my new whats my transformer qiuz

@ John, have you noticed how my mate's Enlite LED bulb provides constant light output from about 6v dc (using a bench power supply) to all the way to 16v dc, ...
... which is, of course, exactly what one would expect if the 'lamp' itself contained some sort of electronic constant-current circuitry.
This means that Enlite LEDs are using an altogether different technique to drive their LED lamps, and as I said it radiates some noise on MW radio receiver so that would indicate some kind of switch mode operation to regulate the output power (voltage and current) from the input power applied, so if the input voltage is high, its converter will use less current and produce constant power to drive the LED, at 16volts it was drawing 300mA from the source.
As above, I assume that the lamp must contain constant-current electronics but, as you go on to say, it's quite probably more complicated than that....
Of course this can be achieved quite easily by applying a constant voltage to the current source driver for the LEDs, such that secondary voltage must remain constant, at a predetermined value, so take an example say there are 4 banks of LEDs and each bank has 4 LED s in series, 16 total LEDs, each bank will require a minimum of 12volts as each LED needs 3v forward voltage and each bank set to pass 100mA maximum at 12V, so 4 banks would need a total of 400mA at 12v.
... simple 'passive' constant-current circuitry, per se, cannot give an output voltage greater than the input voltage, so it probably would not work with 5V /6V input with even groups of two LEDs in series, let alone more. It would, of course, theoretically be possible for all the LED elements to be in parallel (not arranged as groups in series), in which case it probably would work down to 5V/6V input.
Now to get this 12v constant, you would need a converter circuit that converts any input voltage from say minimum of 5 volts to all the way to 20volts and regulates it to 12v before sending it to the LED current driver.
IF the LEDs are arranged in 'groups in series' and IF it works down to 5V, then, as you say, one assumes that there must be a 'voltage step-up' facility within the lamp (certainly if there are more than two LEDs in series) - presumably some sort of SMPS-like circuitry (hence the MW noise you have noticed). However, that all sounds rather complicated - are these lamps expensive?

Kind Regards, John
 
Sponsored Links
Is the problem with these tiny tiny units that a better means of control would be too big to fit
Possibly - but electronics can be remarkable small these days, so I think it's more likely to be a cost issue. Resistors, in bulk, cost virtually nothing.

Kind Regards, John
 
You got me wrong, I agree that LEDs do work with a minimum forward voltage of 3 volts for white LEDs, so if you have say 4 in series, you will need 12volts to overcome forward voltage and a bit more to allow some voltage to drop across a current limit resistor, so let us say we need 12.5V so that 0.5volts can dispense across a current limit resistor, so if we were to pass let us just say 100mA for the sake of making a point, then the value of resistor can be calculated by ohms law, where R= V/I , so 0.5/0.1 (100mA) = 5 ohms, and the power dissipated in that resistor will be V x I = 0.5 x 0.1 = 1/2 watt resistor.

However, you are saying how come that I claim that if a bank of 4 LEDs require 12.5V then how can they work on an input supply voltage of as low as say 6 volts, that was my point that Enlite LED uses a complex circuitry to boost input voltage in an inverter like circuit build inside their lamp, it can take low voltage and boost it to a minimum of 12.5v but at the expense of requiring more current, that is why I stated that i found it strange that when I was bench testing my mate's Enlite bulb, that at 16v it was drawing just 300mA, at 12v it was drawing 420mA, but then at lower voltages like 5.5v or 6v it was drawing over an Amp, (1000mA) So it surprised me as I was expecting it to draw less at lower supply voltage, as did not think it would be much more complex than having just a few diodes and resistors to limit current, or at the most a constant current source.

So that is exactly what i was trying to say, that these characteristic are not that of a constant current source, but rather a different technique, hence I said that these Enlite LEDs have a complex circuitry, and by complex I mean it may just be a tiny chip with a Mosfet switching transistor that boosts input voltage and brings it to a required voltage to drive its 4 Leds wired in series, and so on.

It is likely that it uses a power converter circuit, (converting voltage to current and current to voltage) where it automatically draws more current to maintain 12.5v supply to LEDs, and if input voltage went higher than the required 12.5v then the converter circuit (which it actually is an inverter) draws lower amount of current to maintain the required 12.5v for the LEDs.

A converter circuit uses external inductors and capacitors, and schottky diode to convert low or high voltage to a desired output voltage, through feedback it can regulate when to go into action and when to relax, and I used the term Converter rather than an inverter, but either could be applied here, (normally converters are those that take DC 12 volts from car batteries as an example and convert it to AC 230V) as energy is being transformed I prefer calling it a converter, any way, one may even call it a switched mode power supply, a converter or an inverter transforms power and can generate a steady voltage at the expense of current, it can transform a low input voltage to a higher as in a converter, it can transform a higher voltage into a lower as in an inverter.

