part of the LED assembly, if they wanted it to be a "drop-in" replacement).
--
John Miller
Email address: domain, n4vu.com; username, jsm
It's worth getting Kingbright's LED Application Note, it has current/voltage curves & max specs for most led chemistries, plus tips on fitting, soldering, etc.
formatting link
The 3 white types they produce are each 30mA max If, though you can go considerably higher by pulsing them. But heat is still the problem.
--
email address is a black hole. Change cpemma to cpm.
You should also have thermal resistance data and maximum junction temperature. At least most of these figures are similar enough for "usual" 5 mm LEDs - the thermal resistance from the junction to 5 mm outside the case on the lead that the chip is mounted to is generally around 250 degrees C per watt. Temperature rise of the chip above the temperature at that point is thermal resistance times the amount of power being dissipated in the chip. If you want really long life, have the junction temperature not exceed
85 degrees C. If you can tolerate a shorter life, it's OK to go up to around 100, maybe 110 degrees C or so.
Note that the junctions in a cluster lamp can sometimes get that hot at less than 30 mA through each LED.
The point is that LED assemblies have long been made as a replacement for any incandescent indicator lamp. Same even for traffic lights whereby LEDs now replace incandescent lamps.
The original poster wants an LED light that can operate at
25 volts. Do they make same incandescent lamp replacements for 830 nm LEDs? I don't know. But back when we were first purchasing these incandescent lamp replacements, they were willing to do any existing LED color as a replacement for any incandescent lamp. Often these assemblies contain four or more LEDs meaning they also have good intensity.
LEDs are current, not voltage driven. In other words, you need a current limiting resistor in series with the LED. You can calculate the resistance required by (Vin - Vled)/ current, current is usually 20 mA. If the voltage is likely to change, or if you are to lazy to do the calculation, use a FET as constant current source, I have posted the schematic here a couple of times, use google.
You can also buy LEDs with that FET integrated on the chip, these work on any DC source up to 30 V.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.