or Connect
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › JVC 2011-2012 lamp measurement project
New Posts  All Forums:Forum Nav:

JVC 2011-2012 lamp measurement project - Page 9

post #241 of 390
Quote:
Originally Posted by jim2100 View Post

I don't know how I could have been more clear. What part of "sk576c never signed up for the JVC lamp measurement project, so we should not draw any conclusions about failure rate based on data posted by sk576c" is difficult to understand?

ok, then I did understand and not really sure what your point is then. He posted the same valid data as anyone else, so what difference does it make if he didn't sign up for your project?

the majority of people haven't been posting with regular results, so when a member who has a lamp that goes bad in a similar way that we saw last year has a major drop, it's going to finally draw some attention to this thread which has been relatively dormant for a while.
post #242 of 390
Thread Starter 
Quote:
Originally Posted by WTS View Post

What because one guy doesn't sign up for this thread then you should discount his lamp problems.lol, wow take a pill pal, who cares if you're thread starter, ridiculous.

I am surprised that people are having trouble understanding my statement. I picked my words carefully. Please read EXACTLY what I wrote, and do not try to "read between the lines". I wrote exactly what I meant. I did NOT say that data from sk576c was "invalid", nor did I say that lamp problems of sk576c should be "discounted".

I did say that we should not draw conclusions about the failure rate based on data from sk576c. This study (and this thread) was intentionally limited to only people who signed up by a certain date in order to be able to calculate a reasonably accurate failure rate. Data from anyone who was not signed up cannot be used to calculate the failure rate in this project.
post #243 of 390
Thread Starter 
Quote:
Originally Posted by zombie10k View Post

ok, then I did understand and not really sure what your point is then. He posted the same valid data as anyone else, so what difference does it make if he didn't sign up for your project?

The difference is that only data from people who signed up can be used to calculate the failure rate for this project.

A failure rate is calculated as a ratio: failures / sample_size. You need to know the numerator and the denominator.

If you start including data from people who did not sign up, then you no longer know the denominator and cannot compute a failure rate.
post #244 of 390
Jim, thanks for explaining your method. I can't argue about data collection & failure rates, but I am guessing that there's going to be folks who signed up for the project that aren't going to participate (maybe they don't have time, sold the projector, etc)

Doesn't that affect the results if 50 people signed up and only 30 members participate? If this was a controlled experiment and guaranteed that 50 people were submitting data on a regular basis, I could certainly understand not wanting to skew the results by adding additional data.
post #245 of 390
Thread Starter 
Quote:
Originally Posted by zombie10k View Post

Doesn't that affect the results if 50 people signed up and only 30 members participate? If this was a controlled experiment and guaranteed that 50 people were submitting data on a regular basis, I could certainly understand not wanting to skew the results by adding additional data.

It is not a matter of "wanting". It is a matter of "impossibility". We cannot calculate an accurate failure rate without knowing the sample size.

As for some people not reporting, that does affect the accuracy of the failure rate we compute. But it is always possible to compute a failure rate if we have any people from the sample who report, since we know the sample size and we know the number of failures.

In addition, we can make some reasonable assumptions that increase our confidence in the failure rate that we compute. For example, it is likely that most people who signed up (and got a projector) and had a major drop in brightness are going to post about it, so the number of failures in the sample that are not reported is likely to be small.

Also, I have been planning to contact everyone who signed up and has not posted any data, asking them: did they get a projector? if so, why have they not posted any data? have they been using the projector? have they returned it? if they have not measured the brightness, how long have they had the projector and how many hours have they put on the lamp, and has the brightness seemed to decrease excessively to their eye?

From these answers, I can eliminate people from the sample who did not get and use a new JVC projector, and maybe we can build our confidence that most of the rest of the people who have not reported any data are happily using their projectors without noticing any problems.

As for contacting people who signed up and have not posted any data, zombie10k, where is your first measurement data?
post #246 of 390
I knew that was coming. My first RS55 had some issues right out of the box and didn't bother with the measurements until the replacement arrived.

