Fluorescent light, battery powered

[fdk]

Years ago we had a fluro light running off a half dozen D Cell batteries.  I'm interested in building a circuit to run a 4W tube, and hook it up to few 18650s, maybe 3 for 12ish volts.

There's not really much decent info as everything these days is geared towards LED.  I know that is probably the better way to go, but I want fluro like my childhood memories, dammit!

There are some simple joule thief based circuits online which seem a too simple given that most discussion on how fluro circuits work centres on there being 2 stages, starting and running.

There's no real point to this besides academic pursuits and just doing it, so if anyone has some good ideas/desigs on how to do this I'd be immensely grateful.

P.S. I learn best by doing, screwing up, then doing again until it works.

[Terry Bites]

Many choices. Even class E these days. Just about any inverter scheme can be used to gererate the HT and heater power.
Back in the day.... I think Royer circuits were the most popular. Very simple.
I found this: www.google.co.uk/url?sa=t&source=web&rct=j&opi=89978449&url=https://www.diodes.com/assets/App-Note-Files/an17.pdf&ved=2ahUKEwjkmJfOlsiLAxU6V0EAHVP3PNIQFnoECBkQAQ&usg=AOvVaw3lbOf_DBa5NGCH0ubocgqT

And this:


And a fanciful claim for a 100% efficient circuit. Yes 100%! Who peer reviewed this and didn't noice that BS, why not 146%?

 here www.google.co.uk/url?sa=t&source=web&rct=j&opi=89978449&url=https://eprints.qut.edu.au/62445/3/PaperID227_singer.pdf&ved=2ahUKEwig2evnlsiLAxVjU0EAHfqpJ-0QFnoECBMQAQ&usg=AOvVaw2CeT0f_rHhCWP_djyJ7BIt

[Ian.M]

There is some info on old commercial low voltage fluorescent driver circuits at Silicon Sam's Technology Resource (Samuel M. Goldwasser of Sci.Electronics.Repair FAQ fame), including transformer core and winding data.
https://www.repairfaq.org/sam/samschem.htm#schinv

[SteveThackery]

I recall seeing a circuit in a magazine which didn't use the heaters. Presumably a high enough voltage will strike the tube anyway?  I ask because running the heaters might waste energy when it's a battery-powered circuit.

[CaptDon]

I have a very nice battery powered 4W hand held unit. I have CW, BL, BLB and a clear germicidal bulb for it. Makes a good night light for camping and a good black light for night time mineral and 'critter' investigation. The 4W unit does generate a tiny bit of heater power in two individual windings and has a separate HV A.C. winding. The unit is hard on batteries but gentle on bulbs. I had an 8W 12vdc unit for use in a boat / camper and it would ruin a bulb in about 20 hours of operation. It suffered from metal displacement I assume from mercury ions on the unheated cold cathodes. I threw it out and went with warm white LED strips.

[Gyro]

Starting the tubes without the heaters tends to blacken the ends. Mains high frequency compact florescent lamps typically use a tuned capacitor between the heaters to achieve some pre-heating before the tube strikes.


EDIT: as CaptDon says, secondary heater windings are the ultimate solution for tube life.

[Andy Chee]

My brute force solution would be to tear apart a broken spiral CFL or similar, and rebuild the driver section on a new board. Then build a small 12V-200V step up DC-DC converter to power it.

Running a conventional fluoro in cold cathode mode without filament heating, will definitely blacken the ends, though that may not matter in your experimentation.

[fdk]

Thanks for the replies.  Yes, it is the blackened ends I'd like to avoid.  I take it it is rather easy to light up a fluro, harder to make it last a long time.  I will investigate some of the links and see what I can put together.

[Whales]

I built the last circuit on this page a few years back:  https://ludens.cl/Electron/Fluolamp/fluolamp.html

It still works well.  Full-length fluoro tube, 6 flat AA batteries, enough light to see at night but not particularly bright.  I use to have to strike the tube manually by putting one hand to one terminal and sliding the other hand down the length of the glass it was fun but not safe if you slipped and touched the other terminal.

[ME]

I built one tears ago, I didn't bother with heaters,it seemed to work ok with a high enough voltage.

[armandine2]

...an early Big Clive take on it

[lu1s.p]

That sounds like a fun and nostalgic project! You might want to look into inverter circuits or small CCFL drivers, as they work similarly to fluorescent starters. A Joule thief could help, but a proper ballast circuit would be more reliable.

