Author Topic: Maximum Efficiency of MH  (Read 6099 times)
Lanternbro
Member
**
Offline

Gender: Male
View Posts
View Gallery


Maximum Efficiency of MH « on: May 02, 2015, 11:54:49 AM » Author: Lanternbro
I have been wondering about the maximum efficiency of MH bulbs.

Patents I've seen state 177 lm/w. I think the figure could go higher to rival LED's.

What effects the efficiency of MH anyways?
Logged
BlueHalide
Member
*****
Offline

View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #1 on: May 03, 2015, 10:22:48 PM » Author: BlueHalide
Initially (first 100 hours) metal halide lamps are super efficient, like 140 lumens per watt for quartz Na-Sc lamps. The problem is the gradual lumen depreciation caused by loss of available sodium, arc tube blackening and electrode wear. CDM (ceramic) lamps are a significant improvement over quartz and is the current rival of LED in new installations. Also there has yet to be a white LED that achieves 177 L/w. Lots of LED manufacturers will use the fact that LEDs are a directional source and count "reflector losses" of HID/fluorescent against their lumen/per watt. However, in raw output MH's mean lumen output is currently about the same as the most efficient LED's initial output.

But the thing I like the best about metal halide, is reliability in just about any climate/application. HID lamps and ballasts (magnetic) are not nearly as temperature sensitive as LEDs and their electronic drivers. Obviously HID lamps operate at extreme heat, as much as 25x higher than what most high power LEDs can handle. A dayburning LED fixture due to a faulty photocell in the middle of summer on a 95 degree day is enough to cause every LED in that fixture to fail...ive seen it happen numerous times.  
« Last Edit: May 03, 2015, 10:25:01 PM by BlueHalide » Logged
Medved
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #2 on: May 05, 2015, 01:10:32 AM » Author: Medved
Well, if you want a lamp to just win the contest in just a "most efficient during the measurement" discipline, in deed you may end up in the 200lm/W range, even for CRI above 50 (I would take that as a usable limit for an usable light to be really called "white"), but those lamps won't be usable for lighting at all.
With discharges it would mean exploding within hours, loosing most of the output over that period as well,...
With LED's very low power densities and enormous cooling (for the power).

If you want more practical HID, you can not use that extreme loading and so reactive material combinations and that cost you the efficacy.
Similar way, with the LED's you want reasonable light output, so you end up operating them at higher power densities, what cost the efficacy.

And usually you want better color rendering than the 50, what cost you quite some energy to be emitted in deeper red, but that radiated power uses needs to be fed, but it practically does not contribute to the lumen output, so again the efficacy goes down.

The 100lm/W could be seen as the compromise accepted by the market, so better technology goes either for longer life (HID's), lower cost (LED's) or better color (both), but practically stays at the 100lm/W mark.
Logged

No more selfballasted c***

Lanternbro
Member
**
Offline

Gender: Male
View Posts
View Gallery


Re: Maximum Efficiency of MH « Reply #3 on: May 06, 2015, 05:28:15 PM » Author: Lanternbro
Well, if you want a lamp to just win the contest in just a "most efficient during the measurement" discipline, in deed you may end up in the 200lm/W range, even for CRI above 50 (I would take that as a usable limit for an usable light to be really called "white"), but those lamps won't be usable for lighting at all.
With discharges it would mean exploding within hours, loosing most of the output over that period as well,...
With LED's very low power densities and enormous cooling (for the power).

If you want more practical HID, you can not use that extreme loading and so reactive material combinations and that cost you the efficacy.
Similar way, with the LED's you want reasonable light output, so you end up operating them at higher power densities, what cost the efficacy.

And usually you want better color rendering than the 50, what cost you quite some energy to be emitted in deeper red, but that radiated power uses needs to be fed, but it practically does not contribute to the lumen output, so again the efficacy goes down.

The 100lm/W could be seen as the compromise accepted by the market, so better technology goes either for longer life (HID's), lower cost (LED's) or better color (both), but practically stays at the 100lm/W mark.


That's wrong though, the Cree XP-L LED's achieve 200 lm/w.
« Last Edit: May 07, 2015, 09:13:06 PM by Lanternbro » Logged
Medved
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #4 on: May 07, 2015, 01:05:15 AM » Author: Medved
The CREE datasheet states 1012lm/3A and 85degC (that means about 10W and the expected temperature) of the bare LED for the 4000KI CRI70 type.
If you add the ballast losses, you are at about 80..90lm/W for CRI70...
In the preview on the web states "1260lm/10W", but without any relation of color type and operating temperature, so that would be some of the low CRI high efficacy type at laboratory 25degC.

When I take the output vs current graphs, at 0.5A (just 1.5W, so more expensive fixture) you may get something about 140..150lm/W for the bare LED, so about 130lm/W with the ballast losses...

All that was based on the official CREE datasheet, so something CREE really offers as a contract, so the only thing that could be taken seriously...


I was not able to see the link, but based on above, my guess: The 200lm/W could be a best figure achieved in the lab for some low CRI type, selecting one from the large lot, cooled down and operated at ridiculously low current, so nothing to do with real life use. Just a teaser...
Logged

No more selfballasted c***

merc
Member
****
Offline

Gender: Male
View Posts
View Gallery

Adam


GoL
Re: Maximum Efficiency of MH « Reply #5 on: May 07, 2015, 01:51:31 AM » Author: merc
I think this is the link Lanternbro had in his mind.

Quote: "The XLamp® XP-L LED is the first commercially available single-die LED to deliver breakthrough efficacy of up to 200 lm/w at 350 mA."
Logged
Medved
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #6 on: May 07, 2015, 02:27:34 AM » Author: Medved
I think this is the link Lanternbro had in his mind.

