60 lm/W halogen bulb. TESTED!!!!

finally i have do a overdrive test on the new osram irc series...

first of all i have test them with 3 li-ion cells (D size so low voltage drop) and consider the output 100%

then with 4 cells: output 249%

with 5 cells: 460%

this mean that the ususal table for calculate overdrive remain correct, so with 5 cells the efficiency reach 60lm/w!!!!! more than some hid lamps... incredible!!!!

the lamp life is over 4000 hours with 3 cells, 400 hours with 4 and 30h with 5 cells. the cost of the lamp is around 5$ in germany (plus shipping fees etc...)... quite cheap

here a image of the lamps http://digilander.libero.it/cuxdavide/osram irc.jpg

you can see that the filament is axial, only the glass is slightly frosted and can affect the beam, maybe is a problem of my lamps and others are more clear, i hope so...

osram produce 4 wattage of this lamps: 25, 35, 50, 65W

the 25W is the less efficient overall, the 35 and 65W the more efficient.

the 35W with 5 cells drain around 68W and output 4000 lumens

the 65W with 5 cells drain around 125W and output 7500-8000 lumens

These lamps demand big overdrives to provide a white light. My 65 watt on 12 A cells is still much warmer and yellower than a 62138 on 11 cells.

They do, however, put out a LOT of light!!

That haze or fog is the coating used to absorb/reflect the infrared back into the envelope. It is effective, too!! Because of that coating, the filament stays hot longer after you shut off the lamp, and so glows for quite a while.

I have yet to push these with more than 12 cells, but I suspect they will handle 15-17 NiMH cells without to much problem.

I haven't had much time lately to play, trying to get the next shipment of USL's ready!!

Bill

Not possible.

The highest LPW attainable with a tungsten halogen incandescent is about 42.

You can't trust the re-rating formulas outside of +/- 10 or 20 percent.

Trust me. You haven't got a 60 LPW lamp.

davidefromitaly said:

first of all i have test them with 3 li-ion cells (D size so low voltage drop)

Click to expand...

Have you measured the voltage drop? Being a D size cell doesen't automatically give it a low internal resistance. There can also be a lot of losses in the cables and connectors that go to the lamp. It would be useful if you measured the voltage directly at the two pins of the bulb each time.

davidefromitaly said:

then with 4 cells: output 249%

with 5 cells: 460%

Click to expand...

How did you measure that? Do you have a light meter?

davidefromitaly said:

this mean that the ususal table for calculate overdrive remain correct, so with 5 cells the efficiency reach 60lm/w!

Click to expand...

I don't think that the formulas are correct with IRC bulbs. The efficiancy gain comes mostly from the fact, that at higher voltages, less of the whole output is infrared. But IRC works best if there is a lot of IR radiation, so the gain that comes from the IRC coating will go down when the voltage goes up.

davidefromitaly said:

you can see that the filament is axial, only the glass is slightly frosted and can affect the beam, maybe is a problem of my lamps and others are more clear, i hope so...

Click to expand...

Like bwaites already said: The "frosting" is the IRC coating, all IRC bulbs have that.

davidefromitaly said:

the 25W is the less efficient overall, the 35 and 65W the more efficient.

Click to expand...

How do you come to that conclusion?

davidefromitaly said:

the cost of the lamp is around 5$ in germany (plus shipping fees etc...)... quite cheap

Click to expand...

I bought a few and tested them. I could overdrive them to 21 Volts (measured directly at the twp pins of the bulb) without blowing it. At 18V it drew exactly 70W from my regulated power source.

js said:

The highest LPW attainable with a tungsten halogen incandescent is about 42.

Click to expand...

That's only right for a normal incandescent. If you use an IRC coating to reflect the IR radiation back onto the filament, the efficiency can theoretically be much higher.

I didn't want to burst anyones bubble, but js is correct. IRC lamps may increase the efficiency by increasing the filament temp, but Tungsten starts melting long before you reach a 60 lumen/watt figure.

That is an impressive overdrive!! How white could you make it?

Bill

bwaites said:

I didn't want to burst anyones bubble, but js is correct.

Click to expand...

No.

bwaites said:

IRC lamps may increase the efficiency by increasing the filament temp, but Tungsten starts melting long before you reach a 60 lumen/watt figure.

Click to expand...

IRC doesen't increase the efficiency by increasing the filament temp, but by reducing reducing the current that is needed to heat the filament. In a conventional bulb, all the energy that is used to heat the filament comes from the current. In an IRC bulb, only a part of it comes from the current, the other part comes from the reflected IR. This means that you can maintain the same output while decreasing the input.

YES! Show me a single lamp that approaches 50 Lumens/watt.

But, it is essentially the same thing, isn't it?

Less energy to heat the filament to it's most efficient level from the power source because you are using the IR already produced from the filament to increase the temperature of the filament!

But again, the tunsten vaporizes before you get close to 60 lumens/watt.

