Perfect camera for Pyra/Pandora?


My question is:
Will it work as I've planed it? Especially the USB-Port.
I want 3 wires to be soldered to the port without a switch and only the Power line to have it.
AFAIK connected lines are simply dead wires as long as the camera does not have power and there won't be any interference with an other USB-Device plugged into the port.
I guess it should work. But I can try that for you. I have an old laptop, where I had a USB cable soldered internally to one of the USB ports to have a USB stick work as an internal poor-man's SSD. [1]
I also have a USB stick where I think I can easily block one of the power contacts with tape.


[1] https://www.amilo-forum.de/topic,24492,-USB-Stick-intern.html
 
At least there is a good chance now to get a camera into the Pyra if you want to.
Maybe I'll even offer to do the Modification for anyone if it works out well :)
(As long as you pay for it of course :p)
I got dibs!

(Still got no smartphone and not even my tablet has a camera)
 
Last edited:
News from the Camera front.

I finally had time to order the camera I want to put inside the Pyra.

Good news:

1: Image Quality is OK. It's 5 MP and about as good as my Smartphone camera. Maybe not as good but no big difference.
2: It works out of the box on my Arch Linux with GTK+ UVC Viewer
3: It's tiny and will fit into the Pyra lid (Pictures are attached. They are bad quality but it's only for size comparison)

Further steps:
1: Making it smaller
Can you see the white connector for the USB-cable (see the Amazon Link for good quality). Without it it'll fit into the Pyra lid pretty well
I'll remove it and solder the wires directly onto the PCB

2: Prepare the USB-Cable
I want the wires to be inside the Pyra. Therefore I need to remove the USB-Port and solder the wires directly to the USB 2.0 Port of the Pyra.

3: Add a switch
As I don't want to sacrifice my USB-Port I want to add a switch for the power wire and put it somewhere where it fits.

If it all works out well I'll have a camera inside the Pyra Lid that is connected via internal wires and can be enabled/disabled via a hardware switch.
The USB-Port will be usable as long as the camera is not turned on.

An alternative to the power line switch would be to place a hub between the Camera and the USB-Port.
The Advantage would be that the Camera and USB-Port can be used at the same time.
The disadvantage would be that I have to put all those wires and Cables inside the Pyra and that the Hub will drain more power.
Especially the later issue is what I don't want.

My question is:
Will it work as I've planed it? Especially the USB-Port.
I want 3 wires to be soldered to the port without a switch and only the Power line to have it.
AFAIK connected lines are simply dead wires as long as the camera does not have power and there won't be any interference with an other USB-Device plugged into the port.

If anyone wants to port GTK+ UVC Viewer (Source is in the Link above) to the Pandora I would try the camera with the Pandora.


At least there is a good chance now to get a camera into the Pyra if you want to.
Maybe I'll even offer to do the Modification for anyone if it works out well :)
(As long as you pay for it of course :p)

Real nice! Why not make it 2 cams if it's going to be hardware switched anyway, one front, one back :D
I'd consider a modification like that for sure!
 
Wow good going, @Askarus. That looks like the exact camera I was hoping to use as well. If all goes well, I might ask for help modifying my Pyra, too!
 
Hm, too expensive for me for a test.
Btw you can buy usb hubs as single chip smaller than a fingernail, check reichelt for example.
 
I just checked with my old modified laptop and now I don't think, just disconnecting one (or both) power lines will suffice.


For those who are interested in details, this is what I did:
I took two USB sticks. On one of them I first covered one, then both power pins with a strip of tape. I'll call this the "manipulated" stick. I didn't alter the other one, which I will call the "regular" stick. Attached you'll find outputs of journalctl from Debian Jessie running on the laptop.

1. connecting manipulated usb stick:
nothing happens

2. connecting regular usb stick:
Code:
Sep 27 18:03:11 server kernel: usb 5-1: new high-speed USB device number 119 using ehci-pci
Sep 27 18:03:11 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:03:11 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:03:11 server kernel: usb 5-1: new high-speed USB device number 120 using ehci-pci
Sep 27 18:03:11 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:03:12 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:03:12 server kernel: usb 5-1: new high-speed USB device number 121 using ehci-pci
Sep 27 18:03:12 server kernel: usb 5-1: device not accepting address 121, error -71
Sep 27 18:03:12 server kernel: usb 5-1: new high-speed USB device number 122 using ehci-pci
Sep 27 18:03:13 server kernel: usb 5-1: device not accepting address 122, error -71
Sep 27 18:03:13 server kernel: usb usb5-port1: unable to enumerate USB device
So obviously something is happening here when the regular stick is connected. But the stick can't be enumerated, I guess because the controller sees something funny going on on the data lines.

