Issue with 1 out of 10 touch buttons

WiFive
Posts: 3529
Joined: Tue Dec 01, 2015 7:35 am

Re: Issue with 1 out of 10 touch buttons

Postby WiFive » Tue Apr 03, 2018 3:23 am

Only the host. You would have to remove the uh3n or cut the traces if you want to bypass it. Then you would need to use a jumper wire on gpio0 and the reset button to enter download mode. Probably both dtr/rts should be low.

Archibald
Posts: 110
Joined: Mon Mar 05, 2018 12:44 am

Re: Issue with 1 out of 10 touch buttons

Postby Archibald » Tue Apr 03, 2018 7:51 am

Now looking more closely at datasheets, I see that RTS is an output from the CH340C USB-to-UART chip, not an input as I expected. I therefore agree with WiFive that the only option seems to be to remove the dual transistor.

Both DTR and RTS are active low. If both DTR and RTS go high when USB is disconnected, GPIO-0 will be pulled high through the base resistor and the forward-biased base/collector PN junction.

dave111
Posts: 45
Joined: Fri Mar 03, 2017 1:37 am

Re: Issue with 1 out of 10 touch buttons

Postby dave111 » Tue Apr 03, 2018 8:46 am

Things are starting to make sense! Thank you both vm for your replies.

I am still wondering...

- Why can't Android devices (as host) achieve gpio0 bypass/passthrough? My Android terminal app does not raise dtr or rts (just like minicom on Ubuntu also does not). If I raise rts on Android for test purposes, serial comms come to a halt. Maybe doing so is interpreted as "hey client, stop sending and start listening to me"? IDK. I'm asking because, if raising RTS on the host does not reach the RTS signal on the Lolin32 board, then how is the host supposed to control both switches? Raising dtr on Android for test purposes didn't seem to have any effect.

- Are ch341 and cp210x functionally identical and they exist both mainly for pricing/availability reasons? Or are there some significant technical differences between the two? Are they actually incompatible, really demanding different drivers?

Archibald
Posts: 110
Joined: Mon Mar 05, 2018 12:44 am

Re: Issue with 1 out of 10 touch buttons

Postby Archibald » Tue Apr 03, 2018 9:22 am

I wonder whether you could get another pin instead of GPIO-0 working as a touch sensor using the techniques from the Arduino forum:
https://playground.arduino.cc/Main/CapacitiveSensor
https://playground.arduino.cc/Code/ADCTouch

dave111
Posts: 45
Joined: Fri Mar 03, 2017 1:37 am

Re: Issue with 1 out of 10 touch buttons

Postby dave111 » Wed Apr 04, 2018 10:41 am

Interesting. Thank you. Let me see, if I can make this work. Or maybe I just try to live with 9 touch buttons.

Regarding the Android specific issue I mentioned above: I now consider this to be a shortcoming in the user level USB serial library I am using. It's not really a big issue, once you get used to the idea that you are limited to 9 touch buttons.

jeffmcclain
Posts: 1
Joined: Sat Jun 20, 2020 3:47 am

Re: Issue with 1 out of 10 touch buttons

Postby jeffmcclain » Sat Jun 20, 2020 3:53 am

So, I've recently been working on a touch sensor display (Heltec Lora 32 board) and using GPIO13 (Touch 4) I have been working just fine with it plugged into the USB connector. It toggles just wonderfully. But when I unplug it, no go. I need to read the values that it is returning, and I suspect it changes the levels, but one interesting thing. It works again just fine if I hang onto a ground wire in my other hand to the battery ground (or if I tie the ground pin to a case that is grounded or even have someone else hold it). Likely the issue is a ground reference issue. When it is USB powered, it isn't floating it is EARTH MAINS grounded and the touch is likely counting on that. Without a ground reference, it just moves the whole reference / board down or up relatively. But if you tie the ground plane to something that doesn't move, it holds it. So, likely you would have ground tied to the case or something the user was "holding" to have it held to a good capacitive reference...

Who is online

Users browsing this forum: No registered users and 105 guests