This is topic Barco light sensor in forum Digital Cinema Forum at Film-Tech Forum ARCHIVE.


To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=16;t=002284

Posted by Jay Glaus (Member # 6011) on 04-26-2015, 06:07 PM:
 
I was touching up the alignment on my 32B today, and noticed something strange. I was watching the alignment chart on communicator, and as the bulb is on, it will be reading 10.50 Ft/L, and without really touching anything will plummet to 8.00 within a second. Then, eventually, it will come back up. I know sometimes the arc can wander and you can have a slight flicker, but I would have to believe a sudden drop of slightly over 2 Ft/L would be noticeable to the eye.

The other thing I thought was strange, was a few times it plummeted, no matter what adjustment you made, it really didn't change it until it decided to come back up. I honestly do not believe the brightness is changing as much as it is the number on the sensor. The bulb is fairly new, and i cannot really see any flicker.

I ended up aligning the bulb off my eye, because the number on communicator began to bounce so rapidly, it basically became useless. I tried closing and reopening communicator too, and rebooted the machine once. I'm sure the alignment isn't great, but it will get me through the night.

I don't have a light meter, but I know somebody who does. He's going to bring it out in a day, and we are going to try to align the bulb using an external light meter. I even tried making a new calibration file for the light sensor, on the assumption that i was getting 10 ft/l on the screen. Even though the calibration file would be incorrect, I wanted to see if it would still bounce with a new file, and it did.

I believe the light sensor in that 32 is going. Does anyone else agree?

Thanks,
Jay
 
Posted by Tim Sherman (Member # 569) on 04-26-2015, 06:36 PM:
 
Before you blame the CLO sensor,

First make sure you are not in CLO mode when doing a lamp alignment. It should be in normal mode

Second did anyone ever check the orientation of the magnet in the lamp house. There is a info-t on the 32B, some of them had magnets reversed when they were manufactured and that can cause an unstable arc.
 
Posted by Jay Glaus (Member # 6011) on 04-26-2015, 07:07 PM:
 
Hi Tim,

No, I'm not in CLO. In fact, I actually don't use CLO. maybe i'm old school from the days of film but I like to know right about where the bulb is ,and be the one to take it up if needed. That way I know if I have one darkening bad.

As for the magnet… Yes, ours was in backwards. We changed it shortly after the info-t came out a while back.
 
Posted by Marco Giustini (Member # 4544) on 04-27-2015, 01:15 AM:
 
Which version are you using? I've seen this with 1.13,but it may have been a lamp/ballast/sensor issue, it was the only machine I've seen with 1.13 so I can't be sure it was a software issue.
 
Posted by Jay Glaus (Member # 6011) on 04-27-2015, 02:10 AM:
 
Honestly I believe my 32 is still running 1.12, I need to update it. But I'll double check when I'm out there tomorrow. My communicator is 4.9.14.

My alignment by eye was better than I thought it would be. I'll be interested to hit it with the light meter Tuesday.

I did go around the front of the booth during projection, and stared straight down the lens to see how well I centered the bulb, and it looked pretty good for now. I have a new problem though. Theres a big pesky black spot on the front of my laptop screen that won't seem to go away no matter what I clean it with [Confused]
 
Posted by Monte L Fullmer (Member # 2797) on 04-28-2015, 05:24 PM:
 
Here is a topic from last year on the 32B stabilizing magnet issue.

Barco 32B Magnet

I had 4, 32B units with reversed magnets installed in the module. Bought a compass to check polarity and to help correct the polarity.

-Monte
 
Posted by Dave Macaulay (Member # 813) on 04-28-2015, 10:54 PM:
 
Beware of Barco 8/10 projectors, the CLO sensor is nutty. Light level reported depends on what's projected: black reads lowest and white reads highest. I haven't figured out why or how this happens. The lamp power will vary with scene brightness changes if it's run in CLO mode, I think maybe recent software locks out power changes some time after a macro is executed though.
 
Posted by Marco Giustini (Member # 4544) on 04-29-2015, 01:06 AM:
 
Wow, really! Thanks for the heads up.
 