A simple converter or if you prefer an inverter circuit works by rectifying an ac input and converting it into DC, then chopping up dc input supply and pushing it through through an inductor to charge a capacitor, a capacitor absorbs the current as it starts to charge up slowly, converts the current being pumped into it into voltage. when you pump current into a capacitor its terminal voltage starts to rise, just like a rechargeable battery, the terminal voltage may be zero to start with, when you start pumping in current into it, it will start charging up, and faster the chopping higher the transfer of energy, and its terminal voltage will start to rise, until it is fully charged to the desired voltage, that is how some inverters are made to pump charge into capacitors and by using steering diodes one can also make a voltage doubler,

When a current passing through an inductor stops abruptly, the collapsing magnetic field generates reverse flow of current, (Back EMF) this back EMF is steered into the capacitor via a schottky diode and boosts voltage further, and creates higher transfer efficiency in energy transfer. so the capacitor will charge up to designed voltage (i.e. 12.5v) and when it reached the desired voltage it signals (feeds back) to the chopper circuit to slow down, as no more current is needed to maintain desired voltage across the capacitor terminal, so when a load is placed across capacitor, its voltage will try to go down as current is being drawn out by the load, this immediately signals the chopper circuit to speed up and start pumping in more current (energy transfer) and keep pumping up the capacitor to maintain the desired voltage, this chopper circuit would work on a wide variation of input voltage, and can generate higher voltage than its input voltage but at the expense of higher current demand when the input voltage is low, and lower current demand when the input voltage is higher than desired voltage.

One can have a converter/inverter working on a very wide range of voltage, but individual components must be able to handle higher voltages and currents.


it is interesting to discuss about things in general. That enlite bulb cost just under a fiver.

Now you can also see why such lamp will flicker as when it requires a sudden rush of current then the 12v SMPS may not be able to deliver a sudden rush of amps to satisfy the sudden demand made by an Inverter inside an LED lamp, so the LEDs are starved of current and dim or switch off completely and relight again when the SMPS has delivered enough current to switch them back on fully and the same thing happens again as LEDs switch on fully, current demand increases again and SMPS can't deliver it in time, and so the flickering cycle continues. If the SMPS produces a DC output, one can possibly add a large capacitor of a few thousand microfarads to allow for current surge to be met, I haven't tried this but when I get an opportunity this could be tried, keeping polarity of capacitor correct or else it could vent, and using something like 10,000uF 16V might work good. But, not necessarily, since even the capacitor will require an inrush of current, so the SMPS may go into overcurrent protection mode and starve the supply of current and LED lights may already start drawing the rest of the current and so SMPS may never deliver enough current to top up the capacitor as well as at the same time supply the demand made by LED inverter, so either allow the SMPS to fully charge up before connecting LEDs.

Its like trying to charge a car battery with a 5 amp charger and you have got headlights left on, so any charging current will end up neither charging the battery nor enough current to supply the headlights which may require 10 amps.
 
Last edited:
I have just brought some gu10s at £3.49 from led hut 5W 370lm cool white -- very good
 
Sponsored Links
... that was my point that Enlite LED uses a complex circuitry to boost input voltage in a converter like circuit build inside their lamp, it can take low voltage and boost it to a minimum of 12.5v but at the expense of requiring more current ...
That was my point, too - but I'm not sure I understand the reason for going for that additional complexity, when one could put less LEDs in series (or not put them in series at all) and then not need any voltage conversion, even if one wanted it to work down to 6V or so. However, even though one can get, or fabricate, voltage converters in tiny packages these days (hence could use three or four LEDs in series, I'm not sure why one would want it to work down to 6V (other than for your bench tests!), if it's designed to run from a 12V source!

Kind Regards, John
 
12 volt power supply AC will give 0 ~ 16 volt so if designed for 12 volt you will still get some voltage at 8.5 volt, assuming 9 volt LED then they would still work at 6.3 volt and that is assuming sine wave with other wave forms can be greater. This using of unknown wave form AC is what worries me with extra low voltage LED packages. Just too many unknowns. When you start to look at driver chips
hy3015.jpg
one starts to realise how small they are and how cheap and in some cases unless you can read the writing on the chip working out if a full wave rectifier or a driver chip would be hard.

This chip has just 4 connections
hy402.jpg
when I looked at this web site it made me realise how many options there are, I think it would be quite a hard job to reverse engineer any LED lamp which has a chip unless you can read the markings on the chip. I paid £2.77 including postage for a R7s J78 Energy Saving White 24 SMD LED Halogen lamp bulb 5W replacement by Fly Service it has a voltage range of 85~265 volt and 400 lumen output so around 80 lumen per watt and uses 5050 LED's on an alloy back one wonders how anyone can produce these for the price. Clearly it has some form of PWM controller to get that range of voltage.

