Ki Pro F3 Problem

L

Lenilenapi

Carbonite Member

[TD="bgcolor: #f5f5ff"][/TD]

[TD="bgcolor: #f5f5ff"]

[TD="bgcolor: #f5f5ff"] Posted this last night on DVINfo.net with no responces yet so I'm going here as well.:
I just had a weird issue with a KiPro today. Input would be appreciated:
I was shooting a green screen corporate address to camera at 1080/29.97P. We double recorded to the SxS card and a Ki Pro (not mini). It was just a 10 minute shot but the tech could see the KiPro take numbers changing every minute or so at different intervals throughout the shot. The CEO was a very tempermental guy so we didn't stop especially as we knew we had SxS back-up.
Turns out the KiPro recorded about a dozen little files numbered sequentially "Clip1TK1.mov" , "Clip1TK2.mov" etc. The size varied from 300MB to about 3 Gigs randomly. When I tried to stitch the takes together in Final Cut Pro there seemed to be short gaps between the shots. In other words we were completely hosed!
Fortunately the SxS card is fine and the key will I'm sure cut fine as the guy has not a flyaway hair on the loose. However we don't know what went wrong. The Ki-Pro appeared to be set up correctly and the tech was experienced and knew the KiPro. Moreover when we got back to the shop we tried to replicate the problem and we couldn't. That's actually the scariest part.
Some other details : The camera was connected to and controlled by the standard Sony paintbox (forget the number). Time code was on Record Run The SDI for the KiPro was coming from A output on top and the SDI Out was feeding a monitor. Dual Link was set to 1.5 YPbPr422 & Video SDI Record Control was switched HD SDI Remote I/F (This turned out to be irrelevant to controlling the recorder since it was set to start from timecode anyway) The "Rec 2" note was flashing on the screen as I think its supposed to when the HD SDI Remote I/F is on.
We thought maybe the HD SDI Remote I/F was the problem but when we turned it OFF while still at the location the recorder still broke up into little files.
When we got back to the shop we could not get the reorder to do it again regardless of whether we had the SDI Record Control engaged or not and regardless of whether the cameras time code was free-run or record run and whether the KiPro was set to start from time code or not.
Any ideas? The hard drive did look a little odd - almost as if it could be pried apart very easily. [/TD]

[/TD]
 
I’ve had the same problem happen twice; yesterday I wanted to see if I could find out why. I read on another form that the problem was caused when you played back video recorded on the camera in my case the F3, the TC from the F3 somehow causes the problem through the SDI cable in the AJA KPM. Yesterday I tried to recreate the problem by continually recording clips and playing back those clips on the F3 while the KPM was connected to the F3 via SDI cable. Hours later the KPM functioned normally without problems. I then switch to battery power and tried again to recreate the problem, the KPM worked flawlessly until the Anton Bauer Dionic 90 battery read 1 hour on the power indicator and then you could see the KPM flashing from record to stop and back to record creating multiple files when there should have been 1 take. Soon after the KPM shut down but the F3 and my DP4 EVF continued to work. The F3 has a power consumption of 24w and the KPM is 18w. I’m fairly sure the problem is with voltage, I’m not sure what the minimum voltage is on the KPM is suppose to be. I purchased an Anton Bauer HCX battery which is a 120WH battery and ran my camera which is around 50W (F3=24w, KPM=18w and the DP4EVF is 5w) for over 2 hours. I also had this happen while I was on AC power, I powered down the camera and KPM and re powered up and the problem stopped, I’m thinking it could have been a brown out maybe? I would like to hear AJA’s take on this problem and hear what advice they have to eliminate the problem. The KPM is a wonderful device, I like having it mounted to the rear of the camera. Hopefully soon the version 3.0 software upgrade will come out allowing the KPM to feed via HDMI or SDI out the transport state of the unit to my EVF. It would be nice to know when shooting off the shoulder if the unit is still running. My sound person is getting tired of standing behind me watching the device.
 
scocklin said:
I’ve had the same problem happen twice; yesterday I wanted to see if I could find out why. I read on another form that the problem was caused when you played back video recorded on the camera in my case the F3, the TC from the F3 somehow causes the problem through the SDI cable in the AJA KPM. Yesterday I tried to recreate the problem by continually recording clips and playing back those clips on the F3 while the KPM was connected to the F3 via SDI cable. Hours later the KPM functioned normally without problems. I then switch to battery power and tried again to recreate the problem, the KPM worked flawlessly until the Anton Bauer Dionic 90 battery read 1 hour on the power indicator and then you could see the KPM flashing from record to stop and back to record creating multiple files when there should have been 1 take. Soon after the KPM shut down but the F3 and my DP4 EVF continued to work. The F3 has a power consumption of 24w and the KPM is 18w. I’m fairly sure the problem is with voltage, I’m not sure what the minimum voltage is on the KPM is suppose to be. I purchased an Anton Bauer HCX battery which is a 120WH battery and ran my camera which is around 50W (F3=24w, KPM=18w and the DP4EVF is 5w) for over 2 hours. I also had this happen while I was on AC power, I powered down the camera and KPM and re powered up and the problem stopped, I’m thinking it could have been a brown out maybe? I would like to hear AJA’s take on this problem and hear what advice they have to eliminate the problem. The KPM is a wonderful device, I like having it mounted to the rear of the camera. Hopefully soon the version 3.0 software upgrade will come out allowing the KPM to feed via HDMI or SDI out the transport state of the unit to my EVF. It would be nice to know when shooting off the shoulder if the unit is still running. My sound person is getting tired of standing behind me watching the device.
Click to expand...