Bill

bwaites said:

YES! Show me a single lamp that approaches 50 Lumens/watt.

Click to expand...

Because I don't have one here, doesen't mean it isn't possible.

bwaites said:

But, it is essentially the same thing, isn't it?

Click to expand...

Yes, we are talking about the same thing, only that you think IRC is used to rise the filament temperature compared to a conventional bulb, whereas my focus is on the fact that the filament temperature is the same in both bulbs and the IRC is only used to lower the current.

bwaites said:

But again, the tunsten vaporizes before you get close to 60 lumens/watt.

Click to expand...

Tungston vaporizes at a certain temperature, not at a certain efficiency.

Here's an example as easy as possible: Let's say that we have two bulbs. A conventional one and one with an IRC coating. Both are burning very close to the melting point of tungston. This means that the conventional bulb has an efficiency of 42 lm/w and emits a certain amount of IR radiation. The IRC bulb reflects some of the IR radiation back onto it's filament, where the energy is "reused" to heat the filament, instead of heating your room. If you would drive both bulbs at the same voltage, the IRC bulb would blow, because the filament would get too hot. To avoid this, the voltage of the IRC bulb is lowered (so that the tungston is near the melting point), resulting in a lower current and a lower power.

Now what do we have? We have two bulbs at the same filament temperature, thus emitting the same amount of light. The conventional one is burning at 42 lm/w and the other one draws less power, but emits the same amount of light. This can only mean one thing! The IRC bulb has an efficiency of MORE than 42 lm/w.

Do you agree? If not, please state exactly where the error in my argumentation is.

I agree, but the WHOLE idea of IRC lamps is increasing efficiency, not increasing lumen output, or allowing the same output from less electrical energy.

When I state that tungsten vaporizes before you can get to that efficiency, I am simply stating that in order for the filament to produce that much light, the filament would have to get hotter than it is possible for it to get, not that it is a relationship of efficiency to vaporization point. I assume (not always a good idea, I know) that anyone reading this type of thread would understand the implied.

In this particular case, the whole argument is whether that IRC technology allows the lamp to reach an efficiency of 60 Lumens/watt. My point is that tungsten simply won't allow that, regardless of what technology you use.

The IRC technology simply allows you to do it with less input energy from your power source, it doesn't allow you to get somewhere where you can't go because of the limitations of the tungsten.

If you find an INCANDESCENT lamp that approaches 60 Lumens/watt, we want it!! But so far we can't find one that is close to 50, and in fact most are closer to 25-30, including the overdriven WA and Osram lamps. The overdriven 62138 in the USL, for instance, is around 28-32, and it is incredibly white. Some of the WA lamps approach 35, if I recall correctly, but none 60 or even 50.

Bill

bwaites said:

The WHOLE idea of IRC lamps is increasing efficiency

Click to expand...

That's exactly right.

bwaites said:

not increasing lumen output

Click to expand...

Why not?

bwaites said:

or allowing the same output from less electrical energy.

Click to expand...

Allowing the same output from less electrical energy is exactly what we call "increasing efficiency". I don't understand how you can say that this isn't the purpose of IRC, while at the same time you say that the purpose is "increasing efficiency"!

bwaites said:

In this particular case, the whole argument is whether that IRC technology allows the lamp to reach an efficiency of 60 Lumens/watt.

Click to expand...

I NEVER said that you could reach 60 lm/w with an IRC bulb. I simply said that js was wrong when he said that it's impossible to get more than 42 lm/w out of an incan bulb because his statement is only true for conventional bulb without IRC.

bwaites said:

My point is that tungsten simply won't allow that, regardless of what technology you use.

Click to expand...

Your point is wrong. Let's say we have a "perfect" IRC shield, that reflects all IR and UV radiation back onto the filament. Now all the energy would go into visible light, with no losses from IR or UV radiation, that our eye can't see. Because we are now converting the whole energy into visible light, we would get around 200 lm/w, which is the theorethical maximum for white light.

I know that this technology doesen't exist yet, but this doesen't mean that it's impossible to come close to that.

How is my point wrong? You have a 60 Lumen/watt light?

Now I understand! You are arguing to simply argue!!

NO tungsten filament will EVER do what you THEORETICALLY are proposing, you're into Star Trek technology there!!! A THEORECTICALLY perfect IR/UV reflector which allows visible light to pass goes right in there with teleportation, force fields, and Warp Drives!!

You keep arguing the same point, IRC tech was developed to increase efficiency as you say, and there is NO WAY that it can create 60 Lumen/watt! Unless of course, you are ready to release YOUR new IR/UV coating.

js point was simple, "there ain't no way"!

I agree, there ain't no way, not with today's tech. Of course, when you come up with that coating and some 60 Lumen/watt lamps useful in a flashlight, I'll be first in line to buy some!!!

Until then, there ain't no way!!!

