XL2 Vertical Banding/Dropout at DVinfo.net

Bill Masters · May 16th, 2010, 06:54 AM

Hi all. Our small video production business uses 2 Canon XL2s for most of our productions. On a recent project we had a problem with one of the cameras and I am hoping that someone here might be able to explain what may have happened.

The scenario is 2 cameras running simultaneously then the footage is synced and edited. Both tapes from camera 1 were perfect. Tape 1 from camera 2 was also perfect. However, tape 2 from camera 2 is all but ruined. There is a band of pixelation running vertically on the left side of the footage. The band is around 15% of the width of the frame (16:9) and is about 15% in from the left edge. So, what I have is 15% image, then 15% pixellation then the remaining 70% of the image to the right of the pixellation. The pixellation runs for the duration of the shot which is around 35 minutes. Once the camera was paused and recording recommenced the pixellation disappeared. The camera was not switched off during this time.

The camera is routinely serviced by Canon, so no problems there I am fairly certain. I had run a head cleaner prior to the shoot and used a brand new Panasonic Professional tape. No warnings appeared on the screen at any time.

I am wondering if, when the operator changed tapes, the cause of this issue could be that the recording process was started too hastily and the tape may not have been fully engaged in the carriage mechanism? Or could there have been some foreign matter which freed itself during the pause phase? Any thoughts on this will be greatly appreciated.

The cameras has had no issues and was serviced earlier this year.

Thanks in advance.

Bill.

Don Palomaki · May 20th, 2010, 04:39 AM

Vertical banding is odd. Head clog problems are more likely to appear as a horizontal banding. Is the problem in both fields of the frame?

It does sounds like it could be a tape or tape load issue. I would expect the camera load processes to be timed to not allow too-soon-record-start to be an issue

Since the footage sounds unusable, you could try record over a bit of the bad area and see if the problem persists on the tape. You might wind to a bad portion and see if there is any apparent physical issue (e.g., a crease) on the tape.

Depending on the content, You might be able to sort of salvage some of the footage by using your NLE to zoom/pan in on the 70% or so of the image width that is not bad with some corresponding loss of resolution and scene edge.

Bill Masters · May 23rd, 2010, 08:42 PM

Thanks for your reply Don. I have had a look at the tape and there is no creasing. I taped over a portion of the affected footage and could not replicate the problem. I have posted a frame capture of the problem here...

Panoramio - Photo of Vertical dropout

I will be able to use some of the footage, although no more than is absolutely necessary.

Bill.

Don Palomaki · May 25th, 2010, 04:29 PM

I've not seen that sort of banding before.

It is as if the record heads in the drum lost contact with the camcorder electronics for a portion of the drum rotation. Were it a conventional motor with slip rings and brushes, it is as if the brushes broke contact with the slip ring for a portion of the scan each half rotation of the record head. No idea why.

Was there any thing interesting in the frames at the start and end of the bad stretch, such as a frame or field with a partial vertical bar? Is the bar position stable?

Bill Masters · May 26th, 2010, 08:12 AM

The bar position is stable throughout. There are some flecks popping up in the rest of the frame during the problem period but not what I'd regard as dropouts, but no doubt a part of whatever went wrong. The banding was quite stable for the duration, just a slight bit of chatter in the outer part of it. The banding is there from frame 1 and disappears completely after the pausing, nothing to indicate there was going to be a problem and nothing to indicate there was a problem afterwards.

Bill.

May 16th, 2010, 06:54 AM	#1
Bill Masters Tourist Join Date: Jul 2009 Location: Numurkah Australia Posts: 3	XL2 Vertical Banding/Dropout Hi all. Our small video production business uses 2 Canon XL2s for most of our productions. On a recent project we had a problem with one of the cameras and I am hoping that someone here might be able to explain what may have happened. The scenario is 2 cameras running simultaneously then the footage is synced and edited. Both tapes from camera 1 were perfect. Tape 1 from camera 2 was also perfect. However, tape 2 from camera 2 is all but ruined. There is a band of pixelation running vertically on the left side of the footage. The band is around 15% of the width of the frame (16:9) and is about 15% in from the left edge. So, what I have is 15% image, then 15% pixellation then the remaining 70% of the image to the right of the pixellation. The pixellation runs for the duration of the shot which is around 35 minutes. Once the camera was paused and recording recommenced the pixellation disappeared. The camera was not switched off during this time. The camera is routinely serviced by Canon, so no problems there I am fairly certain. I had run a head cleaner prior to the shoot and used a brand new Panasonic Professional tape. No warnings appeared on the screen at any time. I am wondering if, when the operator changed tapes, the cause of this issue could be that the recording process was started too hastily and the tape may not have been fully engaged in the carriage mechanism? Or could there have been some foreign matter which freed itself during the pause phase? Any thoughts on this will be greatly appreciated. The cameras has had no issues and was serviced earlier this year. Thanks in advance. Bill.

May 20th, 2010, 04:39 AM	#2
Don Palomaki Wrangler Join Date: Sep 2001 Location: Northern VA Posts: 4,487	Vertical banding is odd. Head clog problems are more likely to appear as a horizontal banding. Is the problem in both fields of the frame? It does sounds like it could be a tape or tape load issue. I would expect the camera load processes to be timed to not allow too-soon-record-start to be an issue Since the footage sounds unusable, you could try record over a bit of the bad area and see if the problem persists on the tape. You might wind to a bad portion and see if there is any apparent physical issue (e.g., a crease) on the tape. Depending on the content, You might be able to sort of salvage some of the footage by using your NLE to zoom/pan in on the 70% or so of the image width that is not bad with some corresponding loss of resolution and scene edge. __________________ dpalomaki@dspalomaki.com

May 25th, 2010, 04:29 PM	#4
Don Palomaki Wrangler Join Date: Sep 2001 Location: Northern VA Posts: 4,487	I've not seen that sort of banding before. It is as if the record heads in the drum lost contact with the camcorder electronics for a portion of the drum rotation. Were it a conventional motor with slip rings and brushes, it is as if the brushes broke contact with the slip ring for a portion of the scan each half rotation of the record head. No idea why. Was there any thing interesting in the frames at the start and end of the bad stretch, such as a frame or field with a partial vertical bar? Is the bar position stable? __________________ dpalomaki@dspalomaki.com

May 23rd, 2010, 08:42 PM	#3
Bill Masters Tourist Join Date: Jul 2009 Location: Numurkah Australia Posts: 3	Thanks for your reply Don. I have had a look at the tape and there is no creasing. I taped over a portion of the affected footage and could not replicate the problem. I have posted a frame capture of the problem here... Panoramio - Photo of Vertical dropout I will be able to use some of the footage, although no more than is absolutely necessary. Bill.

May 26th, 2010, 08:12 AM	#5
Bill Masters Tourist Join Date: Jul 2009 Location: Numurkah Australia Posts: 3	The bar position is stable throughout. There are some flecks popping up in the rest of the frame during the problem period but not what I'd regard as dropouts, but no doubt a part of whatever went wrong. The banding was quite stable for the duration, just a slight bit of chatter in the outer part of it. The banding is there from frame 1 and disappears completely after the pausing, nothing to indicate there was going to be a problem and nothing to indicate there was a problem afterwards. Bill.