Verfasste Forenbeiträge
-
Beiträge
-
Update:
– I reviewed the reported problems on board manager 3.0.2.
– First of all I created a fix for the abs() / fabs() problem, which impacts the level crossing implementation. I did not do regression testing on level crossings yet, but I am quite confident that there are no negative side effects. The fix will be included in the upcoming firmware release 0.5.
– The firmware compiles with board manager 3.0.2 now, and I can confirm the reported WiFi connectivity problem. I didn’t find a solution yet, but Joos is also working on it and we might find a solution soon.
– For now I have placed a note in the General Section that only board manager version 2.7.4 is supported, and not version 3.0.0 and above.
– The root cause might have to do with a major WiFi library change introduced with board manager 3.0.0.Thanks for sending me your configuration. The problem here is, that you need to start counting at 1 for each different component type in Rocrail. Imagine you have 1 switch, 2 signal LEDs and 2 sensors. The components may be connected to pin D0, D1/D2 and D3/D4 respectively, but this does not play a role for the correct numbering of the Rocrail ports.
The numbering for the Rocrail ports should look like this:
Switch (pin D0): Port 1
Signal 1 (pin D1): Port 1
Signal 2 (pin D2): Port 2
Sensor 1 (pin D3): Port 1
Sensor 2 (pin D4): Port 2As you can see, the pin allocation on the microcontroller does not play a role for the numbering of the Rocrail ports.
A small lookout: with firmware 0.5, configuration is going to get somewhat easier (I hope).
Hi Joos,
hmm, good question. The configuration file looks quite good as far as I see it… in my setups, the MattzoLayoutController with the attached MCP23017 was running without problems, even with a lot of sensors.I will build and test some more MattzoLayoutControllers in the upcoming future and test it with firmware 0.5, which is to be released soon. Maybe the problem do not show with firmware 0.5. Let me know if you want to help us testing, then I will grant you early access to the firmware.
Thanks for the heads up regarding the wiring diagram.
Cheers,
MattzeHallo Stephan,
bei mir stapeln sich derzeit verschiedene Projekte. Die Boards von Kevin sind super und ich probiere sie auf jeden Fall auch noch aus. Derzeit bin ich leider noch nicht dazu gekommen.Danke für Deine Geduld.
Gruß,
MattzeJepp, it will be ESP-32 based. Please send me an e-mail if you want to have early access to the firmware.
Hi Max,
a good question and some good news for you: we will publish a new controller soon: the “MTC4BT” (MattzoTrainController for Bluetooth). It handles both LEGO Powered Up and SBricks. So the question is: yes, just may put SBricks in your train.There is a promotion on the website (www.mattzobricks.com) for MattzoBricks users to purchase SBricks with a discount. Feel free to use the published voucher code.
Cheers,
MattzeHallo allerseits,
wie oben schon geschrieben ist der springende Punkt die Ausrichtung des Magnetfelds. Dann sollte es keine Probleme mit den Reedsensoren geben. Die Richtung des Magnetfelds ist übrigens auch erheblich wichtiger als die Feldstärke des Magneten. Trotzdem kann es nicht schaden, ggf. mehrere Magnet in Reihe zu schalten. Das macht das Feld stärker und auch geometrisch größer, was in Kurven nicht ganz unwichtig ist.Gruß,
MattzeHi gramels,
thanks for your appreciation of the project.Sure, we have considered using RFID, but the use cases are pretty rare and integrating them into Rocrail would require some significant and in-depth changes to the way we are handling Rocrail messages. We might tackle this in the future, but presently we do not have plans to support RFId.
Cheers,
MattzeSomething like that. Check https://mattzobricks.com/automation/controllers/general, section “MQTT”.
Hallo Hans-Joachim und alle,
zunächst einmal großen Dank, dass Du unsere Technologie für Deine Anlagensteuerung adoptierst und sogar mit Deinen eigenen Signalen anreicherst. Typen wie Du sind der Grund, wieso wir das hier machen. Schön, dass Du dabei bist.Zu Signalen haben wir einiges in der Pipeline – das nächste Update der MattzoLayoutController Firmware wird auch Formsignale unterstützen, und überdies auch Signalbegriffe, die sich über mehr als eine LED erstrecken (z.B. Langsamfahrt bei einigen DB-Signalen, gelb und grün leuchten gleichzeitig). Das wird mich noch einiges an Arbeit kosten, aber das Ergebnis einschließlich Video wird hoffentlich viele Begeisterte finden. Ein bisschen Geduld müsst Ihr aber noch haben.
Danke für Euer Engagement.
Gruß,
MattzeHi Edwin,
can you send me your MattzoLayoutController configuration file?Besides this, keep up the good work with your layout!
Cheers,
MattzeHi folks,
I have read your comments above. I am pretty busy at the moment, but I have just created a bug report in our internal bug tracking tool. I will check out if I can reproduce the problems with the new ESP-8266 firmware (V3.x.x). I will also have a look at the proposed abs() / fabs() work-around.Thanks for your esteem.
Cheers,
MattzeHi Joos,
thanks for your comment and your interest in our technology. I am happy to hear that you are going to build a layout with MattzoControllers.To your question: the issue is that the MattzoLayoutController has been developed for an ESP-8266, while the MTC4PU is designed for the ESP32 board. It would be possible to integrate both firmwares into one and base it all on an ESP-32, but this is complicated, takes a lot of time and testing. This is why we do not have plans to allocate our time on this matter at the moment.
So, my recommendation is for buy a couple of ESP-8266s and one or two ESP-32 for a start. It is useful not to centralize the layout control too much. When your layout grows, you should rather allocate a couple of MattzoLayoutControllers on different parts on your layout. Especially switches do not react too well when the cable lengths become too long.
I wish you good success with your automation. It’s big fun!
And stay tuned for the latest MattzoController which we are (hopefully) going to release very soon. It will be called MattzoController for Bluetooth (MTC4BT), and controls both LEGO Powered Up devices and SBricks.
Cheers,
MattzeHi ROb,
thanks. This part of the wiring is not required at the moment as we are not using the voltage information on A0 of the microcontroller yet. Will will do this in a future release of the MTC4PF. So feel free to just leave that part out for now.Cheers,
MattzeI added download links for older firmware versions on the firmware page.
Ich lade die alten Versionen demnächst mal hoch und baue ein Archiv auf. Meine Behauptung wäre jedoch, dass die 0.4.1 besser ist als die 0.3.
Was mir noch eingefallen ist: als Ihr das Update heruntergeladen habt, habt Ihr da auch die Datei MattzoController_Library.h im Arduino-Library Verzeichnis aktualisiert? Da waren auch Änderungen drin.
Reset button should do, powering the unit off for 20+ seconds might help in severe crash situations.
Hmm, good question. Sure you try to connect to a 2.4 GHz network? ESP32s can not connect to 5 GHz networks.
Hi Chris,
I have my roots in the golden 12V age as well.We have experienced a couple of problems on ESP32 boards when trying to connect to connect to WiFi and Bluetooth concurrently. Usually it helps to reboot the controller a couple of times.
Please make sure you are using Firmware 0.4 or above – we have improved the stability of the connection procedere with 0.4. WiFi connectivity should be working now without a problem. The controller will now establish the WiFi connection at first and only thereafter start to connect to Bluetooth devices. This seems to work more reliably on most boards.
We are actively working on improving to control bluetooth devices. The next version of the MTC4PU will likely also connect to SBricks and is (hopefully) going to work significantly better in terms of reliability and stability.
Cheers,
Mattze