Price does not seem to go hand in hand with quality I was warned it may have live parts but it is in a earthed metal outside lamp with a glass front so in real terms no one can get anywhere near the bulb when in use. OK not as bright as the quartz halogen lamp it replaced but 120W to 5W is some jump and it is enough to light the drive and it has worked longer than the 3 days I got from the quartz halogen tube.

I have 14 different 230 volt LED units in the house, in some cases 8 of the same unit, and non of my 230 volt LED units have failed and all were rather cheap the replacement for the fluorescent tube being the most expensive I think that 24 watt lamp was £17. It seems to me it is only the 12 volt LED where we have problems. So simple, don't use 12 volt versions. Or at least if you do use one of the 10 ~ 36 volt types on a DC supply. Other than the extremely cheap £1 MR16 12 volt LED units I have not had a single one fail. But I note with some 12 volt versions there is a complete lack of information. Either voltage is stated simply as 12 volt where clearly there should always be a tolerance, the lumen is missed out or the watts is missed out I would expect to have problems where details are withheld.
 
Good on you to post those LED driver chip circuits, makes a complete nonsense of the old technology using discrete parts. Yes not only do they have the sensing and feedback to regulate current and voltage but also the driver or the switching transistor that switches load through an external inductor and a schottky diode, where in the top diagram it is external and the bottom one it is probably built into.

The biggest problem used to be Power dissipation, but over the years they now have devices that have very low ON resistance (like MOSFETS) unlike Bipolar transistors, so once the on resistance is in order 0.2Ohms, and you are passing say 2 amps peak currents, you have only 0.4watt dissipation compared to many watts in bipolar switches.

Biggest advantage for 12v LED bulbs would be in them able to run off batteries, so they could do away with Emergency lighting requirement, one can run a whole house lighting off a car battery.
 
A quick tally up tells me that I have 51 gu10 leds around the house with a a total of around 255W, if they were 50w halogen we are looking at 2.5kw
The oldest are 4 years plus old.
One of my last rooms was the kitchen which you chaps have helped me with, the reason I have nowt done it until now is that they are 12v mr16s, luckily it turned out ok due to a WW transformer but if I had to change the transformer that would of have to wait because of access.
Forget all this government forcing energy companies to provide insulation, just get the nation onto LEDs -which in turn will bring the price down and the technology better as costs fall.
 
Amazing how much electricity bills you are saving.


any way a little cheer up for all:

Two atoms are walking down the street, and one says to the other, "Wait, wait, we have to go back. I've lost an electron somewhere."
The second atom says, "Really? Are you sure?"

To which the first atom replies, "Yes. I'm positive."
 
Im sure Stephen Hawking is ****ing himself at that (as well as you)
I suppose that if I did not have led downlighters then there could of been a single light in each room instead of 5 or 6 downlights. But still 4x5W downlights in a fitting against a 100W bulb
 
Well I never knew that. I said a swear word and it automatically swapped letters for some ****. Thats rather nifty
 
... one wonders how anyone can produce these for the price.
I suspect that if got a quote for buying, say, 100,000 driver chips (and 1 million LED elements) every month that you might start understanding at least part of the "how" :)

Kind Regards, John
 
I have no idea either why the bulb worked all the way down to 5.2V ....
... we've discussed the "how" of that to some extent, and chips such as eric has mentioned could obviously be the means of realising that "how" - but I'm far less sure about the "why?" - particularly if, as you say, the minimum operating voltage is stated as 12V (and they are presumably always going to be run from supplies with a nominal ~12V output).

Maybe it's just that the available (and perhaps cheapest) chip that will do everything else they want also 'happens' to be able to step up voltages from as low as 5.2V, even though that's not functionality that they actually want or need??

Kind Regards, John[/QUOTE]
 
Perhaps I should have rephrased that "I have no idea why the bulb (LED) worked all the way down to 5.2v despite the minimum voltage specified as 12v by Enlite.

But technically it is not a problem to see why it actually works at minimum of 5.2v. since by design it would be the voltage at which the booster circuit starts to kick into action and the chopper circuit energises the inductor which starts pumping up charge into a capacitor, the capacitor can be the self capacitance of LED diodes and other associated parts within or a dedicated external cap, but since it must oscillate at good few kilo hertz, it therefore requires a very small value to store charge being pumped into it and to run a few hundred mili amps as opposed to many amps where you would need a lot bigger cap and several MOSFETS mounted on heat sinks.
 

DIYnot Local

Staff member

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.


Are you a trade or supplier? You can create your listing free at DIYnot Local

 
Sponsored Links
Back
Top