LED Longevity Overrated? - Page 3 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #61 of 81 Old 09-16-2010, 06:48 PM
 
BobL's Avatar
 
Join Date: Nov 2001
Posts: 3,797
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 52
You can replace the LED module on the Samsung DLP RPTV but they are pricey ~$900 from Samsung parts.
BobL is offline  
Sponsored Links
Advertisement
 
post #62 of 81 Old 09-17-2010, 07:02 AM
AVS Special Member
 
Verge2's Avatar
 
Join Date: Jun 2004
Location: arkansas
Posts: 1,560
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 20
Quote:
Originally Posted by DaViD Boulet View Post

* LED has instand on/off ability which allows single-chip DLP to be driven without a mechanical color-wheel completely eliminating spoke time and any hint of "rainbow" artifacts for the majority of viewers who are sensitive to color-wheel discomfort with single-chip DLP designs.

How fast do they cycle?



With 3d being the new fad, how exactly are they planning on getting the lumens required?
Verge2 is offline  
post #63 of 81 Old 09-17-2010, 10:10 AM
Member
 
Mendo's Avatar
 
Join Date: Jun 2009
Location: Canton, OH
Posts: 52
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Verge2 View Post

How fast do they cycle?

Looking around on Google, the predominant answer seems to be under a microsecond. So somewhere around a million times a second.
Mendo is offline  
post #64 of 81 Old 09-17-2010, 11:10 AM
AVS Special Member
 
Verge2's Avatar
 
Join Date: Jun 2004
Location: arkansas
Posts: 1,560
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 10 Post(s)
Liked: 20
Quote:
Originally Posted by Mendo View Post

Looking around on Google, the predominant answer seems to be under a microsecond. So somewhere around a million times a second.

Doesn't mean the projector manufacturer will have them cycling at such a speed. That was my point.

There are rainbows on the cheapy LEDs out right now.
Verge2 is offline  
post #65 of 81 Old 09-17-2010, 01:54 PM
AVS Addicted Member
 
stanger89's Avatar
 
Join Date: Nov 2002
Location: Marion, IA
Posts: 17,311
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 115
Quote:
Originally Posted by Verge2 View Post

How fast do they cycle?

Cine4Home's tests had the Vivitek 9080 running 800-900Hz, or about 3x faster than the fastest current DLP color wheels. That would, I think, put it in the 20x range.

See what an anamorphoscopic lens can do, see movies the way they were meant to be seen
stanger89 is offline  
post #66 of 81 Old 09-17-2010, 02:41 PM
AVS Club Gold
 
JapanDave's Avatar
 
Join Date: Jul 2009
Posts: 2,321
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 98
Quote:
Originally Posted by Verge2 View Post

Doesn't mean the projector manufacturer will have them cycling at such a speed. That was my point.

There are rainbows on the cheapy LEDs out right now.

No rainbows on the Vango.

My build thread

My 8 x RE XXX 18" Subwoofers, IB build
Couldn't pour water out of a boot with instructions on the heel.

Do you know what Nemesis means?

JapanDave is online now  
post #67 of 81 Old 09-17-2010, 03:35 PM
AVS Special Member
 
MCaugusto's Avatar
 
Join Date: Aug 2001
Location: BOSTON --- MA ---
Posts: 1,196
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 5 Post(s)
Liked: 30
There are at least two online articles concerning "On/Off PhlatLight Chipset Cycling Speed" which stated that they cycle on/off at speeds of 2.9 KHz/second, or almost three million times per second.
The speed of the first generation 1x color wheels was 60 times/second (namely 60 Hz), or 3.600 RPM; Then they came up with 2x color wheels (120 Hz) or 7.200 RPM, and finally 3x color wheels (180 Hz) spinning at 10.800 RPM.
So, these PhlatLight chipsets cycling on/off at 2.9 KHz are theoretically 16 times faster than 3x color wheels running at 180 Hz.
Apparently color wheels cannot spin safely any faster than 10.000 to 15.000 RPM/second, so 3x color wheels running at 10.800 RPM is about as fast as the technology allows.
Am i correct in these figures and assumptions ?
BTW : Luminus Devices state that "PhlatLight chipsets run very cool because they emit directly into air unlike conventional LEDs which emit into an epoxy-based encapsulant, have very low failure rates and median lifetime that are well over 100.000 hours under extreme high current operating conditions and passed a rigorous suite of environmental and mechanical stress tests".
---------------------------------------------------------------------------------------------------------------------------
Marcos
MCaugusto is offline  
post #68 of 81 Old 09-17-2010, 10:32 PM
Advanced Member
 
