Current controlled bulbs -- can someone explain?

[Tirodani]

Hey folks,

Just wondering how current controlled bulbs work. Can they be direct driven if the required voltage is 'just right' (ex. 6.6A 48w on two 3.7v cells)? Can they be overdriven?

For examples, see here. Osram has a few IRC bulbs of this type as well (but only PK30 base or reflectorized).

EDIT:

This now feels like a dumb question to me. A filament is a filament, so the 6.6A 48w bulb should be the same as a 7.2v 48w bulb. I'm just confused by the the way these are spec'd and marketed.

 

[LuxLuthor]

Hey folks,

Just wondering how current controlled bulbs work. Can they be direct driven if the required voltage is 'just right' (ex. 6.6A 48w on two 3.7v cells)? Can they be overdriven?

For examples, see here. Osram has a few IRC bulbs of this type as well (but only PK30 base or reflectorized).

EDIT:

This now feels like a dumb question to me. A filament is a filament, so the 6.6A 48w bulb should be the same as a 7.2v 48w bulb. I'm just confused by the the way these are spec'd and marketed.

I'm not exactly sure what you are asking, but most all the bulbs I have seen have a default voltage spec, and then variation in wattage ratings. So for example this Osram IRC-12V 50W has default 4.17A whereas this Osram IRC-12V 65W has default 5.42A

Both of those are rated at 4,000hrs if run at default spec, but you can find them in the destructive test chart in my sig and see how much they can be overdriven with higher voltages. I think the main point is that two different 12V 50W bulbs will both run 4.17A at that voltage. So they are mainly sold based upon Voltage & Watt specification. Using Power (watts) = Voltage x Current, if a bulb is listed as a default Amps & Voltage, you can still figure out what the wattage is...just like your home screw in lamp bulbs.

Beyond the volt & watt rating of one bulb to the next, the differences then relate to their life rating, color, shape/thickness/quality of filament, quality/purity of filament alloy, halogen gas, shape/thickness/quality of glass envelope, type of socket connection, etc.

 

[Tirodani]

If you take a look at the link above, you'll see Osram, Philips and Thorn bulbs rated between 6 - 6.6A, 30 - 200W.

In the Osram catalog they are listed under "current controlled bulbs". I think this is just because they're designed to be used with a particular type of power supply, but now I think there's no reason they couldn't be used otherwise.

Looking at specifics, though, the bulbs are all a bit large, in the 11 to 13mm range, so perhaps not useful. The IRC bulbs (PK30d base, which are just dressed up bi-pin, right?) start at 45w (6.8v) at 3000 hours -- which I thought was interesting because other IRC bulbs are 12v, right?

 

[LuxLuthor]

I have not actually seen this category of lights, so it is a good question. Looking at the transformer listings page here, this whole design revolves around a requirement of fixed 6.6 Amp system, and using transformers to vary the delivered voltage to get the specified watt.

There must be a safety reason why airfields use this fixed 6.6Amp, and then achieve the desired watt (& lumen output) bulb based upon making it tolerate specified voltage according to Watts/6.6Amps...but I don't see it applying to our battery powered flashlights.

 

[Benson]

Very interesting find. It looks like airfield lighting is commonly done with current-regulated series strings -- I can't see how the bulbs would care (well, except that constant-current eliminates the need for a soft-start), but it does give some interesting default voltage options for voltage-regulated or DD applications; maybe some of them would match nicely to convenient battery voltages...

The easiest to use in Mag mods would be the ones with G6.35 base -- that's a regular bi-pin. PK30d seems to be some sort of flange mount. Some of those dichro reflector bulbs work easily, too.

 

[LuxLuthor]

And how are you going to control delivery of only 6.6A without a transformer?

 

[Benson]

And how are you going to control delivery of only 6.6A without a transformer?

Take one rated at, say, 6.6A, 36W, then the default voltage is just 36W/6.6A=5.45V; you can either (with a regulated voltage driver) regulate to 6V or so, or direct drive from maybe 2 A123 cells.

Obviously, if I could afford it, I'd send you a couple bulbs to first, so I know how much overdrive I have to work with.

Of course, it's also possible to devise a current-controlled buck regulator, similar to the ones used with P7s but rather beastier. But like Tirodani said, a filament's a filament, so voltage regulation should be fine for 1 bulb.