This projector has 6 hours, but I have zero hour lamp I am starting with and the next time I fire it up, will post the beginning lumens. I have a number of other projectors to play around with and only use the RS55 for 2D BD, so it's going to be a while before I even hit 100 hours.
post #247 of 390
My 3rd measurements are as follows:

1. 02/24/12
2. JVC RS-45
3. 100IRE Internal test pattern on RS-45
4. 460 Hours on lamp
5. Normal lamp primarily used
6. measurement taken with normal lamp
7. lens aperture set to 0
8. measurements taken approximately 60 minutes after warm up
9. 628 lumens using center screen measurement method

***Previous lumen measurement was 795 at 245 lamp hours; approximate decrease in lumens is 21%*** Decrease in lumens from baseline measurement of 907 lumens is approximately 31% at 460 lamp hours
post #248 of 390
Quote:
Originally Posted by jim2100 View Post

If you start including data from people who did not sign up, then you no longer know the denominator and cannot compute a failure rate.

If you're trying to compute a failure rate based on participation in a web forum, you're probably already dealing with fairly meaningless data.
post #249 of 390
Date 25-Feb
Hours 89
Screen Size 137"
Lux 106
Lumens 554
Model JVC X90
Test Image AVS709 Startup Screen center between "S" and "H"
Lamp Power Normal
Lens Ap 0
Image Setting Natural
Warm Up 60
post #250 of 390
It appears there is a new generation of lamps coming from JVC for the 2011/2012 projectors. I talked this week to JVC reps in several locations (Chicago, Texas, Calif.) and was told by a technical type at the JVC repair facility in Long Beach that they are expecting a new generation of lamps to arrive within a few weeks. A JVC customer service center rep. in Texas told me the lamps are now out of stock and they expect it will be April when they have a new supply to fill current orders. Its not clear however if this will simply be another minor update from the same lamp manufacturer in China that JVC has been using for the 2011/2012 models or perhaps the lamps will be sourced from another supplier (we can always hope). The tech guy at JVC in Long Beach volunteered he hoped (and seemed to have some reason to believe) the new lamps will correct the issues they have been having with the lamps for the past year.
post #251 of 390
Good info Ron, I don't mean to sound jaded, but it might be more false hopes.
Sorry, but that's just me, hope I'm wrong...
post #252 of 390
Quote:
Originally Posted by Ron Jones View Post

It appears there is a new generation of lamps coming from JVC for the 2011/2012 projectors. I talked this week to JVC reps in several locations (Chicago, Texas, Calif.) and was told by a technical type at the JVC repair facility in Long Beach that they are expecting a new generation of lamps to arrive within a few weeks. A JVC customer service center rep. in Texas told me the lamps are now out of stock and they expect it will be April when they have a new supply to fill current orders. Its not clear however if this will simply be another minor update from the same lamp manufacturer in China that JVC has been using for the 2011/2012 models or perhaps the lamps will be sourced from another supplier (we can always hope). The tech guy at JVC in Long Beach volunteered he hoped (and seemed to have some reason to believe) the new lamps will correct the issues they have been having with the lamps for the past year.

What troubles me about this information is that, if true, it is more or less an admission of a faulty design of the original lamps. They should now provide a free replacement for anyone who has an issue with the old design. I do not get the chance to use my home theater as much as I would like and, as such, only have ~230 hours on my second lamp (the first one died at 100 hours). I know the second lamp is going to die soon and I sure as hell am not paying $450 for a new one. It doesn't see fair that I should have to when I will more than likely get less than 500 hours out of the first two lamps.

John
post #253 of 390
Quote:
Originally Posted by nohjy View Post

What troubles me about this information is that, if true, it is more or less an admission of a faulty design of the original lamps. They should now provide a free replacement for anyone who has an issue with the old design. I do not get the chance to use my home theater as much as I would like and, as such, only have ~230 hours on my second lamp (the first one died at 100 hours). I know the second lamp is going to die soon and I sure as hell am not paying $450 for a new one. It doesn't see fair that I should have to when I will more than likely get less than 500 hours out of the first two lamps.

