![]() |
Forum Index : Microcontroller and PC projects : Console connection?
Author | Message | ||||
davematt Regular Member ![]() Joined: 27/09/2011 Location: AustraliaPosts: 55 |
Hi all, Has this happened to anyone else? Recently I've been driven to despair by stange console behaviour. Using the excellent MMedit, I can work for some time programming and debugging code, and then things go sour. Sometimes MMedit has trouble finding the com port, followed by complete inability to get a command prompt. Rebooting the mite will give me the normal copyright message, but no prompt! The scope shows the break signal getting to the mite, it shows 2 negative going pulses of a few microsecs. Is this the normal picture? The details: Using Mmedit latest version, mmbasic 5.04.04, on a Mx170 freshly flashed. Any suggestions most gratefully received... Dave PS same behaviour in teraterm... |
||||
Andrew_G Guru ![]() Joined: 18/10/2016 Location: AustraliaPosts: 871 |
Hi Dave - I'm a rank novice and haven't seen that specific problem but have you tried re-flashing the MM hex file to the chip, or even a different chip? (Edit: I now see your comment about a freshly flashed chip but how about a different chip?) Cheers, Andrew |
||||
TassyJim![]() Guru ![]() Joined: 07/08/2011 Location: AustraliaPosts: 6283 |
If MMEdit is having trouble finding the com port, it points to USB adapter issue. Even a brief loss of the adapter causes Windows to sulk. What adapter are you using and how are you powering your 'mite. Some USB adapters don't have very good 3.3V available and that can cause grief if you are trying to run any sort of load from it. The other common problem is the capacitor on pins 19/20 I assume by 'break signal' you are referring to control-C Jim VK7JH MMedit |
||||
davematt Regular Member ![]() Joined: 27/09/2011 Location: AustraliaPosts: 55 |
Further to the above and thanks for these replies. Sorry for the delay, Grand final weekend in sunny Victoria NOT! In the middle of testing the ideas above I tried a Win 10 PC. Well I'm desperate... ![]() So far it's working fine! To me this would suggest a driver problem, but from that point I'm stumped. I've used the CP2102 converter and the Microbridge on three Win7 PCs with identical results. I don't understand how a standalone mite able to transmit the copyright message could receive the control C character, and then transmit nothing... Plan H at the moment is to buy a Win 10 laptop, so you can see things is crook! Dave. |
||||
cosmic frog Guru ![]() Joined: 09/02/2012 Location: United KingdomPosts: 302 |
I've had a lot of problems with the USB TTL serial things, from driver problems to sometimes working and then not working. In the end I started using the "old fashioned" 9 pin serial connector, this works great everytime. Now I have a rock solid, always there connection. Dave. |
||||
Azure![]() Guru ![]() Joined: 09/11/2017 Location: AustraliaPosts: 446 |
Can you elaborate a little more on the setup, is it a known MM PCB or your own config. Are you powering it completely from the CP2102? What is connected to the MM (like an LCD and any other peripherals). Do you have access to a scope to check things? |
||||
sagt3k![]() Guru ![]() Joined: 01/02/2015 Location: ItalyPosts: 313 |
Hi davematt Can you try last version of Micromite Console? link Thanks Antonio |
||||
davematt Regular Member ![]() Joined: 27/09/2011 Location: AustraliaPosts: 55 |
@Azure, I have one dil chip in a breadboard, and one SOIC chip on one of Mick's serial backpack boards. In each case, nothing else connected, external power, 10uf ceramic on pin 20, and check the ctrl C signal with a couple of different scopes. At the moment I get no response at all with any of the Win 7 adapters, while the Microbridge and a Win 10 machine work normally. At 38.4 kb, Ctrl C produces (consistently), a 25usec negative pulse, followed 50usecs later by 160usec pulse. This normally results in a burst of characters from the console Tx pin, and the arrow prompt in MMedit chat window. Communication and loading will then work. Not so if otherwise. If the above is the correct pulse waveform, why will a fresh chip not respond? Or are my pulses different from everyone elses? Hmmm... |
||||
Azure![]() Guru ![]() Joined: 09/11/2017 Location: AustraliaPosts: 446 |
Someone else will hopefully be able to chime in on the Ctrl-C response waveforms (I have not checked them before), if not I can check on Wednesday (but not before). You mention external power, what are you using for power and just to clarify - you are not using the power from your CP2102 adapter board to power the MM. |
||||
![]() |
![]() |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2025 |