Why ask us? Ask jim thorn from aja directly. Also whynot record manually not thru thesdi trigger? I stopped recording thru sdi feed cause its a falsesafe that im recording on both
 
eddavid; I wasn't asking about the problem I was offering an explanation to what might be causing the problem.
 
Thanks for the help but:
1- I was using the full size Ki Pro not the mini and both camera and Ki Pro were on AC throughout.
2 - You're right I didn't need to have the SDI trigger enabled and won't in the future, but since we can't replicate the problem it doesn't seem to be the problem at least so far as we can see.
3 - I did send the same note to AJA support. Can't hurt to ask people on these forums either of course. Good idea to go directly to Jon Thorn though and I just looked up his address through cml ( which was going to be my next post btw.)

We are wondering whether it could have just not been seated well enough or something having to do with the hard drive itself.

There is a line in the F3 manual on pg 127 under the section about the dual link outs where it says that changing the SDI output setting may make the sync signal for dual link disontinuous. But just what does that mean? Changing what? and how soon after changing is it discontinuous? When are you safe?
It also says "when using the HD SDI A/B connectors, use the BPU60 battery pack or supply power via the DC IN connector." What other way is there to power the camera?
 
For Lenilenapi (which I think is Leonard Levy) and All:
Sorry, but it's very difficult to find these types of posts on all of the possible forums that are out there... since, unlike dvinfo.net, there isn't a dedicated AJA sub-forum or even a "Tapeless Video Recording Solution" forum that I'm aware of here... the best I can do is try to periodically watch daily posts to dvxuser.com - not very efficient for trying to help you guys I'm afraid... your best solution is to always contact AJA Technical Support with questions or concerns; the support is offered free of charge during normal PST business hours. (Also if you posted to dvinfo.net, I didn't see it in the aforementioned forums, so I'm guessing it's somewhere else on that site... I'll have to go looking there too.)

Back to the topic: if the Ki Pro (or Ki Pro Mini) is set to Arm Recording via TC/REC KEY, it uses the incrementing timecode values coming from the camera and perceives the cameras starts and stops of timecode and then starts and stops its own recordings. From the sound of what you describe, the Ki Pro wasn't entirely stopping recording (which would rule out media issues or signal issues), but it was starting and stopping unexpectedly. This would tend to lead one to believe that the timecode coming out of the camera may have been interrupted, or inconsistent, etc. Remember that SDI timecode is part of the ancillary data in the SDI signal that is passed to the Ki Pro and so your picture might be perfectly fine, but your timecode could be compromised. The fact that you have trouble replicating the issue again when you try would tend to lead to the assumption that there is an intermittent issue with the timecode value being received; I sometimes refer to this as timecode "burping" and with tape based cameras (which I know the PMW-F3 is not), a "burp" can often occur when a tape goes from an idle state to "record ready" state (quite common on HDV cameras). So one simple test to perform is this: does the Ki Pro Mini record as expected if you don't try to use the ancillary SDI timecode data to arm the recordings? In other words, if you use the physical record button, do you see the same behavior? If you don't, this is another clue that the issue is isolated to the timecode value's consistency.

I'll post this information to dvinfo.net as well as I think found the thread there in the Sony XDCAM forum.

For sconklin (which I am guessing is Steve Conklin whom I have helped before on the dvinfo.net AJA sub-forum):
This sounds like a few things related to power: perhaps not all devices draw the amount, in watts, that are stated for them at all times. For example, some device may operate at a certain wattage (which is listed in manuals, etc.) but may peak above this in certain instances, like power up, etc. The Ki Pro Mini should not be exceeding 18 watts as an absolute maximum. The device idles around 11-12 watts, operates at 15 watts and maxes out at 18 watts. Having said that, if the values of other items - like the camera or EVF - are higher than noted at certain points, and you are already around 50 watts using the stated values, then I could see that a non-fresh battery, losing charge as it's being used, might "sag" momentarily, causing an issue. This might explain why a 90 watt battery performed differently with three devices attached to it compared to a 120 watt rated battery. If you are questioning power, it's worth running a test to see if the issue resolves itself when supplying power to each item from it's own source and then add items into the mix until you see the issue; this would tell you what wattage pushes the battery over the edge. And yes, the Super Out via SDI is coming in version 3.0 firmware so that you can see the transport state of the device and the timecode value (shown at IBC and in QA now). This should be a help for many users.