Posted by Steve Guttag (Member # 268) on 04-29-2015, 06:52 AM:
 
I wonder if they are trying some primitive form of contrast ratio improvement. That said, it is normal for CLO to be locked out 5 minutes after a macro change (historically...I haven't verified this on an S series).
 
Posted by Pete Naples (Member # 761) on 04-29-2015, 08:01 AM:
 
I wonder if the CLO sensor is located after the DMDs? I too have observed this behaviour.

On a B or C you can happily line up the lamp with content playing, try it that way on a 8 or 10 and see how far you get!
 
Posted by Tim Sherman (Member # 569) on 04-29-2015, 10:38 AM:
 
Clo sensor is attached to the folding mirror before the Prism and the DMDs
 
Posted by Pete Naples (Member # 761) on 04-30-2015, 04:07 PM:
 
That's true for B & C but is it for S?

I don't follow why the CLO reading jumps all over the place with content.
 
Posted by Carsten Kurz (Member # 5396) on 04-30-2015, 04:53 PM:
 
Probably some light-spill or back-reflection. Just why didn't they catch this issue in time? Having a functioning light sensor/CLO is not the most exotic feature.

- Carsten
 
Posted by Marco Giustini (Member # 4544) on 05-01-2015, 03:17 PM:
 
It's a shame, the light calibration has always been very accurate on a Barco. Something Christie don't have.

If I service a projector with an old lamp and I calibrate the light, if the light level is - say - 12fL, on a Barco I can dial 14fL on the macro: when the lamp gets changed the CLO will adjust to reach 14fL. Same the other way round, when I find that the lamp installed is too powerful and I recommend for a smaller one: the projector will adjust to the correct level when that is done (within a reasonable accuracy). This cannot be done on a Christie.
 
Posted by Steve Guttag (Member # 268) on 05-02-2015, 01:57 AM:
 
That is not entirely correct. Christie uses a "target number". It is just not referenced to any particular fL reading. When I install a Christie projector, I set up any desired light loc calibrations I may need and name them appropriately. That is, if I want a 14fL Flat and a 14fL scope...I'll manually set the light to 14fL and then label the setting something like FLAT14fL and set the lightloc to that number. Likewise for the Scope format (and on down the line for any/all lens zoom or anamorphic settings). So while the Christie will not read out in fL (you can "calibrate" its sensor for ONE format, that is indeed inaccurate) you can set up as many suitable lightloc calibrations that are as accurate as your light meter. The projector will keep the lamp at that level to the best of its ability (and potentially to the lamp's detriment if allowed to go over 100%)

For sure though, Barco has everyone beat on the lamp calibration thing and they let you have it read out in fL with reasonable accuracy (not taking into account the optical items after the sensor...i.e. screen, port glass).

THE most worthless sensor is NEC's. It too is just a number but it has no basis on anything, including brightness. Setting for the highest number often doesn't mean the brightest or most even picture. That said, NEC will, like Christie allow you to "mark" your desired setting, save it, recall it and use "Feedback" to account for lamp aging. From what I can tell in my limited tests, it seems to lock onto your settings pretty well.

What Christie really beats the others on is lamp alignment. What a pleasure that thing is...it gets you 95-100% there and it does it pretty darn fast on the series 2 projectors. Barco's automatic lamp adjustments (if you buy the rather expensive upgrade) is horrendously slow (15 minutes or so) and is hit/miss at best on the alignment and seemingly never gets it 100%. That said, Barco is the clear winner on manual alignment...the lamp adjustments combined with the "green dots" makes it rather quick and accurate. NEC is the clear loser there with the most sloppy lamp adjustment that seemed to copy Christie's old "H" series lamp adjustments...with all of their failings.
 
Posted by Marco Giustini (Member # 4544) on 05-02-2015, 04:14 AM:
 
Sure Steve, that's what I do.

But what if you turn up servicing the projector and the lamp is due to be replaced and all you get on screen is 12fL? Can you tell the Christie to aim for a higher LiteLOC number for when the lamp is replaced or upgraded?