Drexler's Avatar
 
Join Date: Jan 2008
Location: Copenhagen, Denmark
Posts: 912
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 22
MCAugusto,

They have double sets of color filters on the wheels (two reds, two blues, two greens) which makes the color wheel speed at 6X even though the wheel only rotates at 3X.
Drexler is offline  
post #69 of 81 Old 09-18-2010, 05:49 AM
AVS Special Member
 
MCaugusto's Avatar
 
Join Date: Aug 2001
Location: BOSTON --- MA ---
Posts: 1,196
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 5 Post(s)
Liked: 30
Quote:
Originally Posted by Drexler View Post

MCAugusto,

They have double sets of color filters on the wheels (two reds, two blues, two greens) which makes the color wheel speed at 6X even though the wheel only rotates at 3X.

---------------------------------------------------------------------------------------------
Yeah, i'm aware of the implementation of double-segment color wheels (RGBRGR) but is it fair to classify them as 6x wheels when they're still running at 3x speeds ?
If the idea behind double-segment color wheels was that RGB would be displayed at faster rates, which would imply less visible artifacts ("rainbows"), then why not make triple-segment wheels, quadruple-segment, etc ?
And whatever happened to that Archimedes color wheel design, which i believe had RGB spokes emanating from its center that progressively widened towards the edges of the wheel ?
Didn't Marantz had a 7-segment color wheel some years back, as in RGB/RGB/deepG ?
MCaugusto is offline  
post #70 of 81 Old 09-18-2010, 08:53 AM
AVS Addicted Member
 
stanger89's Avatar
 
Join Date: Nov 2002
Location: Marion, IA
Posts: 17,311
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 115
Quote:
Originally Posted by MCaugusto View Post

---------------------------------------------------------------------------------------------
Yeah, i'm aware of the implementation of double-segment color wheels (RGBRGR) but is it fair to classify them as 6x wheels when they're still running at 3x speeds ?
If the idea behind double-segment color wheels was that RGB would be displayed at faster rates, which would imply less visible artifacts ("rainbows"), then why not make triple-segment wheels, quadruple-segment, etc ?

Yes it is fair, for the reason you say the "x" refers to how many times RGB are repeated per frame, it's got nothing to do with the speed the wheel actually spins.

As for why not make more segments? Probably because the light beam that passes through it is of a finite, non-zero area, so each segment first has to be large enough to completely encompass the light beam, and has to do so, while moving/spinning for long enough to display a frame. I'd guess RGBRGB is about as small as you can practically get away with and still be able to display frames and keep everything in sync.

See what an anamorphoscopic lens can do, see movies the way they were meant to be seen
stanger89 is offline  
post #71 of 81 Old 09-18-2010, 10:17 AM
Senior Member
 
pjpoes's Avatar
 
Join Date: Feb 2007
Posts: 309
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Having been involved in some of the statistical analysis of product longevity data, one project involving a related LED technology, I can tell you that they have not actually tested their life span. Phatlight doesn't likely know, its estimated. The estimates are based on sophisticated statistical models which take into account actual known parameters, which have been measured. This makes the models relatively accurate.

For those familiar with statistics, you may also know that one potential pitfall in such models is that if you extrapolate out passed known ranges, the predicted values can be way off. If you utilized human growth characteristics for the first 10 years of life to extrapolate height and weight at 45 years, people would be 50 feet tall and 1000lbs. That isn't true because our height and weight from birth to adulthood is non-linear. The big difference is that much of this is known for devices like LED's, and so they are far more linear devices. Additionally, if we find that they do decay in a non-linear fashion (most bulbs do, but LED's are very different), we can account for that in the model and create more accurate predictions.

My understanding is that the specified life of the LED bulbs for these projectors is actually far less than what is being statistically predicted. LED's are known to dim very little over their life time compared with other bulbs, and we will not see them needing replacement every 2000 hours by any means. Getting 10,000 or 50,000 or even 100,000 hours is another story, I think we will see the low end, not the high end. Still, from even own look at the statistical data for the "related" devices, I'm fairly certain you won't see 50% loss in output at 50,000 hours, it will actually be more likely to just die than dim greatly.

Because of how they are currently being implemented, I am fairly certain they will be very expensive to replace. I'm sure these manufacturers are doing their best to make this as reasonable as possible, but its not going to be as simple as replacing the LED chip. An entire assembly including parts of the cooling components are going to be replaced, however the boards can be rebuild with new LED chips, as most of the other components will not age much.
pjpoes is offline  
post #72 of 81 Old 09-18-2010, 01:11 PM
AVS Special Member
 
DaViD Boulet's Avatar
 
Join Date: Aug 1999
Location: Washington DC area
Posts: 6,427
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 38
Quote:
Originally Posted by Verge2 View Post

Doesn't mean the projector manufacturer will have them cycling at such a speed. That was my point.