 

[Tirodani]

Very interesting find. It looks like airfield lighting is commonly done with current-regulated series strings -- I can't see how the bulbs would care (well, except that constant-current eliminates the need for a soft-start), but it does give some interesting default voltage options for voltage-regulated or DD applications; maybe some of them would match nicely to convenient battery voltages...

The easiest to use in Mag mods would be the ones with G6.35 base -- that's a regular bi-pin. PK30d seems to be some sort of flange mount. Some of those dichro reflector bulbs work easily, too.

Regarding the PK30d base, I've read that you can take apart a PK22s base (as used in H1 and H3 bulbs) to get at the creamy bi-pin filling. If we could do the same with the PK30d, that could give access to the 45w IRC (64317 IRC/64319 IRC, same bulb different connectors). It's rated 3000 hours at 6.8v; if it follows the overdrive habits of its 12v brothers, it might be good in the 10-11v range (3x LiFePo4 or Li-Ion, 9x NiMH.)

The G6.35 offerings could be useful but I think they might be a bit fat and long for what they offer. Take a look and see what you think.

And how are you going to control delivery of only 6.6A without a transformer?

That was along the lines of what I was asking in my first post; on further reflection, I don't think it's a pertinent question. Voltage and current are already positively correlated in any incan bulb. I think this whole "current controlled bulb" category exists only because a certain market exists that uses them with a current controlled supply; they're just regular bulbs spec'd for and marketed towards aircraft ground lighting systems.

This is my best guess; any other opinions?

 

[LuxLuthor]

I see what you are saying if you used a fixed voltage output, but something about this is not having me . I think my uncertainty revolves around not knowing if they are more susceptable to even a slight inrush current above 6.6A, but I already have too many incan bulbs to divert into this category, but it is interesting. Of course these all had to just be over AW's Li-Ion 5.5A PCB protection limit.

 

[Tirodani]

I see what you are saying if you used a fixed voltage output, but something about this is not having me . I think my uncertainty revolves around not knowing if they are more susceptable to even a slight inrush current above 6.6A, but I already have too many incan bulbs to divert into this category, but it is interesting. Of course these all had to just be over AW's Li-Ion 5.5A PCB protection limit.

Definitely IMR territory. But just imagine 80w of overdriven IRC goodness in a 3C mag... or a Megalennium.

And yes, I know that you have a lot of bulbs to play with; hopefully someone out there will want to experiment. I don't have a Mag host yet, so I can only scheme. Actually, the Philips 6137 is an MR-11 which might be just right on 2xIMR26500; I'd need to find a way to fit those cells into my FM 2C, though!

Re: your concerns about sensitivity to higher currents: the way I look at it, if a voltage controlled bulb can tolerate higher voltages, why couldn't a current controlled bulb tolerate higher currents? This may be an overly simplistic view, of course.

 

[js]

Guys,

Regulating the current of a bulb is just a different way of driving our very own well loved filament bulb. Osram IRC airfield lamps are indeed all driven by pushing a set current through them. You don't need a transformer to do this.

Current regulating has a number of advantages:

1. Inherent soft-start. By it's very nature, you will have a soft-start driver.

2. Regulation circuit is unchanged whether you have a long string of 50 lights on the long landing strip, or a shorter string of 20 lights. The same current goes through all of them, and it's the same lamp, so you have the same driver circuit (but you'd obviously need a lot more voltage for 50 lamps).

and finally (and most importantly)

3. inherent CONSTANT OR INCREASING brightness over the life of the lamp. When a filament is about to die, it develops a thin spot of higher resistance. If you are regulating on voltage, this means that you put less current through the same filament than you did when it was new, which equates to lower power, which equates to less light. But if you drive the same current through the filament, you have the same current * a higher voltage = MORE power = more light.

Of course, if all your lights are in series, and one blows, they all go out! So, I'm quite sure that air fields must use a series / parallel arrangement of some sort or another.

As applies to flashlights, a lamp that is meant to be current controlled is physically no different than a normal one. But it does mean you'll have to experiment a little to find the right voltage drive level and be aware of insta-flash issues.

 

[Kremer]

Guys,
Of course, if all your lights are in series, and one blows, they all go out! So, I'm quite sure that air fields must use a series / parallel arrangement of some sort or another.

or maybe an automatic shorting means in the bulb for when the filament pops.

 

[js]

Mmmm. Good thought! But it wouldn't be in the bulb. Maybe in the bulb holder, though.