disconnecting manipulated:
Code:
Sep 27 18:05:26 server kernel: usb 1-1: new full-speed USB device number 125 using uhci_hcd
Sep 27 18:05:26 server kernel: usb 1-1: not running at top speed; connect to a high speed hub
Sep 27 18:05:26 server kernel: usb 1-1: New USB device found, idVendor=13fe, idProduct=1d00
Sep 27 18:05:26 server kernel: usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Sep 27 18:05:26 server kernel: usb 1-1: Product: USB DISK 2.0    
Sep 27 18:05:26 server kernel: usb 1-1: Manufacturer:         
Sep 27 18:05:26 server kernel: usb 1-1: SerialNumber: 07731D4E001E
Sep 27 18:05:26 server kernel: usb-storage 1-1:1.0: USB Mass Storage device detected
Sep 27 18:05:26 server kernel: scsi45 : usb-storage 1-1:1.0
Sep 27 18:05:26 server mtp-probe[31959]: checking bus 1, device 125: "/sys/devices/pci0000:00/0000:00:1d.0/usb1/1-1"
Sep 27 18:05:26 server mtp-probe[31959]: bus: 1, device: 125 was not an MTP device
Sep 27 18:05:27 server kernel: scsi 45:0:0:0: Direct-Access              USB DISK 2.0     PMAP PQ: 0 ANSI: 0 CCS
Sep 27 18:05:27 server kernel: sd 45:0:0:0: Attached scsi generic sg2 type 0
Sep 27 18:05:27 server kernel: sd 45:0:0:0: [sdb] 8060928 512-byte logical blocks: (4.12 GB/3.84 GiB)
Sep 27 18:05:27 server kernel: sd 45:0:0:0: [sdb] Write Protect is off
Sep 27 18:05:27 server kernel: sd 45:0:0:0: [sdb] Mode Sense: 23 00 00 00
Sep 27 18:05:27 server kernel: sd 45:0:0:0: [sdb] No Caching mode page found
Sep 27 18:05:27 server kernel: sd 45:0:0:0: [sdb] Assuming drive cache: write through
Sep 27 18:05:27 server kernel:  sdb: sdb1
Sep 27 18:05:27 server kernel: sd 45:0:0:0: [sdb] Attached SCSI removable disk
The regular stick is coming back right after the manipulated one is disconnected. This is actually more than I would have expected. I would have thought, that I'd have to re-plug the regular one too.

reconnecting manipulated:
Code:
Sep 27 18:06:39 server kernel: usb usb1-port1: disabled by hub (EMI?), re-enabling...
Sep 27 18:06:39 server kernel: usb 1-1: USB disconnect, device number 125
As expected, the hub gets confused again and decides to shut down the port.

