Hi all, I recently picked up an Arducam 8MP Synchronized Stereo Camera Bundle Kit (model B0195S8MP, SKU B0298R) for my Raspberry Pi 4B and I honestly can’t get it working at all. I’ve been at this for a while now and I’m starting to think the unit is faulty.
My setup:
Raspberry Pi Model 4B
Clean SD card with Raspbian OS legacy (Buster)
Camera enabled in configuration
Here’s the strange part. The very first time I run:
raspistill -t 0
I get a preview window, so it looks like the camera works. But if I run the exact same command again, the preview goes black and no picture gets taken or saved. Running:
raspistill -o image.jpg
at that point gives me:
Camera control callback cmd=0x4f525245 mmal: No data received from sensor
The only way I can recover is to reinstall a clean OS. After that it shows once and then breaks again on the next run.
vcgencmd get_camera reports supported=1 detected=1 the whole time, even though the camera clearly isn’t working.
Things I’ve already tried:
Camera enabled in configuration
vcgencmd get_camera → supported=1 detected=1
Ran raspistill -o image.jpg a second time (that’s where the sensor error shows up)
Loaded Bullseye OS and tried the libcamera apps, got similar errors
Tried the Arducam MIPI Camera Driver, no output from the camera at all
Followed the first-use instructions on the product page (with official driver)
Any idea what’s going on? Did I miss something obvious?
Hi, thanks for the very detailed report — that helps a lot.
Just so I have it straight: on a Raspberry Pi 4B with a clean Buster install, the Arducam 8MP Synchronized Stereo Camera Bundle Kit (B0195S8MP / SKU B0298R) shows a preview on the first raspistill -t 0, then the next run gives a black preview and no capture, with mmal: No data received from sensor. Meanwhile vcgencmd get_camera still returns supported=1 detected=1.
That “works exactly once, then dies” pattern is usually either a cabling / CSI port issue or a driver/config conflict rather than the sensor itself.
Could you start by working through our official troubleshooting guide for this kit?
Pay close attention to the CSI cable orientation on the HAT and the driver install steps. Let me know which steps you’ve done and what result you got, and we’ll go from there.
Thanks for the link. I went through it and I’m still stuck.
What I did from the guide:
Re-did the cable seating following the article
Tried the Arducam MIPI Camera Driver again — still no output from either camera
On Bullseye with the libcamera apps I get similar failures: it will show something once, then go black, then no data
The only consistent thing is that a clean OS gives me exactly one working preview, and then it’s dead until I re-image the card
vcgencmd get_camera still says supported=1 detected=1. Am I chasing a hardware fault at this point, or is there something in the driver/config side I’m missing?
Camera control callback cmd=0x4f525245 mmal: No data received from sensor
A few things can cause this. On this HAT there are multiple camera CSI ports. If even one of the connections on one of those ports is abnormal, the image data transfer will be abnormal, and you’ll get exactly this “no data received from sensor” error. A hardware issue is also possible.
Two things I’d like you to try:
Swap / change the cable on the extra camera. Because there are multiple CSI ports on the HAT, one bad connection is enough to break the whole thing.
Check the board’s status LED. If the board is running normally, you should see the LED flash when you power it on.
Also, could you put together a detailed video demo of your setup and the exact steps you’re doing? That way I can check the cabling and the sequence in detail instead of guessing.
Thanks Bin. I checked the LEDs on the board on power up, here’s what I see:
On power up, the red LED comes on immediately and just stays on — it doesn’t flash at all.
The green LED flashes twice and then stays off.
Not sure if that’s normal or not. I also went over the cables again after reading your reply. Does this info help you figure out where the problem is? I’m not 100% sure what I should be looking for here.
Normally, when the board is running correctly you should see the status LED flash on power-on. A red LED that comes on steady and never flashes, with the green only blinking twice then going dark, is not the usual “all good” pattern — it points at the board/connection side rather than the sensor.
Let’s narrow it down:
Reseat every CSI cable on the HAT — both the Pi-to-HAT ribbon and the HAT-to-camera cables. Make sure the contacts are facing the correct way and the connector latch is fully closed on both ends.
Test the two cameras one at a time, using the detection script from the troubleshooting article.
Re-check the LED behaviour on power up after each change, so we know whether anything shifts.
And please still send me that operation video when you can — seeing the actual cable seating and command order will help me spot anything the text description misses.
OK, spent a lot more time on this and it’s actually gone backwards.
I loaded a clean OS again, set it up, enabled the camera in settings, and ran vcgencmd get_camera. Now it says:
supported=1 detected=0
So it doesn’t even detect the camera at all now. And because it isn’t detected, I can’t check the LED status while running the raspistill command either.
I also tried the Arducam script to find out whether one of the two cameras is abnormal (from the troubleshooting article), but it just throws a Syntax error when I run it.
Cables: I checked all of them, removed all of them and put them back — I’m 100% sure they’re seated correctly — and rebooted the Pi with no success.
Thanks for sticking with it and for the screenshot. Here’s where things stand as far as I can tell:
The change from detected=1 to detected=0 after a clean OS install points at the connection path (Pi ↔ HAT ↔ cameras) rather than the sensor itself.
The No data received from sensor error combined with your LED readings (red steady, green double-flash then off) is consistent with an abnormal connection on one of the HAT’s CSI ports, or a hardware fault in the kit.
The detection script throwing a Syntax error is a separate issue — make sure you’re calling it with the right interpreter (e.g. python3 script.py) and that you copied it exactly, since a copy/paste artifact will produce that error and isn’t related to the camera.
Suggested next steps before we call it a hardware problem:
Test each camera individually on a known-good single CSI port setup if you can, to isolate whether one specific camera/cable is bad.
Re-run the detection script with the correct interpreter and share the output.
Confirm the LED behaviour again after every cable change so we can see if the pattern ever changes.
If after that the board still reports detected=0 and the LEDs behave the same, it’s likely a hardware fault rather than a configuration problem. Send over those results and the video demo and we’ll take it from there.
