Because I can!
For a project I need the VGA signal my old tablet computer can put out.
Thanks to the idiots at ViewSonic I need a special, proprietary adapter cable for that. This in itself wouldn’t be too big a problem, if the tablet wasn’t ten years old. Plus, ViewSonic discontinued the device shortly after I’d bought one.
Also, the necessary adapter cable wasn’t even too available while the tablet was still sold.
So the idea was to remove the original connector, figure out the wiring and hook it to a regular VGA connector. Easy-peasy.
Well … somewhat.
Step 1 – Get Rid of the Culprit
The connector might look a bit like a Micro USB or a Mini-HDMI or Mini-DisplayPort. But it’s nothing of the sort. It is in fact proprietary. Despite evil lead free solder, the connector came out quite easily. Without damaging the traces.
In the right image note the two MOSFETs Q8 and Q7. You can easily trace out their connections to the VGA port. That helped me a lot in determining what the signals were. From the way they were connected it seemed to be a I2C bus. Which is used these days by the display driver to talk to the monitor.
Unfortunately, I couldn’t just put the scope on the pins and look what’s going on. Because usually – and sure enought in my case too – the graphics card deactivates the output drivers for the port if no device is connected. I’m not sure exactly how the detection is done, but it’s not too hard to imagine that the driver (the driver chip on the main board) recognizes the 75 ohms of input impedance of the monitor.
The other problem was that I had to mount the board upside down into the case in order for the CPU to be in contact with the heat sink whenever I was testing something. This meant I couldn’t reach the pins at all.
So, I would have had to solder wires to all of the pins BEFORE I knew what the signals were. I was not in the mood for that. Soldering on 12 tiny wires and then just using 8? Not gonna happen!
Step 2 – Wire Breakout
After a little bit of investigation I could exclude a couple of ground pins and I soldered one wire to a resistor next to the connector. This gave me enought room to get all necessary signals connected.
For the first signal testing I taped a strip of PCB with short traces to the back of the tablet PC as a solder base. The strip is actually the PCI connector from an old PC card. Who knew that this would be useful someday !?
Step 3 – First Contact
To start of I connected what I assumed to be the red, green and blue signals. After connecting that to an actual VGA display I could in fact measure a signal on each of the color channels.
After that it was easy to figure out what the other signals were. The sync signals are easily recognizable on an oscilloscope screen. The I2C was a bit more complicated because I had to disconnect and then reconnect the monitor to get the controller to send data on the I2C bus.
Step 4 – Victory
By the way, there are people that argue, my workspace would be a bit crowded.
Step 5 – Making it Pretty
Since I have no intention to use that tablet as a tablet computer anymore, and VGA plugs are quite bulky, I decided to slab on an ugly box that holds an ordinary 15 pin VGA connector.
For stuff like this I use raw circuit board material. It’s really easy to handle and – as you can see – can be fixed together with simple solder joints.
Since I was making the project up as I went along, I had no good plan how to close off the housing. Finally I quickly designed a little cover and put it through the 3D printer. As if this project wasn’t crazy enough.
I had to print it two times. The first one was a hair too small and one of the braces hit the “nut” of the VGA connector.
But that’s the beauty of 3D printing: Too small? Print it again!
Especially such a small part. That took just 20 minutes to print.
Step 6 – Be happy, go to sleep
Thanks for reading.