GeneS

Since we last typed, I have conducted tests with the recommended Neewer intervelometer. Using the fastest setting of 1 second photo intervals, the actual frame rate is about 1 photo every .96 seconds or so. After over 25k images shot on the same card without formatting in the camera, I have had no image dropout. I have let the camera shoot for 45 mins on several runs, as well as for about 85 mins, which completely filled my 64G card shooting JPEG only, extra fine. It seems that the formatting step was not the problem. This leads me to believe there was/is some error in the flight control command that issues the camera trigger function. Not really any closer to a solution, but at least I know that the camera can write at least 1 image a second pretty much forever. Thanks again for all the comments and suggestions, I'll keep you posted on further updates.

Yes, i do the same, copy the entire parent folder. Once it has copied i delete it in the computer interface. Should I just skit this step and reformat the card once it is back in the camera to clean the memory?

I like the controller based time lapse test. Definitely worth the cost. You said, "If you have been taking images off the card and then deleting them in the folder of the SD card then that might well cause it to lose sync as the database thinks they are still there." When you say deleting them in the folder, do you mean in Windows or when using the camera as interface? Also, do you recommend taking the SD card out and copying with a card reader and computer, or tethering the camera to a computer and ripping images via USB and not taking out the SD card? I now know that the camera should be powered off before removing SD and that it should be reformatted before the next shoot. Thank you for answering my basic questions.

Thanks BTM_Pix. I did not reformat the SD card, only copied and deleted data. Also, I didn't check if the memory held the amount of data expected for the full number of images or not. That is a good idea though. Will reformatting reinstall the database system and wipe the corruption from the SD card if it is there? I may be able to create a ground mission. It is tricky to arm and execute this however, as the drone senses where it is and compensates based on the gps. It will obviously not be moving, so I don't think it will work. It wont spin up and arm if all system checks don't pass, and I don't think they would. Might be possible with the arms and props removed, not sure though. Is there any sort of log of commands and messages that can be downloaded from the camera after use for diagnostics and troubleshooting? Like a black box recording for the camera? I do think that the IBIS may be a possible culprit, both for the capture/write issues as well as for some image processing issues after the fact. The photogrammetry software I am using back calculates the internal lens geometry for each frame, and when that geometry changes slightly as a result of steady shot movements, it creates some trouble for the point matching algorithm that stitches together all the frames and generates a 3d model based on this mosaic. Good to know that 1 frame every 2 seconds shouldn't be a problem. Has anyone done short interval time lapse for long periods of time that would mimic my setup. Again, thinking 1 frame per second for about 45 mins to replicate. Has anyone ran into a functional limit on photo interval past 30-40 frames or so when the memory cache runs out? Thanks for the input and keep it coming!! GS

Hello and thank you for the reply. To hopefully clarify and give you a better picture of what is going on: The camera trigger command is given through the gimbal which is attached by both HDMI and micro USB. The camera is also powered from the drone through the gimbal. Trigger commands are generated by a mission planning software that calculates trigger interval based on desired overlap, groundspeed, image footprint size, etc. A mission plan file is generated, then loaded into memory on the drone. It then executes this flight plan, including camera trigger locations based on either gps for position based camera trigger or intervalometer for timing based trigger. Our ground station controller (how battery, telemetry, satellite signal strength and all other drone info is monitored) displays when a camera trigger command is given, and this is also recorded in the drone flight log file. The trigger commands are approx 2 seconds apart. So, 1 photo every 2 seconds for about 45 mins. I am shooting in JPEG, as I don't need RAW for mapping. My problem is with inconsistency of image capture. For example when flying an approx 160 acre site that required multiple flights to cover; the first flight of approx 45 mins captured 1394 images based on 1401 camera trigger commands. So, 99.5 of the camera trigger commands yielded a photo on the SD card. After changing batteries and putting the drone back up to finish the second half, the drone returned with only 271 images out of a similar number of camera trigger commands. All drone and camera settings were identical. Manual focus set to infinity on our Sony Sonnar T* FE 35mm f/2.8 ZA Lens, ISO 250, f/8, 1/1000sec shutter. Also copied and cleared all existing images from previous flight from the SD card, so full memory capacity. I'm using SanDisk Extreme PLUS 150MB/s, XC1, V30, U3 Class 10 SD card. If the problem lies in the buffer or memory card, why was it able to successfully write 99.5% of the images on the first flight? I've had this issue in other locations with flights of varying times as well, so it isn't limited to long flights and 1000+ image attempts. Works as expected sometimes and not at others. All other variables seem to be the same. I appreciate your thoughts on the issue. The community of those knowledgeable about higher end cameras as it relates to drone applications is definitely on the smaller side. Thanks for your time. GS

I know this thread is old, but I hope it still has some traction. My question is a bit different and has to do with rapid image capture over a long flight. I have a large hexacopter capable of flying with the A7Rii for about 40 mins. The problem is, I am often losing images. What i mean is, a camera trigger command is given, but the image is not written to the SD card. I have had success capturing with full resolution on the A7Rii for about 1300 frames @ 1 frame every 1.5 seconds, but other times I am only getting between 20-25% of the images I am expecting. I have a V30 150MB/s SD card, class 10 U3, so pretty fast, but I think this may be the problem. I am wondering if anyone else has run into a trigger interval limit, where any faster the images don't capture. Thanks for the input. I am using the unit for high resolution mapping and 3D terrain models. Thanks in advance for the input.