disconnecting regular:
nothing

disconnecting manipulated:
nothing

reconnecting regular:
Code:
Sep 27 18:09:33 server kernel: usb 5-1: new high-speed USB device number 123 using ehci-pci
Sep 27 18:09:33 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:09:33 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:09:33 server kernel: usb 5-1: new high-speed USB device number 124 using ehci-pci
Sep 27 18:09:34 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:09:34 server kernel: usb 5-1: device descriptor read/64, error -71
Sep 27 18:09:34 server kernel: usb 5-1: new high-speed USB device number 125 using ehci-pci
Sep 27 18:09:34 server kernel: usb 5-1: device not accepting address 125, error -71
Sep 27 18:09:35 server kernel: usb 5-1: new high-speed USB device number 126 using ehci-pci
Sep 27 18:09:35 server kernel: usb 5-1: device not accepting address 126, error -71
Sep 27 18:09:35 server kernel: usb usb5-port1: unable to enumerate USB device
Sep 27 18:09:35 server kernel: usb 1-1: new full-speed USB device number 127 using uhci_hcd
Sep 27 18:09:36 server kernel: usb 1-1: not running at top speed; connect to a high speed hub
Sep 27 18:09:36 server kernel: usb 1-1: New USB device found, idVendor=13fe, idProduct=1d00
Sep 27 18:09:36 server kernel: usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Sep 27 18:09:36 server kernel: usb 1-1: Product: USB DISK 2.0    
Sep 27 18:09:36 server kernel: usb 1-1: Manufacturer:         
Sep 27 18:09:36 server kernel: usb 1-1: SerialNumber: 07731D4E001E
Sep 27 18:09:36 server kernel: usb-storage 1-1:1.0: USB Mass Storage device detected
Sep 27 18:09:36 server kernel: scsi46 : usb-storage 1-1:1.0
Sep 27 18:09:36 server mtp-probe[32040]: checking bus 1, device 127: "/sys/devices/pci0000:00/0000:00:1d.0/usb1/1-1"
Sep 27 18:09:36 server mtp-probe[32040]: bus: 1, device: 127 was not an MTP device
Sep 27 18:09:37 server kernel: scsi 46:0:0:0: Direct-Access              USB DISK 2.0     PMAP PQ: 0 ANSI: 0 CCS
Sep 27 18:09:37 server kernel: sd 46:0:0:0: Attached scsi generic sg2 type 0
Sep 27 18:09:37 server kernel: sd 46:0:0:0: [sdb] 8060928 512-byte logical blocks: (4.12 GB/3.84 GiB)
Sep 27 18:09:37 server kernel: sd 46:0:0:0: [sdb] Write Protect is off
Sep 27 18:09:37 server kernel: sd 46:0:0:0: [sdb] Mode Sense: 23 00 00 00
Sep 27 18:09:37 server kernel: sd 46:0:0:0: [sdb] No Caching mode page found
Sep 27 18:09:37 server kernel: sd 46:0:0:0: [sdb] Assuming drive cache: write through
Sep 27 18:09:37 server kernel:  sdb: sdb1
Sep 27 18:09:37 server kernel: sd 46:0:0:0: [sdb] Attached SCSI removable disk
So this is the normal scenario of just connecting one regular stick to a USB port.
It is only registered as "full-speed", because to reach the internal port I attached an extension cable to the port. To make sure this doesn't influence the behavior in an unfair way I attached the same type of cable to the external port. These cables seem not to be "high-speed" compatible.
The stick without the cable however is (for reference):
Code:
Sep 27 18:17:45 server kernel: usb 5-1: new high-speed USB device number 9 using ehci-pci
Sep 27 18:17:46 server kernel: usb 5-1: New USB device found, idVendor=13fe, idProduct=1d00
Sep 27 18:17:46 server kernel: usb 5-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Sep 27 18:17:46 server kernel: usb 5-1: Product: USB DISK 2.0    
Sep 27 18:17:46 server kernel: usb 5-1: Manufacturer:         
Sep 27 18:17:46 server kernel: usb 5-1: SerialNumber: 07731D4E001E
Sep 27 18:17:46 server kernel: usb-storage 5-1:1.0: USB Mass Storage device detected
Sep 27 18:17:46 server kernel: scsi48 : usb-storage 5-1:1.0
Sep 27 18:17:46 server mtp-probe[32193]: checking bus 5, device 9: "/sys/devices/pci0000:00/0000:00:1d.7/usb5/5-1"
Sep 27 18:17:46 server mtp-probe[32193]: bus: 5, device: 9 was not an MTP device
Sep 27 18:17:47 server kernel: scsi 48:0:0:0: Direct-Access              USB DISK 2.0     PMAP PQ: 0 ANSI: 0 CCS
Sep 27 18:17:47 server kernel: sd 48:0:0:0: Attached scsi generic sg2 type 0
Sep 27 18:17:47 server kernel: sd 48:0:0:0: [sdb] 8060928 512-byte logical blocks: (4.12 GB/3.84 GiB)
Sep 27 18:17:47 server kernel: sd 48:0:0:0: [sdb] Write Protect is off
Sep 27 18:17:47 server kernel: sd 48:0:0:0: [sdb] Mode Sense: 23 00 00 00
Sep 27 18:17:47 server kernel: sd 48:0:0:0: [sdb] No Caching mode page found
Sep 27 18:17:47 server kernel: sd 48:0:0:0: [sdb] Assuming drive cache: write through
Sep 27 18:17:47 server kernel:  sdb: sdb1
Sep 27 18:17:47 server kernel: sd 48:0:0:0: [sdb] Attached SCSI removable disk
 
Askarus, when you say it will fit in the lid, is that somewhere where the screen isn't ? I don't see the screen and camera fiting one on top of another.
The Pyra has a thicker lid. The LCD-PCB will be cut to save weight and the Camera PCD will fit inside that hole.
Maybe the Camera will stand out a little but that's ok I think. Maybe I'll need a cover.

Can you share also some sample pictures taken with the webcam ?
Will do so tomorrow.

Can this camera be used externally?
It can be used wherever you want.
If you want it for an other purpose you can also buy it with a case.

If you mean if It can be taken out of the Pandora I might make it so I can easily remove it and flip it 180° so it faces me for video chat.

Btw you can buy usb hubs as single chip smaller than a fingernail, check reichelt for example.
McObit did a USB Hub mod with the Pandora.
That's a solution but I don't want ti to constantly drain power and I would have to solder a lot and put a lot of cables inside the Pyra.

It's a possible solution if the one wire switch method does not work as long as I can turn off the USB-Port completely as I could do with the Pandora.
That little extra power consumption is no issue then if it only lasts as long as the USB-Port is used.