With any of these questions, I strongly encourage people to contact AJA Technical support via phone or email. I try to watch these forums, but there are so, so many of them... it's not reasonable to expect that we will catch all of your questions I'm afraid.

Hope this is helpful,
Jon Thorn
Senior Product Manager
AJA Video Systems, Inc.
 
Jon,
We have considered that some kind of time code interruption may have been happening and will go back and compare the AJA files with the SxS recording on the F3. However we have been unable to replicate the problem at any setting since that time . This is actually the most disturbing part .
 
Thanks Jon, actually I was happy to finally see the issue re-occur when the battery level of my Dionic 90 started dropping to around the 1 hour level, I will call Anton Bauer tomorrow to see what the voltage and wattage characteristics are of the battery when it reaches that point but switching over to the 120wh HCX batteries and in the near future installing the 3.0 software should resolve those issues.
 
No interruption of any kind on the SxS recording. hiccups of loss around 1/2 sec each on the Aja recording clips.
 
"No interruption of any kind on the SxS recording. hiccups of loss around 1/2 sec each on the Aja recording clips."

The internal recordings aren't relying on SDI ancillary data for their timecode value, but rather the camera's internal generator. If the SDI ancillary data is problematic, it would impact external devices not internal recordings on the camera.

Regards,
Jon Thorn
Senior Product Manager
AJA Video Systems, Inc.
 
Last edited:
Jon,
I appreciate your help, but The problem for me is that this remains an abstract theory. It makes sense, but I have No explanation of why the time code would be discontinuous nor evidence that it was.
Since I've been unable to repeat the problem I still don't really know if it was a time code problem or perhaps something else. Since similar things can happen on a mini with low power I' can't be certain something else could be operating.
It would help if you had come across something similar or an explanation of why there might be a time code problem.
I don't feel I can go back to the client and say , oh I know now what happened and how to prevent it, lets use the Ki Pro again.
Lenny
 
Lenny I requested an evaluation PMW-F3 from Sony, but have not received a unit yet. Testing with the camera is the best way to determine what may be the cause of the issue. Based on the information you provided, it certainly sounds like a possible inconsistent timecode issue. This type of issue has not been reported when using other cameras (other than the tape-based cue up mentioned previously where timecode is sometimes "burped" as the tape-based camera moves out of idle). It's worth noting that if the Ki Pro Mini is not set to arm recording via timecode start/stop then "burping" timecode from a camera would not impact recordings. I don't think, based on the information provided, it's power related. However, as noted before, it may be worth contacting AJA Technical Support to make note of your configuration; which format/frame rate and output used on the PMW-F3 along with the menu settings, the Ki Pro Mini settings used, how all items are cabled and powered, the firmware used on the camera and Ki Pro Mini, etc. Sorry, but this is the best information and suggestion I can provide until a PMW-F3 arrives for us to test. Our lab has other Sony PMW XDCAM cameras on hand and no such issue has ever been noted in extensive testing.
 
Lenny, you know what would do it? A bad BNC cable. They are notorious for getting loose over time. It would both interupt the time code and drop a couple frames.
 
Well my low voltage theory is wrong, Today I set up a test with the Sony camera, DP4 EVF and AJA KPM all running continually using a fully charged BP90 battery with a digital volt meter reading an empty port on my Anton Bauer quad port. The KPM worked flawlessly all the way down to 12.7 volts when it read low battery and shut down. The BP90 ran the entire setup for two hours; this is about right considering the system is drawing around 45W. But other than having to format the cards several times the KPM did not start rapidly starting and stopping creating multiple takes. This means something else is causing the erratic behavior. I've also checked my BNC cable, pulled and tugged on it while the camera was running.
 
Jon if I'm having the same problem as Lenny and we are using the Ki Pro and Ki Pro Mini doesn't it sound like the problem might be originating with the F3, could AJA just rent an F3 and try to recreate this problem?
 
AJA tends to work with manufacturers directly to obtain test cameras. In this case, if an evaluation Sony PMW-F3 cannot be supplied to us shortly, we will certainly rent a unit to perform testing.
 
Jon, I'm impressed that you are getting an F3 to test to resolve this. When I mentioned I would get in touch with you, one of the producer/editors who share our shop just looked up from his computer and said "AJA customer support is incredible!" and went back to work.
I did send the same message I started this thread with, to your customer support and they replied simply that it was probably a time code problem.

A bad cable is always a possibility and it was a regular BNC not a dedicated HD SDI cable. We will invest in those in the future for recording. I don't know if the fact that we were using a paintbox could possibly have any relevance. We did use the same cable again when we tested later and could not repeat it. It is a little funky seating the BNC in the A connector though , perhaps it was not all the way in?

However if the timecode theory is correct then I am drawn again to the ambiguous and disturbing note in the F3 manual on pg 127 under the section about the dual link outs where it says that changing the SDI output setting may make the sync signal for dual link disontinuous. That's a question for the Sony engineers, and what do they mean by changing the setting? when?

I would just feel better if I could get it to happen again.

Lenny
 
You must log in or register to reply here.
Back
Top