Bill

Tungsten technology only sets a limit on the maximum temperature of the filament, which translates into maximum color temperature of the light source and maximum "power density" of the light source (power radiated per unit surface, following the kwown T^4 law of heat radiation) - a "point-source quality measurement".
IRC achieves a reduction of the electric power needed to reach the same filament temperature of non-IRC bulbs, hence getting the same light (if the filament is the same) using less power: more lumens-per-watt or a higher efficiency.
Am I missing something? Please correct me if I'm wrong.
I don't think there is a dramatic difference in efficiency between standard halogen and IRC bulbs, but IRC does permit higher efficiencies than standard halogen technology.

Somebody should experiment with overdriven IRC bulbs and throw in some real world numbers!

Sanny

Sanny said:

IRC achieves a reduction of the electric power needed to reach the same filament temperature of non-IRC bulbs, hence getting the same light (if the filament is the same) using less power: more lumens-per-watt or a higher efficiency.

Am I missing something? Please correct me if I'm wrong.

Click to expand...

What you're missing is that when you overdrive the filament a smaller percentage of its emission falls within the infrared spectrum which is reflected by the IR coating. In other words, the IR coating offers a smaller percentage of efficiency increase over a standard bulb as the filament temperature increases. At "normal" filament temperatures I've heard these bulbs with IR coatings can offer efficiencies in the low 30s compared to mid 20s for regular halogen bulbs. However, their efficiency doesn't scale the same as regular halogen bulbs for the reasons I said. By the time you reach the filament temperatures where regular halogens are in the 40+ lm/W realm, the IR coating is offering little if any additional benefit. My guess is you'll be lucky to break 45 lm/W with any incandescent bulb, IR coated or otherwise.

Now if we use unobtainium filaments which can run at 6500K you can obtain efficiencies of about 93 lm/W.

I'm totally out of my league here but :-


Pic extracted from my thread :
https://www.candlepowerforums.com/threads/110551 (for more info on the test)

156w --> 10,585 lumen
= 67.85 lm/w

Haha... nah, just kidding!! :lolsign: I know that's not an integrating sphere!

But then again, that isn't bulb lumens and there's light leaking out from the back.... so could it be possible...... Someone chuck this type of bulb into a real IS....

Pls don't flame my noob claims! I don't want to anger the incan gods!

And thus we see why you cannot calculate Lumens from Lux readings!

But there is a table somewhere here in CPF where the maximums with todays technologies were all posted: Incans, LED's, Metal Vapor, fluorescent, etc.

Bill

Yeah, I knew 10k lumens was too good to be true... but well, a man can dream can't he

these bulbs do look really interesting.. for outdoor use, you do not need the 'white'.. the 625 lamp though outputting more than the 138 does not have the detail and definition, and i get a lot more white flashback.

A study of lumen per watt on these would be interesting, also with the 623 lamp for that matter.. however i will say this about the re-rating formulas.. they were FAR off for my osram lamps until i actually made a baseline MEASUREMENT of wattage rather than making the incorrect assumption that at 12V they were 100W.

the osram 62138 which i run at 13.2V re-rates to 32.1 L/W, a reasonable estimate.. however if you use 100W and 12V like the label says it re-rates to 33.7L/W.. not tragically different.. now let's compare to the 64625... if using 100W and 12V like the box says you will get a re-rated value of 40.6 Lum/Watt.. at 12.76V.. a far cry from the far more accurate 36.6 L/W.

plugging the numbers into the re-rating formula properly.. if you measured 18V and 70W exactly.. that means 3.89A.. it also means to achieve that with the re-rating to fit in, the "35W" is actually 37.3W with 12.0V and 3.11A.

It still re-rates to 3720 bulb lumen at 53.1 L/W.. which for incandescent is not possible as far as i know, so more than likely a wall is hit and the efficiency is not up to snuff.. basically .. with the color temp being only 3000 and 900L baseline.. i would bet it's not as high up on the efficiency curve.

What needs to be done is take one of these lamps overdriven nice and hard and compare to a baseline we can measure, even with as simple and crude as a ceiling bounce with a light meter to get a realistic estimate of lum/watt at the over drive levels. osram claims 900L base, but a measurement at 12.00V of what current it's pulling is an absolute requirement at first.. the '625 lamp showed over 40L/W before finding out that the "100W" really mean "111W".

The 4000 hr life rating will mean it's capable of some extreeeem over drive.. do your overdrive calculations with POWER not voltage and the will represent reality.. voltage doesn't create light, power does.. hence lumen per WATT not lumen per VOLT.

The bulb looks really interesting, i'd be up for a couple for testing.

-awr

here's the chart bill referred to:

Can someone explaine to me why the melting point of tungsten is even relavant? While I have a hard time believe and Incan can hit more then 45 l/w, I would like to understand how the melting point of tungsten matters. If your saying the wire would just melt, and drip off, fine. Then 45 l/w is not possible with a tungsten wire, there have to better alloys to emite light somewhere...