Hi everyone, we’re trying to get the Arducam Mega 3MP and 5MP cameras working but we’re stuck on two major issues. I’ve put together a detailed report on both problems and attached it. We did get some earlier support but we still can’t move forward — these two issues are blocking the whole project. Really hoping to get them closed out.
Thanks for the detailed report. I’ve forwarded it to our engineers and asked them to look into it as a priority.
For Issue 1: the sensor on that module is a standard OV sensor, and we’ve verified it works on Windows OS. So this looks like a software compatibility problem on your side. Could you check the compatibility of the software you’re using with it?
For Issue 2: we suspect there’s something wrong in your code. We tested the 5MP camera with our demo code and it works fine — screenshots are attached. Can you try running our demo code on the 5MP and let us know how it behaves?
Thanks. Actually I don’t think I received the highlighted code or attachments you mentioned — our email gateway seems to have marked your earlier message as spam. Could you resend the highlighted attachments? Following up again just in case my first request didn’t go through either.
No problem. I’ve re-sent the highlighted demo code via private message — please check your messages and confirm you received it. If it still doesn’t come through, let me know and I’ll upload the file to Google Drive and send you a download link instead.
Quick update, plus a couple of technical questions on Issue 1.
Issue 1: The ArduCAM Mega captures images using the standard OV camera. There’s an FPGA on the camera module connected to the OV sensor, and it appears to act as a bridge — capturing the image and transferring it over the SPI interface. Looking at the OV3640 and OV5640 datasheets, there aren’t many settings to control the JPEG header, and it seems to depend on the firmware running on the FPGA. On the OV3640 specifically there’s no configuration available for this at all.
So my questions are: is it possible to control or configure the JPEG header in the FPGA or on the camera module? A Windows PC can open all these image formats easily, but our microcontroller doesn’t have the capability to render the image the way a PC does. Is there any other way to handle this in the application itself?
Issue 2: Good news — the 5MP AF camera module is working now. The OV camera wasn’t properly connected to the main FPGA board; it looks like the connector came loose during shipment. Once we reseated it, the module worked with our firmware with no modifications.
Great to hear Issue 2 is sorted — a connector coming loose during shipping would explain a lot.
For Issue 1: just to clarify how the FPGA behaves, no additional processing is done on the image data inside the FPGA. It operates in bypass mode — it only acts as a bridge to capture the image and transfer it over the SPI interface. That means we have no access to control or configure the JPEG header in the FPGA or on the camera module. The JPEG header is governed by the sensor/datasheet behavior, not by the FPGA.
Thanks for confirming. Issue 1 has been holding up our progress for a long time now, so could we get an update on whether there’s any path forward? I’ve also sent you a private message about scheduling a call so we can go through it together.
Apologies, it looks like my earlier reply may have been missed. To repeat: the FPGA runs in bypass mode and does no additional processing on the data, so the JPEG header cannot be controlled or configured from the FPGA or camera module. For Issue 2, that one is confirmed resolved — the root cause was the OV camera connector coming loose during shipment, and it now works with your firmware without any changes.