[Rick Law]

That sounds like a fun and nostalgic project! You might want to look into inverter circuits or small CCFL drivers, as they work similarly to fluorescent starters. A Joule thief could help, but a proper ballast circuit would be more reliable.

++1 to that!

Eons ago, I took the CCFL backlight from a dead laptiop to play with.  Everything I needed to get a 18650 driven CCFL "night light" were inside the laptop lid.

If you search for CCFL inverter circuits, you will find a lot of helpful info and parts should you need them to get the CCFL light going.

[lu1s.p]

Reusing components from old laptops is always a great idea.

CCFL inverters can be quite useful for these kinds of experiments.

Did you have to modify the original circuit much, or did you just adapt the power supply with the 18650?"

Thank you

[Rick Law]

It was back in year 2000, or perhaps just before.  I don't remember much from that far back.  I would have been using Sony or HP laptops in those days.

I don't recall having to make any circuit alterations -- I was not equipped to do much then, I had just a soldering iron and a Radio Shack 2000 count DMM back then.  I replaced the inverter once during the laptop's life-span.  So when the laptop went kaput, I knew what to do to get inside the laptop-lid and stuff inside the lid.  The cute tiny CCFL bulbs was what interested me.  I just had to remove the two CCFL bulbs (long skinny tube tubes one on each side of the screen).  The inverter circuitry was inside the laptop lid at the bottom (around the hinge of the lid) wired with a plug to the laptop's mother board to an approx 5V power source -- it was just around 4.6 volt when the laptop no longer runs.  Not sure what it should be if the laptop was still alive..  I recall deciding 3x AA should drive it ok and it did.  Outside the protection of the laptop-lid, the CCFL tubes didn't fair well.  One was quickly broken but the inverter worked fine with just one tube.  Years later when I starting playing with 18650's, I switched it over to a single 18650 recovered from a later model Sony laptop power pack and it worked.  That was the last time I touch those parts.  With the 18650's I switched over to playing with LED's.

Tonight, I just pulled out that project-box again from the basement, all I have left in that box now is the BMS boards from the old Sony laptop battery packs.  So I must have discarded the CCFL stuff when the second tube also broke as well.

The CCFL tubes were cute, but the high intensity LED emitters were a lot more fun to play with -- particularly with the MCU equipped driver boards.  I am still using my 5 originally cheap but guts upgraded Cree T6/XM2 flashlights, driver upgrade to NANJG 105c (ATTINY 6 on the NANJG also upgraded to ATTINY 85, yeah, you have to bend the pins to fit the ATTINY85 to match the ATTINY 6 pads on the board) running on home-made programs.  Mine have silly features I added such as blinking out the voltage - 3 long and 8 short blinks would mean the 18650 is at 3.8 volt.  Lot more to play with that setup than a simple CCFL tube...

[richard.cs]

Take a look at some of Manfred's designs: https://ludens.cl/Electron/Fluolamp/fluolamp.html

3 different options, described in a decent level of detail.

[fdk]

Thanks all very much!  I've gotten a bit busy with other things, so I will get to this when I can.  Thanks heaps!!

[Andy Chee]

FWIW, Bigclive dropped a new video on fluorescent tubes

[coppercone2]

I have a easier solution, get rechargeable D-cells from tenergy. The ones in the normal stores are pretty lame, they make them cheap and they have a way lower capacity, like only 4000mAh, its some horseshit about leveraging consumer expectations (that people never need the high capacity D cells in the modern world )

IMO those tenergy D cells are so nice its bearly worth the effort to convert it to lithium (is it more convenient?). Their 10000mAh

It should run for like 15 hours


a 18650 is about 3 amp hours @ lets say 3.8V
A nimh D cell is about 10 amp hours @ 1.2v

so if you run 6x nimh you get ~ 10amp hours @ 7.2v (72)
and if you run 3x 18650 you get approx 3amp hours at 11.4v (34.2)


I assume there is power converter working both sets. Now its probobly going to be worse then 72 but its still might beat down 34.2 IMO

Modding a existing device might be counter productive

[RJSV]

My RADIO SHACK portable lamp set had 2 bulbs, (maybe 5 1/2 inches each).
It ran on 4 regular 'C' size 1.5 volt batts.

   I don't have schedule open to help search,  but might be an older vintage outlet that mentions it.   I know that the Lamp had a separate driver for each lamp,  and featured a folding style that helps when aiming the lamp at something.
The C type cells seem a bit ignored,  the last 28 years.