John

The reason I did not post the name of the JVC tech guy that offered his opinion of the lamps was because what he was saying is not the official JVC party line and I don't want to get the guy in trouble. When I talked to the JVC professional products marketing types about the lamp issue back in Sept. at CEDIA, they made is clear it is not "JVC's way" to do a general recall or officially announce a warrenty extension and they planned to handle lamp issues on a case-by-case basis. I do know they are still replacing failed/degraded lamps beyond not only the official 90 day warranty period for lamps but even beyond 1 year as long as the hours on the lamp are reasonably low.
post #254 of 390
just wondering if this is a good lux meter to get.

http://www.amazon.com/gp/product/B00...=ATVPDKIKX0DER

I also have a syder3. Can I take measurement with that and how.

Thanks.
post #255 of 390
Quote:
Originally Posted by space2001 View Post

just wondering if this is a good lux meter to get.

http://www.amazon.com/gp/product/B00...=ATVPDKIKX0DER

I also have a syder3. Can I take measurement with that and how.

Thanks.

I don't have experience with that model, but HERE is a less expensive model that several of us are using that seems to be fairly accurate.
post #256 of 390
26 bucks to measure how much misery I,m going thru..
It's nice to know suffering is not to expensive!
I feel better all ready!!

Regards
post #257 of 390
Quote:
Originally Posted by Ron Jones View Post

It appears there is a new generation of lamps coming from JVC for the 2011/2012 projectors. I talked this week to JVC reps in several locations (Chicago, Texas, Calif.) and was told by a technical type at the JVC repair facility in Long Beach that they are expecting a new generation of lamps to arrive within a few weeks. A JVC customer service center rep. in Texas told me the lamps are now out of stock and they expect it will be April when they have a new supply to fill current orders. Its not clear however if this will simply be another minor update from the same lamp manufacturer in China that JVC has been using for the 2011/2012 models or perhaps the lamps will be sourced from another supplier (we can always hope). The tech guy at JVC in Long Beach volunteered he hoped (and seemed to have some reason to believe) the new lamps will correct the issues they have been having with the lamps for the past year.

Ron,

I have seen reference to a -003 lamp in posts on the rs40 thread. I assume your post above means the April bulb will be a -004 model. Or, does it mean that they ran out of -003's? A post on other thread said -003 bulb had some kind of new part inside the lamp enclosure.

My rs40 went dim at 584 hours. I am not registered for this survey so please don't let this info mess up the ratios.
post #258 of 390
Quote:
Originally Posted by Ron Jones View Post


I don't have experience with that model, but HERE is a less expensive model that several of us are using that seems to be fairly accurate.

Wish I could order that one but it won't ship to Canada
post #259 of 390
Quote:
Originally Posted by Ron Jones View Post

The tech guy at JVC in Long Beach volunteered he hoped (and seemed to have some reason to believe) the new lamps will correct the issues they have been having with the lamps for the past year.

Yeah, but this isn't the first time we've heard that.
post #260 of 390
Quote:
Originally Posted by JackB View Post

Ron,

I have seen reference to a -003 lamp in posts on the rs40 thread. I assume your post above means the April bulb will be a -004 model. Or, does it mean that they ran out of -003's? A post on other thread said -003 bulb had some kind of new part inside the lamp enclosure.

I'm not sure if those numbers even mean anything. I ordered a spare lamp a few weeks ago, and got a -002.
post #261 of 390
Quote:
Originally Posted by space2001 View Post

Wish I could order that one but it won't ship to Canada

HERE is an eBay vendor selling the same meter that does ship to Canada.
post #262 of 390
With all this talk about lamps dimming at a much faster then normal rate... Bottom line...

Would this issue with the current generations lamps steer you in a different direction then JVC if you could do it over again?

I personally have the JVC DLA-HD1 (RS1) and have been very happy with the lamp life and lumens on my projector. I know that the problem seems to have started in the last years models and continued on to this years. I was looking into upgrading my HD1 to a DLA X30 but this could be a solid reason to perhaps wait till next years models. Thoughts?
post #263 of 390
You have to be very leery of believing the lux numbers these cheap meters read when measuring projectors. They are designed and calibrated for environmental light measurements, not spectra like we get with projectors. I think these meters are ok for measuring the dimming of your pj lamps as long as the same meter is used each time.

Here is a little experiment I ran today to show how much different these meters can read when measuring different spectra.

