No data received from camera. (on A3200 IS)

[cballenar]

I was able to install SpreadPi from the latest image ( 27-May-2015 04:52 ) and I was able to configure Spreads to use my Canon PowerShot A3200 IS.

Once I start, access Spread from browser, and try to take any pictures, I get the error "No data received from camera".

Full details:

• I tried with a USB hub and without it
• Both cameras seem to be firing, at first simultaneously, then the even one fires again.
• Display connected to RPi shows the following messages

reset high-speed USB device number 4 using dwc_otg
"usbfs: process 272 (spread) did not claim interface 0 before use

• The logs in the web interface show the following messages:

tornado.access | 500 POST /api/workflow/63eeb59f-93f4-405b-a5d7-d5458aa710bd/capture (192.168.1.90) 11192.73ms
spreadsplug.web | No data received from camera.

• I access Spreads via 192.168.1.33:5000

I'm not sure what to look for next :\ Has anyone encountered this before?

[cballenar]

After some research, I arrived to this post, I thought the problem could be the power. However I've tried again using a powered USB hub, and changing the RPi adapter to one that outputs 2A, and I'm still having the same issue.

The messages on the RPi, about reset high-speed USB device occur when the web interface starts and then when I log into it and begin the workflow. Once I'm trying to take the photos (and failing) those errors don't show up anymore. I get one for each of the cameras.

I tried, once again with and without a USB hub. What else can I try?

[duerig]

Somebody else is having trouble with their camera on the forum and I think a lot of the same can be applied to your situation. We need to figure out if the problem is on the camera or in Spreads. And if the problem is on the camera, there are a few things that are easy to get wrong that you can double check:

http://www.diybookscanner.org/forum/viewtopic.php?f=35&t=3192&start=10#p19098

I'll look more into this as well. I haven't see that reset high-speed USB device error before.

-D

[cballenar]

Hi duerig, thanks for the response.

I made a few tests with TwoCamControl and I think I may have something...

• During the first test, TCC had trouble configuring one of the cameras (zoom in/out, etc) and once in shooting mode, only one of them was working.
• During the second and subsequent tests, there weren't any problems when configuring the camera, however the problem remained when shooting. Here's their console output:

working one

con 4> rsint -cmdwait=600 -tv=1/160
rsint> path C:/MYPROJECTPATH/001L
rsint> s
rsint>

broken one

con 4> rsint -cmdwait=600 -tv=1/160
ERROR: init failed
con 5> s
ERROR: unknown command 's'

con 5>

As you can see, after trying to initialize and fail, the command switches back to con instead of rsint and thus it's not able to take any pictures. Once I close the program the working camera takes one last picture which is not saved. Then both of the shut down.

I'm not sure what to make of it. On one hand, we know that at least one of them is working. But would that be enough to break the Spreads process? I suppose that'd also explain why they were firing differently (first both simultaneously, then one again). The cameras DO have different firmware. The one working in this test is 100c, while the other one is 100a.

I'll continue researching and I'll post any other findings. I'll appreciate any other suggestions. Thanks!