Quote: "The XLamp® XP-L LED is the first commercially available single-die LED to deliver breakthrough efficacy of up to 200 lm/w at 350 mA."

In that press release they just "forget" to mention "... at 25degC or colder". That statement is just about a lab experiment (with commercial product, but still just a lab excersise). These conditions are by far not realistic for real life. Plus using 5x more LED's and excessive heatsink is a bit economic nonsense, with present costs (it may make some sense just for emergency light applications or so, where the battery cost is the dominant part, but not for general lighting).
Logged

No more selfballasted c***

dor123
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Other loves are computers, office equipment, A/Cs


WWW
Re: Maximum Efficiency of MH « Reply #7 on: May 07, 2015, 11:00:57 AM » Author: dor123
This is why I still stand behind my opinion that LED products can't reach more than 100lm/w outside laboratory conditions.
Logged

I"m don't speak English well, and rely on online translating to write in this site.
Please forgive me if my choice of my words looks like offensive, while that isn't my intention.

I only working with the international date format (dd.mm.yyyy).

I lives in Israel, which is a 220-240V, 50hz country.

Lanternbro
Member
**
Offline

Gender: Male
View Posts
View Gallery


Re: Maximum Efficiency of MH « Reply #8 on: May 07, 2015, 09:10:17 PM » Author: Lanternbro
So CREE is pulling a marketing trick I see. Seems that Medved is correct.

Do CMH bulbs have IR reflective coatings like LPS?
Logged
BlueHalide
Member
*****
Offline

View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #9 on: May 07, 2015, 09:54:52 PM » Author: BlueHalide
No IR coating on any metal halide lamps (CDM or Quartz), infrared reflection is only required in quartz lamps around the electrodes to raise the cold spot temp (hence the white zirconium coated pinch-ends). CDM lamps operate at a higher pressure and therefore higher temp than quartz lamps as the alumina tube allows for the higher loading, no need for any reflective coatings.
Logged
Medved
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #10 on: May 08, 2015, 01:26:13 AM » Author: Medved
@BlueHalide: The CDM does not operate at higher pressures, but just at higher temperatures. The pressures in quartz is in 10's of atmospheres, ceramic just about an atmosphere or so. That is one of the main reasons, why ceramic do not tend to explode so violently as the quartz do.
And it is not only it does not need the IR coating, but actually it is not technically feasible to apply any on the ceramic arctube, so they should live without. Plus there is yet another aspect: The halide salts tend to dissolve the alumina, so eat up the arctube wall where the halide pool is residing during operation. To ensure it won't go through during the rated life, the material has to be thick enough there. And because the thick material absorbs more light, it is quite beneficial to locate this place somewhere, where the light does not go through anyway (that is e.g. the electrode area). And to ensure it will be really sitting there, you have to have that as the coldest spot in the tube. Hence the heat losses there become quite handy for keeping the pool off the thinner walls around the arctube center.

@Lanternbro: These "We have reached that performance" is common in all marketing all around, it is just demonstrating how far the technology could be pushed. There is nothing wrong with it, when it is still presented in that way and the datasheet (= the contract) is listing the realistic performance.
With the datasheets it was not always the case, about 5 years ago it was common to list just the lab-environment performance...

These days I see it is the tradditional light industry, who tend to tell only part of the truth: If you look in detail, you will see the minimum performance of each piece, not just the average over the production.
But with the LEDs you always knew each of them will emit as minimum at some reference condition, how it will behave on not that perfect ballast (= a bit different power than intended, due to any reason), so you have the datasheet and that is enough to design any product and guarantee the performance of each individual piece.

That is, what the "classic" lighting makers are severely lagging: Where do I find the minimum light output of e.g. the 100W incandescent I may count on with each piece? WhenI have to guarantee EACH of my product will perform, I have to know the worst performance of each lamp over the production.
Or do they want to say all have exactly the rated 1600lm? Then I immediately know that they are lying - because it just can never be exact, that is plain impossible.
Logged

No more selfballasted c***

Lanternbro
Member
**
Offline

Gender: Male
View Posts
View Gallery


Re: Maximum Efficiency of MH « Reply #11 on: May 09, 2015, 07:50:49 PM » Author: Lanternbro
The halide salts tend to dissolve the alumina


Are there any salts which don't?
Logged
Medved
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #12 on: May 10, 2015, 06:46:25 AM » Author: Medved
I don't know about any, apparently not among those usable in the lamps, otherwise they would be already widely used instead of the halides...

The problem with the salt selection is, they should readily vaporize so be able to carry the active elements into the arc stream, do not attack other lamp parts, even in the dissociated form do not interfere with the light generation,...
Logged

No more selfballasted c***

Lanternbro
Member
**
Offline

Gender: Male
View Posts
View Gallery


Re: Maximum Efficiency of MH « Reply #13 on: May 27, 2015, 07:21:41 PM » Author: Lanternbro
So why do salts for cooler CCT's give worse lumen deprecation?
Logged
Medved
Member
*****
Offline

Gender: Male
View Posts
View Gallery

Re: Maximum Efficiency of MH « Reply #14 on: May 28, 2015, 03:28:26 AM » Author: Medved
Nothing is perfect, so the reactivity of the active metals with the quartz (if they reach it before combining back with the halogens) or the alumina dissolving into the halides is still a problem.
The metals tend to react with quartz and turns it black. The ceramic (where the degradation does not mean lumen loss) could be used only up to ~4200K CCT, so for the highe you hae to use quartz.
Logged

No more selfballasted c***

Print 
© 2005-2024 Lighting-Gallery.net | SMF 2.0.19 | SMF © 2021, Simple Machines | Terms and Policies