I just checked with my old modified laptop and now I don't think, just disconnecting one (or both) power lines will suffice.
Thanks a lot for testing.

I guess I'll have to go for the hub method then.
I still would prefer a solution with a switch and only one extra wire but hub is good as well as long as everything will fit inside and I can power off the USB-Ports completely.
 
Short of dimensional folding - can you describe how you're going to physically fit that into the Pyra? You said it would go in the lid - but to my understanding, the lid is already full. Is the plan to build a hump, side car or other special case piece?

Can you provide actual measurements of the device? I would like to know the board dimensions & thickness, lens diameter and length from the board if convenient.
 
I guess I'll have to go for the hub method then.
I still would prefer a solution with a switch and only one extra wire but hub is good as well as long as everything will fit inside and I can power off the USB-Ports completely.
I think it should be ok to cut both data lines and one or both power lines. To be on the safe side you'd have to make sure though, that the power lines are connected first and disconnected last.

Edit:
Just a thought:
There's a lot of empty space left of the stylus. What about routing the camera cable this way and putting a USB socket and plug there? With a little case mod this connection could be attached to a slider that's accessible from the outside.
 
Last edited by a moderator:
I think it should be ok to cut both data lines and one or both power lines. To be on the safe side you'd have to make sure though, that the power lines are connected first and disconnected last.

Then I would need a switch with 8 lines (4 in and 4 out) or better 2 with 4 lines ( 1: power in and out 2: data in and out)
Would that really better?
Unfortunately I don't think so as It's 8 wires then as well.

Here's a hub that seems sufficient.

Short of dimensional folding - can you describe how you're going to physically fit that into the Pyra? You said it would go in the lid - but to my understanding, the lid is already full. Is the plan to build a hump, side car or other special case piece?

Can you provide actual measurements of the device? I would like to know the board dimensions & thickness, lens diameter and length from the board if convenient.

It isn't full.
PCB will be between LCD and case. The CPU PCB is cut out to only a frame in order to make the LID lighter.
That's where the Camera PCB will go.
Then there is the Plastic thickness. I'll cut a square there and put the lens through it.
The lens won't stand more than a milimeter I think.

I will measure it as good as I can tomorrow.
 
Then I would need a switch with 8 lines (4 in and 4 out) or better 2 with 4 lines ( 1: power in and out 2: data in and out)
You could do all that via circuits, though. You might even find ICs that take care of that, e.g. TI's TS3USB30 takes care of switching the data lines with only a single switching signal input, if you only switch the 5V line and keep GND connected you're down to switching 2 lines.
 
Here you have some Pictures taken in my house.
They are all made under bad conditions and from the static position where my Laptop is.
My Phone is not much better.

One image shows a photo of the screen with all the options available in GUVCview.

I can clearly read the text of the book. That's important for me.

I do not want it for great photos but for a quick picture if I need one.
It's pretty sufficient for that I think.
 

Attachments

  • my_photo-10.jpg
    my_photo-10.jpg
    761.6 KB · Views: 227
  • my_photo-11.jpg
    my_photo-11.jpg
    691.2 KB · Views: 225
  • my_photo-12.jpg
    my_photo-12.jpg
    414.7 KB · Views: 221
  • my_photo-13.jpg
    my_photo-13.jpg
    359.8 KB · Views: 214
  • my_photo-14.jpg
    my_photo-14.jpg
    449 KB · Views: 226
  • my_photo-15.jpg
    my_photo-15.jpg
    704.5 KB · Views: 224
  • my_photo-16.jpg
    my_photo-16.jpg
    425.2 KB · Views: 226
  • my_photo-17.jpg
    my_photo-17.jpg
    421 KB · Views: 211
  • my_photo-18.jpg
    my_photo-18.jpg
    1.3 MB · Views: 213
  • my_photo-19.jpg
    my_photo-19.jpg
    677.5 KB · Views: 235
Thank you !
This quality is fine for scanning document or to use as a webcam, not really to take photos.
 
Here you have some Pictures taken in my house.
They are all made under bad conditions and from the static position where my Laptop is.
My Phone is not much better.

One image shows a photo of the screen with all the options available in GUVCview.

I can clearly read the text of the book. That's important for me.

I do not want it for great photos but for a quick picture if I need one.
It's pretty sufficient for that I think.
What's causing the magenta tint?
 
What's causing the magenta tint?
Looks like it could be a missing/bad IR filter.
CCD sensors also record parts of the near IR spectrum on the red channel. This creates unusually red images. To counter this, cameras have IR filters that block the frequencies the human eye can't see. Some people remove these filters intentionally for various reasons (see: "full spectrum camera").
 
Back
Top