You are not logged in.

#1 2015-10-20 07:28:07

rallyemax
Member
Registered: 2012-11-09
Posts: 12

Found hardware solution for erratic Synaptics touchpad / clickpad

After a while of battling a very inconsistent bug where my Synaptics clickpad would sometimes behave erratically (jumping cursor, sometimes would continue to jump around after I stopped touching the clickpad), I had an epiphany today. It just got chilly around here, so I put on a fleece sweater. I turned my netbook on, the the erratic jumping started almost immediately. In the past, it has always taken some time after boot / resume for the problem to begin.

Dry Colorado air + sweater = static charges all over the place. The touchpad is a capacitive device. Static will mess with it in a bad way...if the device is poorly grounded to the rest of the netbook's hardware. In other words, it's the hardware, stupid.

A quick search, now that I was looking for an electronic and not software issue, verified my suspicions: http://www.iq-tm.de/TP%20freeze.htm

I opened my netbook, had the same setup as that in the picture (conductive foil tape being the only ground path from the touchpad module to the motherboard). I measured the resistance from a ground pad on the clickpad's PCB to the motherboard's ground, and was shocked to see a number north of 400 Ohms. Yes, full Ohms. I was shocked to have to change my range from 200 to 2000 Ohms to get a reading.

Frankly, I'm actually quite impressed that this clickpad has actually worked the majority of the time. I might have to take back all the unspeakable things I've said about Synaptics' software engineers in the past (and say them about their electrical engineers instead). It's quite impressive that the thing worked at all, when the _capacitive_ sensor's path to ground was more difficult than it would have been to go through a series of three relay coils.

The problem, then appears to be with certain Synaptics modules. The brand of the netbook is relevant as well. Mine's an Acer ES1-111 (cheapest one out there, more or less). And it's relevant because, in this case, even if Synaptics had managed to ground the module PCB to the sensor's chassis correctly, well, Acer's solution was similar to Synaptics in grounding it to the motherboard (it screws into a brass screw set fully into plastic chassis, so no ground there) using sticky conductive tape. Even cheaper than Synaptics' fancy foil. It seems that, in both cases, the conductive adhesive was the culprit, its conductivity declining over time.

I soldered the Synaptics PCB's ground pads to the Synaptic module chassis, and ran a wire from the Synaptic's mounting point to a motherboard ground (also at a mounting point). End-to-end resistance, 2-3 Ohms. No more problems with an erratic cursor. The driver is fine.

TL;DR -- if you know what you're doing with a soldering station, and if you have a laptop or netbook with an occasionally erratic / chaotic / bouncy / weird touchpad or clickpad (particularly Synaptics, but perhaps not necessarily), check resistance between the touchpad's PCB ground and the computer's motherboard ground. In my case grounding was with janky sticky tape, and I showed more 400 Ohms resistance. Fix and done.

Offline

Board footer

Powered by FluxBB