I have 3 meters that will measure lux directly: a Konica-Minolta CL-200 (a $3000 meter that presumably is the most accurate), an AEMC CA813, and an LX1330B. The latter two are popular here on the forum, the CA813 for a long time and the LX1330B is the one Ron Jones recommended a few posts ago. The LX1330B is the newest as I just bought it today.

First, measuring projectors. I used a 100% white window from my Lumagen XE, using the smallest window it produces. This allows me to precisely position each meter's sensor. I have 2 projectors: a Sim2 MICO 50 and a Sim2 C3X720. The spectra of these two are quite different with the MICO using RGB LED's as the light source and the C3X using a UHP lamp. Here is what I measured:

MICO 50
Konica Minolta CL-200
113 lx equating to 509 lumens for my screen

CA813
103 lx (-9% from CL-200), 464 lumens

LX1330B
128 lx (+13% from CL-200), 576 lumens

C3X720
Konica Minolta CL-200
213 lx equating to 960 lumens for my screen

CA813
176 lx (18% lower than CL-200), 792 lumens

LX1330B
224 lx (5% higher than CL-200), 1008 lumens

Thus, for my MICO 50 I can get readings from 464 to 576 lumens depending on which meter is used.

And, for the C3X720 I can get readings from 792 to 1008 lumens.

Which one is right? I believe the closest is the Konica-Minolta CL-200 which correlates with off the screen measurements using my i1 Pro. However, calculating lumens from the i1Pro requires me to know my screen gain, which I think I know pretty close.

As a further check, I used all three meters to measure the lumen output from one of the incandescent spots in my home theater. In this case, the KM CL-200 and the LX1330B were very close, with the CL-200 measuring 201 lx and the LX1330B measuring 196 lx, within 2% or so of each other. This incandescent light is probably closer to what the meters were calibrated to. The CA813 measured quite low at 167 lx, down 17% from the CL-200.

The only point I'm trying to make here is to not fixate on the absolute lux measurement from these cheap meters as they are probably wrong. Comparing what your projector puts out to someone else's using different meters is probably not a good idea. Using these cheap meters to measure trends, or how your lamp is ageing, is a much better idea.

Bob

Addendum: In the manual for the LX1330B it says: "Calibrated to standard incandescent lamp at color temperature 2856K". The Konica-Minolta CL-200 is calibrated with two user-selectable modes: one with a UHP lamp from a projector and the other using an 800lux standard illuminate A. I always use the mode that's calibrated with the UHP lamp. This further supports my opinion that the CL-200 is the more accurate of these meters.
post #264 of 390
My Lux readings X90 Batcave.

iris open
standard/6500k/normal/
110"diag 1.4 BD 16:9 screen/15ft throw
Calman 100 IRE white pattern
60 mins warm up
Center screen readings

Low/High lamp
171/295 100 hours Jan 21 2012
163/252 150 hours Feb 04 2012
142/210 200 hours Feb 14 2012 Starting to dim!!!
120/182 250 hours March 01 2012
post #265 of 390
Late to the party, and not posting as a plug since I expect most people here already have the DVD but...

For measuring ANSI lumens, there is a specific pattern on the GetGray calibration DVD for same. It has the 9 (or 13) circles marked where you are supposed to measure. Further the pattern is encoded "above white" as to maximize the brightness. After all that is what you want to know, the maximum lumens. If you have your PJ calibrated to show above white information as most people do, and then you use a normal 100%/100IRE pattern you will not be capturing the full brightness of the unit. By using an above white pattern you will ensure you have teh correct pattern to measure your PJ's true max white point (and thus max lumens).

From www.calibrate.tv:

post #266 of 390
Quote:
Originally Posted by Citation4444 View Post

The only point I'm trying to make here is to not fixate on the absolute lux measurement from these cheap meters as they are probably wrong. Comparing what your projector puts out to someone else's using different meters is probably not a good idea. Using these cheap meters to measure trends, or how your lamp is ageing, is a much better idea.