There are rainbows on the cheapy LEDs out right now.

Sigh. Can we keep to legit conversation rather than distraction?

The cheap LED PJs *use color wheels*. They are not using the actual RGB LEDs to sequence like higher-end HT models... they just push "white" light through a color wheel like old-fashioned bulb PJs.

Regarding brightness for 3D, LED will continue to get brighter and brighter. Does that need explaining???

1080p and lossless audio. EVERY BD should have them both.
DaViD Boulet is online now  
post #73 of 81 Old 09-18-2010, 01:14 PM
AVS Special Member
 
DaViD Boulet's Avatar
 
Join Date: Aug 1999
Location: Washington DC area
Posts: 6,427
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 38
Quote:
Originally Posted by stanger89 View Post

Yes it is fair, for the reason you say the "x" refers to how many times RGB are repeated per frame, it's got nothing to do with the speed the wheel actually spins.

As for why not make more segments? Probably because the light beam that passes through it is of a finite, non-zero area, so each segment first has to be large enough to completely encompass the light beam, and has to do so, while moving/spinning for long enough to display a frame. I'd guess RGBRGB is about as small as you can practically get away with and still be able to display frames and keep everything in sync.

Color wheels waste light because the DMD "goes black" while the color transitions from one color to another. The more color segments you add to a wheel the more time is wasted. This is called "spoke time".

LED would be instant on/off with *no* spoke time at all. It would mean you could have a DMD that cycles colors incredibly fast bcs you wouldn't have any disadvantage with faster color sequencing as long as your DSP engine could handle it.

1080p and lossless audio. EVERY BD should have them both.
DaViD Boulet is online now  
post #74 of 81 Old 09-18-2010, 06:38 PM
AVS Addicted Member
 
mark haflich's Avatar
 
Join Date: Dec 2000
Location: brookeville, maryland, usa
Posts: 19,281
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 112 Post(s)
Liked: 312
David. Does that mean that bulbs will be getting brighter and brighter or that posters will be getting brighter and brighter?

Mark Haflich
markhaflich@yahoo.com
call me at: 240 876 2536
mark haflich is online now  
post #75 of 81 Old 09-19-2010, 12:58 PM
AVS Special Member
 
DaViD Boulet's Avatar
 
Join Date: Aug 1999
Location: Washington DC area
Posts: 6,427
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 38
Quote:
Originally Posted by mark haflich View Post

David. Does that mean that bulbs will be getting brighter and brighter or that posters will be getting brighter and brighter?


1080p and lossless audio. EVERY BD should have them both.
DaViD Boulet is online now  
post #76 of 81 Old 09-21-2010, 09:52 AM
Advanced Member
 
mlang46's Avatar
 
Join Date: Aug 2006
Location: Boulder, Colorado
Posts: 905
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by pjpoes View Post

Having been involved in some of the statistical analysis of product longevity data, one project involving a related LED technology, I can tell you that they have not actually tested their life span. Phatlight doesn't likely know, its estimated. The estimates are based on sophisticated statistical models which take into account actual known parameters, which have been measured. This makes the models relatively accurate.

For those familiar with statistics, you may also know that one potential pitfall in such models is that if you extrapolate out passed known ranges, the predicted values can be way off. If you utilized human growth characteristics for the first 10 years of life to extrapolate height and weight at 45 years, people would be 50 feet tall and 1000lbs. That isn't true because our height and weight from birth to adulthood is non-linear. The big difference is that much of this is known for devices like LED's, and so they are far more linear devices. Additionally, if we find that they do decay in a non-linear fashion (most bulbs do, but LED's are very different), we can account for that in the model and create more accurate predictions.

My understanding is that the specified life of the LED bulbs for these projectors is actually far less than what is being statistically predicted. LED's are known to dim very little over their life time compared with other bulbs, and we will not see them needing replacement every 2000 hours by any means. Getting 10,000 or 50,000 or even 100,000 hours is another story, I think we will see the low end, not the high end. Still, from even own look at the statistical data for the "related" devices, I'm fairly certain you won't see 50% loss in output at 50,000 hours, it will actually be more likely to just die than dim greatly.

Because of how they are currently being implemented, I am fairly certain they will be very expensive to replace. I'm sure these manufacturers are doing their best to make this as reasonable as possible, but its not going to be as simple as replacing the LED chip. An entire assembly including parts of the cooling components are going to be replaced, however the boards can be rebuild with new LED chips, as most of the other components will not age much.

