Compact fluorescent - voltage

They are current dependent.

They run off standard 50/60Hz A.C. line voltage/current e.g. 120V or

What did you *really* want to know ?

Graham

You can run them off a small 220v/12v mains transformer in reverse, wired in a flyback configuration. That means, a transistor, and letting the thing self oscillate through one of the windings. The frequency all depends on how the transformer is built and will vary from manufacturer to manufacturer. But you can run a small tube off of a 9v radio battery.

The terminology is roughly right there and someone will offer a better explanation. All I know is that from tinkering that it works. These handheld flashlights will have a small specially built transfromer meeting the specifications of the the tube the manufacturer chose.

:-)

There's a bit on this on the DIY hobby sites somewhere. Google it.

I want to know what is applied to the *tube*, because I want to possible use the ballast for other purposes. Like 48V dc -> 12V dc, after all they seem to have all the stuff for a dc/dc converter.

An LCD backlight is essentially the same thing AIUI. You'll find plenty of info on drivers for them. IR have some dedicated CFL drivers too.

Graham

CCFL backlights typically run at 400 - 700Vac (depends on the backlight) and the start voltage is usually about twice the run voltage give or take a handful.

Normal run currents vary - a very common run current is 5mA.

I have one I am using (although I won't be soon - going to white LED) that has a nominal run spec of 680Vac with a start voltage guaranteed at 1500Vac (typical 1250 Vac)

Cheers

PeteS

Have you tried googling about it or using Wiki? My recollection is that the compact fluorescents operate in the 20kHz area and operate at a low power factor, but I did what you should have done, I think. I went to wiki and google and quickly identified this page:

That probably provides as much detail as you may need.

But please try spending a little of _your_ time searching _and_ informing us first about the scope and limitations of what you found, next time. You may have done some searches already, but you said nothing about that.

Jon

His ridiculously long username, his sloth, and posting to s.e.d all may make him look like a n00b--but he ain't.

...and you're right: He should know better.

I ain't no ex spurt but I took one of them suckers apart.

The one I played with. Had no isolation! first off - you'd need a lot of experimenting to make a DC converter out of it. What they did was rectify the line in and then chop it (on and off real quick with discrete semiconductors) then routed the chopped DC through a capacitor to the tube. and had another circuit to work the heaters from the line (yes, it wasn't cold cathode)..

That was my first (or second) thought when I had one burn out. How do I make a 12 PS out of this? Conclusion: ain't too likely.

Now those little 5.7 volt isolated 800 madc converters they make for Nokia phones? The local store sells them for $5 and when I get the time . . . .

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

I have been searching on google/wiki a lot but I didn't find anything useful. I figured if I know the voltage on the tube I can approximate the transformer ratio in the ballast. I need the transformer for an PoE project. But I also figured CFL ballasts can be used for cheap dc/dc (48->12V) converters. Another source for cheap dc/dc can be those mobile phone chargers. But I suspect they have too high ratio on input/output (230->4V) voltage. And they are harder to get by in large amounts.

LCD drivers have some nice big transformers on their pcb. However they are harder than CFL ballasts to come by, and are more cumbersome to handle. As there usually is a whole laptop included ;)

to

info

Graham

Well, I had no way of knowing any of that. It would help a lot to know what you didn't find useful and where you had looked, already. It's better to write more, than less -- up to a point. You respect the time of others, that way.

You haven't mentioned whether or not the article I pointed out was any use. It's quite possible it isn't.

I don't imagine a "voltage," here. There is a necessary cold starting voltage and perhaps some other steps early on, but once ionized, isn't it the case that current control is the important issue?

I remember from doing a flash lamp many years ago (30) that the voltage supply is something like 300V or thereabouts. My guess is that compact fluorescents with tube lengths somewhere in the neighborhood of a similar length will need roughly similar voltages, with starting voltages much higher (I had to use about 4kV to trigger the lamp, memory serving, supplied by a trigger transformer driven off of the 300V on its primary winding.) What's actually required to start these lamps, I don't know.

If my memory serves on another score, to avoid depositing needed gas atoms on the tube surfaces, it's important to use AC and, if possible, fairly high frequency AC is better than low frequency (up to a point, which seems to be about in the area of the 10-30kHz.) The better frequency to keep the atoms from depositing is a matter for physicists to discuss, but I think that frequency is a typical area.

So these do NOT seem appropriate for dc to dc converters. The parts may be useful, such as the current limiting inductor, for example. But the design topology seems all wrong to use in its entirety.

I don't know what a "PoE project" is, but I do know what a 48V to 12V dc/dc converter is, generally anyway. Since you are into adaptations of one kind or another, why not just design and build one and scavage parts from where you can?

Jon

scavenge. Sorry about the typo.

Jon

I tried for some time to find the answer so to say. But I figured I had reached something not well documented. And it was proberlbly a better idea to just ask then to search in absurdum. I tried to limit the requested data instead. A quick way to find out is with a scope across the cfl tube. But high voltages >1kV makes it risky.

It's proberbly useful. I haven't yet read it in depth yet thoe.

Yep, in many places it says have current limiting or else *P0FF*

So far my guess is fire voltage ~800-1500V, and run voltage ~300-600V.

The useful threshold seems to be 20kHz. But due audioable noise higher frequency is proberbly prefered.

My thought was to reverse the transformer and put the rectifier + capacitor on the output. Guess that idea is toast thoe ;) There are designs for dc/dc but they still require an inductor.. thus looking into cfl ballasts.

The PoE project is about power over ethernet. You send some sense signals and then apply 25-60V dc max 400mA over cat.5. But one need a 3:1 isolated transformer. I just want to make a quick bench test of the schematic before ordering the real stuff.

While looking into this I figured a 48V dc distribution network would be useful for power distribution to computer equipment in the building useing thicker cables than cat.5. And cheap sources for 48V -> 12 or 5V dc conversion would be nice. So I thought maybe the cfl ballast could be modded into this. However it seems this was not the case ;)

Nope. There was a model that used an incadescent lamp as a ballast for a 4 foor fluorecent lamp. Ran off 230 VAC - no igniter IIRC.

If they ran off about 50 or 60 VAC I wouldn't be surprised.

Hmm. I just went over to that PDF file I'd mentioned and they seem to talk about 300V there. So perhaps I'm not far off base.

Well, I'm almost dead certain that compact fluorescent (at least, those in the US) run at high frequency. The long ceiling tubes _do_ run off of 50-60Hz, as you say. But not the compacts, as I understand it.

Jon

Sure. Even the 4 footers can come with electronic ballasts now - less flicker and buzz. But there's no sign of an igniter system so they strike off 110 VAC.

Wouldn't bet my life on it but it seems so. Still, it's hard to figure what is happening at that frequency.