Bob

Bob is absolutely correct here. There's an old thread on these meters in the calibration forum that goes into more detail, but I couldnt' find it May be archived.
post #267 of 390
Quote:
Originally Posted by Kevin 3000 View Post

My Lux readings X90 Batcave.

iris open
standard/6500k/normal/
110"diag 1.4 BD 16:9 screen/15ft throw
Calman 100 IRE white pattern
60 mins warm up
Center screen readings

Low/High lamp
171/295 100 hours Jan 21 2012
163/252 150 hours Feb 04 2012
142/210 200 hours Feb 14 2012 Starting to dim!!!
120/182 250 hours March 01 2012

This is already unacceptable...once again like another X90 user, showing around 30% drop within 100 hours. JVC should/will replace the bulb at this point.
post #268 of 390
Quote:
Originally Posted by Citation4444 View Post

You have to be very leery of believing the lux numbers these cheap meters read when measuring projectors. They are designed and calibrated for environmental light measurements, not spectra like we get with projectors. I think these meters are ok for measuring the dimming of your pj lamps as long as the same meter is used each time.

Here is a little experiment I ran today to show how much different these meters can read when measuring different spectra.

I have 3 meters that will measure lux directly: a Konica-Minolta CL-200 (a $3000 meter that presumably is the most accurate), an AEMC CA813, and an LX1330B. The latter two are popular here on the forum, the CA813 for a long time and the LX1330B is the one Ron Jones recommended a few posts ago. The LX1330B is the newest as I just bought it today.

First, measuring projectors. I used a 100% white window from my Lumagen XE, using the smallest window it produces. This allows me to precisely position each meter's sensor. I have 2 projectors: a Sim2 MICO 50 and a Sim2 C3X720. The spectra of these two are quite different with the MICO using RGB LED's as the light source and the C3X using a UHP lamp. Here is what I measured:

MICO 50
Konica Minolta CL-200
113 lx equating to 509 lumens for my screen

CA813
103 lx (-9% from CL-200), 464 lumens

LX1330B
128 lx (+13% from CL-200), 576 lumens

C3X720
Konica Minolta CL-200
213 lx equating to 960 lumens for my screen

CA813
176 lx (9% lower than CL-200), 792 lumens

LX1330B
224 lx (13% higher than CL-200), 1008 lumens

Thus, for my MICO 50 I can get readings from 464 to 576 lumens depending on which meter is used.

And, for the C3X720 I can get readings from 792 to 1008 lumens.

Which one is right? I believe the closest is the Konica-Minolta CL-200 which correlates with off the screen measurements using my i1 Pro. However, calculating lumens from the i1Pro requires me to know my screen gain, which I think I know pretty close.

As a further check, I used all three meters to measure the lumen output from one of the incandescent spots in my home theater. In this case, the KM CL-200 and the LX1330B were very close, with the CL-200 measuring 201 lx and the LX1330B measuring 196 lx, within 2% or so of each other. This incandescent light is probably closer to what the meters were calibrated to. The CA813 measured quite low at 167 lx, down 17% from the CL-200.

The only point I'm trying to make here is to not fixate on the absolute lux measurement from these cheap meters as they are probably wrong. Comparing what your projector puts out to someone else's using different meters is probably not a good idea. Using these cheap meters to measure trends, or how your lamp is ageing, is a much better idea.

Bob

Addendum: In the manual for the LX1330B it says: "Calibrated to standard incandescent lamp at color temperature 2856K". The Konica-Minolta CL-200 is calibrated with two user-selectable modes: one with a UHP lamp from a projector and the other using an 800lux standard illuminate A. I always use the mode that's calibrated with the UHP lamp. This further supports my opinion that the CL-200 is the more accurate of these meters.

Most general purpose Lux light meters, such as the LX1330B or CA813, claim to have a curve based on the CIE standard light curve (goes back 80+ years) that is based on the sensitivity of human sight across the visible spectrum. The peak sensitivity on that curve corresponds to a color temp near 3000K while the 6500K we use for video is considered toward the end of the spectrum that is considered to produce cool to very cool colors. 6500K is said to correspond to the color of natural light on a cloudy day. So with any of these meters when they are used to measure a 6500K source the peak of the light energy being output from the projector occurs on the side skirt of the meters sensitivity curve. This can magnify what appears to be small differences in the overall shape of the sensitivity curve from one meter to the next into larger differences in the meter's readings.

