Other troubleshooting or questions #26
Replies: 11 comments 11 replies
-
|
hi , Hi, I have a problem after about 60s the device enters sleep mode and cannot exit this mode. The connection is rather correct because after uploading the software from link ESP works normally and comes out of sleep mode. Is there anything you can do to help me with this problem? |
Beta Was this translation helpful? Give feedback.
-
|
Note that the system will not come out of sleep mode after booting until a 'no car' situation is detected. In other words, once the system goes to sleep, it remains asleep until the car leaves all zones and then returns. As long as a car is detected, the system will always remain asleep. Rebooting or reloading the software will wake up the system as it boots, but it will return to sleep mode and will not wake again until the zones are first cleared and then entered again. If no car is present and the system does not wake up when a car enters, then you need to check the TFMini sensor and see what distances it is reporting. If the TFMini isn't reporting properly, then the application may never see a 'no car' situation and therefore will not wake up. Let me know what you find... and what troubleshooting steps you have tried. |
Beta Was this translation helpful? Give feedback.
-
|
I tried many combinations and unfortunately did not find the cause of the problem. the board after sleep does not wake up the only difference I noticed is that I have TFmini version 2.0 vs. the version described in your documnetation. |
Beta Was this translation helpful? Give feedback.
-
|
Are you able to get any reported distances from the TFMini? If you have an MQTT broker, enabling MQTT is one easy way to do this. Without knowing what distances are being reported, it is impossible to tell whether there is a problem with the TFMini or it is simply measuring something within the specified wake/active zone so it is not going to a 'no-car' situation. You might try setting your parking zones to very short values for testing... something like wake=32", active=28", park=16" and backup=14". Then point the sensor at the ceiling or somewhere where you are absolutely sure it isn't measuring some object that is within 3 feet or so. Once the system boots and goes into sleep mode, try moving your hand in front of the sensor, about 24" away, and see if the system wakes up. I cannot find anything regarding a v2 of the TFMini-s. Do note that this code has only been tested with the TFMini-s.... and likely won't work with a base TFMini or the TFMini-Luna. It hasn't been tested with the TFMini-Plus, so it may or may not work. This is covered in issue #4 (closed - release 0.40 and later supports both v1.7 and v1.8.1 of the TFMini-s... but it must be the "-s" version). If you like, you can post a link to the TFMini that you purchased and I can take a look. If there truly is a v2 of the TFMini-s, then I may need to look into making another library update in the code. |
Beta Was this translation helpful? Give feedback.
-
|
Resinchem, just wanted to chime in on this same issue. Like SenseyxX, I purchased the TFMini-s using your amazon affiliate product listing (from Blog) and received a TFMini v2_0 in July 2024. My TFMini-s unit definetly works as I have connected it to an arduino UNO and can read the serial output as expected. In addition, I have tested 3 seperate D1 Minis v4.0, also found from following your hardware listing, all have the same result. I attempted to use Arduino IDE, but the .ino just would not compile for me. Finally, I followed all of your tips in this thread including reading the HA sensors which did successfully publish to HA. The parkasst sensors only report; Car presence: Off, Park distance: 200.0 in absolutely no movement is detected in front of the TFMini-s. I have also attempted the reverse the RX/TX wires to no avail. I am happy to send you a picture of the v2_0 printed on the TFMini-s circuitboard if needed. I am strongly thinking that this new version may be the culprit as everything else works as expected. Thanks for this great project, let me know if I can do anything to help you help me. Thanks! |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for providing the feedback and all the various things that you tried. I just want to verify that you ordered the TFMini-s using the link from my blog, correct? I will order another and hopefully receive the v2.0 and not the previous v1.8. And if you get the chance to post a picture here of the back of the board with v2, it will just help me to verify that I have the exact same version as you. There aren't any posted issues or updates to the Arduino TFMIni library that I'm using, so until I can get my hands on the newer sensor, I won't be able to troubleshoot and/or make changes to my code. But once I do get the newer version of the sensor and have a chance, I will test and update my results here. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for your quick response! Attached is a picture of the link that I
clicked from your project blog that took me to the affiliate Amazon link
pictured below. I am also including an image of the back of the circuit
board as it is today.