Do they perform accelerated (increased Temperature tests) on Leds as they do on lasers than extrapolate the room temperature lifetimes from those tests.
mlang46 is offline  
post #77 of 81 Old 09-21-2010, 03:12 PM
AVS Special Member
 
Brandon B's Avatar
 
Join Date: Apr 2001
Location: LA lalalalala
Posts: 3,971
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by DaViD Boulet View Post

Sigh. Can we keep to legit conversation rather than distraction?

The cheap LED PJs *use color wheels*. They are not using the actual RGB LEDs to sequence like higher-end HT models... they just push "white" light through a color wheel like old-fashioned bulb PJs.

Sigh back. No they don't. At least not any of the dozen or so models of pocket and pico LED projectors we are buying. They are all RGB LED sequencing models that emulate very low color wheel speed and rainbow like hell. And I know this for a fact because we have taken all of them apart.

We have seen only one pico unit that used white LEDs and LCOS out of all of them.
Brandon B is offline  
post #78 of 81 Old 09-21-2010, 03:17 PM
AVS Special Member
 
Brandon B's Avatar
 
Join Date: Apr 2001
Location: LA lalalalala
Posts: 3,971
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by pjpoes View Post

Having been involved in some of the statistical analysis of product longevity data, one project involving a related LED technology, I can tell you that they have not actually tested their life span. Phatlight doesn't likely know, its estimated. The estimates are based on sophisticated statistical models which take into account actual known parameters, which have been measured. This makes the models relatively accurate.

. . .

Getting 10,000 or 50,000 or even 100,000 hours is another story, I think we will see the low end, not the high end.

As I posted earlier in this thread, we actually ran a few units out to their projected lifetimes of 20,000 hours and saw brightness drops in the range the manufacturer claimed.

But these were <100 lumen pocket models, not HT units where they are pushing the envelope to get usable brightness, so you are probably correct in your predictions nonetheless.
Brandon B is offline  
post #79 of 81 Old 09-21-2010, 04:10 PM
AVS Special Member
 
DaViD Boulet's Avatar
 
Join Date: Aug 1999
Location: Washington DC area
Posts: 6,427
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 38
Quote:
Originally Posted by Brandon B View Post

Sigh back. No they don't. At least not any of the dozen or so models of pocket and pico LED projectors we are buying. They are all RGB LED sequencing models that emulate very low color wheel speed and rainbow like hell. And I know this for a fact because we have taken all of them apart.

We have seen only one pico unit that used white LEDs and LCOS out of all of them.

Thanks for the correction. However, even if there are cheap designs using LED light source that, even in the absence of a color wheel, don't bother to implement sequencing speed fast enough to eliminate color separation artifacts, that doesn't negate that LED light source makes possible the elimination of color-separation artifacts for a product designed with high-quality images as an objective. There are products in every technology category that fall well below the performance levels possible when cost or size or some other non-fidelity criteria is the primary objective of the product designers.

Otherwise, we can't talk about the benefits of *any* technology because there's a bad implimentation of everything out there that doesn't live up to the technology's capability.

In a few short year's LED lamp sources will be able to replace bulbs in single-chip DLP HT oriented projectors and in doing so eliminate color separation artifacts for virtually all viewers. Naturally there will be cheap products available that don't live up to the technology's capability. Just like there are cheap bulb-lit DLP units today that don't live up to bulb-based DLP performance levels we're accustomed to in HT products. That doesn't say anything about "bulbs", it just demonstrates how cheap products cut corners.

1080p and lossless audio. EVERY BD should have them both.
DaViD Boulet is online now  
post #80 of 81 Old 09-21-2010, 07:08 PM
AVS Special Member
 
FiveMillionWays's Avatar
 
Join Date: Jun 2004
Location: Dallas, Texas
Posts: 1,184
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I use my led tv every day and leave it on over night and I have yet to notice any light output change. DLP LED TV!!!
FiveMillionWays is offline  
post #81 of 81 Old 09-22-2010, 12:34 AM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Sequoia, CA
Posts: 30,062
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 35 Post(s)
Liked: 382
Quote:
Originally Posted by FiveMillionWays View Post

I use my led tv every day and leave it on over night and I have yet to notice any light output change. DLP LED TV!!!

Thank goodness we have an accurate, calibrated test of light loss. Your efforts are appreciated.

There is no difference in HDMI cables. If you can see the picture without visible dropouts or sparklies, the cable is working at 100%. No other cable will display a better version of that picture. You're simply wrong if you think there is a better digital cable than one that is already working.
rogo is offline  
Reply Digital Hi-End Projectors - $3,000+ USD MSRP

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off