I installed a Barco with a 3kW lamp, light level was about 56fL. The customer purchased that machine in case they wanted 3D. The machine arrived with a 3kW lamp from Barco. I told the customer to install 2kW lamps instead. I did the colours, defocussed the lamp a little to have a more reasonable light output and then set the CLO to 14fL. The projector was running the 3kW lamp at min power. When the 3kW lamp was replaced with a 2kW one, CLO simply adjusted to 14fL. Sure, they may have not been precisely 14fL but close enough.

Had it been a Christie... I couldn't have done that. I would have probably disabled the LiteLOC altogether.

Agree on everything else!
 
Posted by Steve Guttag (Member # 268) on 05-02-2015, 11:37 AM:
 
Not so...defocus the lamp/lower the power until you get your light meter to read 14fL...then set the Lightloc there. If you want you can let it go back up to the original reading on the over-sized lamp...LightLoc will try to turn the lamp down within its range and, like the Barco, never get there...when the new smaller lamp is installed, it will still try to hit the target number you set up originally...just like the Barco. In truth, what Barco has is the ability to assign a meaningful number (Foot Lambert) to the internal number. Christie has a pathetic calibration procedure that can only be meaningful for one format (I tend to use FLAT as my calibration point). However, since they use Max/Min as their set points, they presume a linear scale, I guess because 14fL is never 14fL...I've occasionally tried to get it to read right but just didn't find the time worth it.

There is nothing to stop you from creating LightLoc files for 12fL, 13fL ...etc of each format too...you just need to do it with a calibrated meter to find what the LightLoc number is.
 
Posted by Marco Giustini (Member # 4544) on 05-02-2015, 11:42 AM:
 
Steve,

That is correct - even though cumbersome.
But the most common scenario is the other one: servicing a machine with an old lamp which does not reach 14fL anymore.
 
Posted by Steve Guttag (Member # 268) on 05-02-2015, 01:25 PM:
 
Okay...so how is that different than on a Barco? If the lamp doesn't reach the target level it doesn't reach it. The only difference is that you can specify your desired level in fL.
 
Posted by Marco Giustini (Member # 4544) on 05-02-2015, 05:47 PM:
 
That's exactly my point.

I described two scenarios:

1) the lamp is due to be changed - or overdue - and light levels in the room are only, say, 12fL. On a Christie you cannot tell LiteLOC to aim for 14fL. On a Barco you can.
2)The lamp cannot reach 14fL and you advise the customer to upgrade to a bigger one. You don't want to go back just to set light levels, do you? Just dial 14fL on CLO and when the new lamp is fitted it will be driven to the proper level with reasonable accuracy.
 
Posted by Steve Guttag (Member # 268) on 05-03-2015, 11:43 AM:
 
But you did calibrate the Christie sensor to some light level, right? So lets say you set Flat to 12fL and set LightLoc your example. You can then increase the power until it reads 14. Then increase the power until it reads 2fL more then call that setting 14fL. Likewise for Scope...if it normally reads something like 19fL with the old lamp...increase until it reads 21fL and make a new setting.

It isn't rocket science and in neither case are the light sensors as accurate as recalibrating with a proper light meter but sufficient for your customer to keep the light good and in the safe operating range.
 
Posted by Marco Giustini (Member # 4544) on 05-04-2015, 03:02 PM:
 
Maybe I'm on the wrong end of the stick here.
What if your lamp is producing 12fL at max power because it's an old lamp?
 
Posted by Jay Glaus (Member # 6011) on 05-05-2015, 12:05 AM:
 
Just an update…

Aligned the lamp on the screen using a light meter, and the number was pretty much rock solid. So, that has pretty much told me the bulb does not have much wander or flicker, but I figured it didn't.

I've been told to unplug and plug back in the CLO sensor, and see if that changes anything, if not, I may end up changing it.

The 32B is 3 years old and we haven't run anything smaller than a 6.5k bulb in her, I've heard the sensors can cook over time.
 
Posted by Marco Giustini (Member # 4544) on 05-05-2015, 11:41 AM:
 
the light sensor is BEHIND the fold mirror, it receives the 'leftover' from it. I don't think it can overheat.
 




Powered by Infopop Corporation
UBB.classicTM 6.3.1.2