As for using a i1Pro (or any similar probe) to do the readings off of the screen, I find that this can difficult to get accurate lumens because if the screen has any gain/directionality and you tilt the probe (ie. the normal method) then you will be measuring off axis and you would need to account for the reduced screen gain.

By the way I did note that in your results with the C3X-720 projector (which has a light source typical with what most of our projector's have) the lux/lumens reading with the LX1330B and the Minolta meter actually agreed well. However your stated % difference appears to be incorrect. Specifically you said you measured 213 lux with the Minolta and 224 lux with the LX1330B which is (224/213) = 1.0516 or the LB1330B measured 5% higher than the Minolta. I also see that you results were even more closely matched when measuring your HT incandescent light output. The spectrum of an incandescent light is a fairly close match for the CIE standard curve so meters that use the CIE curve should be most consistent between meters and also most accurate with such light sources. It does appear from your results, projectors using LED light sources may have a unique enough spectrum to have an significant impact on the readings from such meters.

I do agree with your bottom line conclusion that the best use for any of these meters is to compared two or mores sets of readings using the same meter to determine relative changes/differences. These meters should also work well for comparing the light output of different projectors that use the same type of light source. However, I would add the following caveat: as the lamp ages the color temp will change. So, for accurate readings of lumens over the life of the lamp on the same projector, the color temperature would need to be kept to near 6500K which may require periodic recalibration.
post #269 of 390
Quote:
Originally Posted by Ron Jones View Post

As for using a i1Pro (or any similar probe) to do the readings off of the screen, I find that this can difficult to get accurate lumens because if the screen has any gain/directionality and you tilt the probe (ie. the normal method) then you will be measuring off axis and you would need to account for the reduced screen gain.

By the way I did note that in your results with the C3X-720 projector (which has a light source typical with what most of our projector's have) the lux/lumens reading with the LX1330B and the Minolta meter actually agreed well. However your stated % difference appears to be incorrect. Specifically you said you measured 213 lux with the Minolta and 224 lux with the LX1330B which is (224/213) = 1.0516 or the LB1330B measured 5% higher than the Minolta. I also see that you results were even more closely matched when measuring your HT incandescent light output. The spectrum of an incandescent light is a fairly close match for the CIE standard curve so meters that use the CIE curve should be most consistent between meters and also most accurate with such light sources. It does appear from your results projectors using LED light sources may have a unique enough spectrum to have an significant impact on the readings from such meters.

I do agree with your bottom line conclusion that the best use for any of these meters is to compared two or mores sets of readings using the same meter to determine relative changes/differences. These meters should also work well for comparing the light output of different projectors that use the same type of light source. However, I would add the following caveat: as the lamp ages the color temp will change. So, for accurate readings of lumens over the life of the lamp on the same projector, the color temperature would need to be kept to near 6500K which may require periodic recalibration.

Thanks for catching my typos in the % errors in measuring the C3X720. It resulted from cut and pasting the MICO 50 results and not correcting the % errors. The lux measurements were correct, however, no typos there. I have corrected my original post.

As far as the i1Pro and measuring off the screen, I fully understand all the difficulties involved, and pointed this out in my post. However, for most of my historical measurements I used a fairly non-directional screen with 1.3 gain. But your point is well taken.

LED light sources are sometimes quite challenging depending on the specific LED and design of the meter. With the increased use of matrixed LED's in general lighting applications, the literature (and of course the 'net) is littered with discussions as to which meters are appropriate for measuring such lights. Many discussions of building integrating spheres, etc., are quite interesting. Our hobby, i.e. HT projectors, doesn't even budge the meter in these discussions. I am always sceptical of reviews stating the lumen output of LED-powered projectors when they don't specify the measuring equipment.

The LX1330B does appear to be a decent meter at a bargain price.

It's amazing the technology changes I've witnessed since I started my engineering career some 50+ years ago.
post #270 of 390
Quote:
Originally Posted by Citation4444 View Post

...........................It's amazing the technology changes I've witnessed since I started my engineering career some 50+ years ago.

I guess you have a decade up on me. I only got my EE degree 41 years ago and starting working as an engineer after another couple of years in graduate school.
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › JVC 2011-2012 lamp measurement project