TFMini-s link: https://amzn.to/3TzdMZe
Have a great weekend,
Bryan
…On Fri, Jul 5, 2024 at 4:31 PM Resinchem ***@***.***> wrote:
Thanks for providing the feedback and all the various things that you
tried. I just want to verify that you ordered the TFMini-s using the link
from my blog, correct? I will order another and hopefully receive the v2.0
and not the previous v1.8. And if you get the chance to post a picture here
of the back of the board with v2, it will just help me to verify that I
have the exact same version as you.
There aren't any posted issues or updates to the Arduino TFMIni library
that I'm using, so until I can get my hands on the newer sensor, I won't be
able to troubleshoot and/or make changes to my code. But once I do get the
newer version of the sensor and have a chance, I will test and update my
results here.
—
Reply to this email directly, view it on GitHub
<#26 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADWOARZJPAJFINVZIEJT4EDZK4NEFAVCNFSM6AAAAABC5CXT72VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TSNZRHEZTO>
.
You are receiving this because you commented.Message ID:
<Resinchem/ESP-Parking-Assistant/repo-discussions/26/comments/9971937@
github.com>
|
Beta Was this translation helpful? Give feedback.
-
|
Thanks. Placed order. Hopefully it will be the same version that you received. I hope to take a look at the issue next week (assuming I get the newer sensor). |
Beta Was this translation helpful? Give feedback.
-
|
Awesome, thanks for looking at this! If you don’t receive the newer sensor
let me know and I will donate mine to you!
…On Fri, Jul 5, 2024 at 7:49 PM Resinchem ***@***.***> wrote:
Thanks. Placed order. Hopefully it will be the same version that you
received. I hope to take a look at the issue next week (assuming I get the
newer sensor).
—
Reply to this email directly, view it on GitHub
<#26 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADWOAR4FA5P4KH7YNDT5MG3ZK5EJJAVCNFSM6AAAAABC5CXT72VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TSNZSGQ4TM>
.
You are receiving this because you commented.Message ID:
<Resinchem/ESP-Parking-Assistant/repo-discussions/26/comments/9972496@
github.com>
|
Beta Was this translation helpful? Give feedback.
-
|
Received new TF-Mini... and it was the newer 2.0 version. Fortunately (or I guess unfortunately depending on perspective), the new v2.0 worked just fine for me. I simply swapped out the older v1.8.1 sensor for this one in my test system and system functioned exactly the same and as expected.
I even took the new v2.0 and temporarily installed it in my production version in the garage. Again, it worked with no issues. So that means that something else is going on with your system. I'm more than happy to try to help troubleshoot, but it doesn't seem to be the newer version of the TF-Mini that is causing the issue. You have basically covered most of the troubleshooting steps that I'd initially recommend, but just to confirm/verify:
If you have Arduino/C++ experience, you could use something like an ESP32 or Arduino board that has multiple serial ports available (I don't do much with Arduinos, so not sure of different models/capabilitites). You could read the distances from one serial port that is connected to the TF-Mini and then use the second serial port to output to the serial monitor so that you could see the actual distances being measured. I can try to come up with a simply "TF-MIni Test App", but I'm not sure when I might have time to get to it. Verifying that the TF-Mini is reporting actual and expected distances would be a good start. I should also come up with a better way to indicate an issue with the TFMini either through particular LED colors/patterns or via the web application. Right now the way it works, is that if the TF-Mini is reporting a communication error (value > 50,000 at least in the prior versions), I simply output that value to MQTT and you'd see something > 200 in MQTT. If the distance is beyond 200", then it simply reports 200 to MQTT. This is to prevent weird values at the outer range of the TF-Mini (actual distances mixed in with range errors). Otherwise, it simply outputs the true measured distance via MQTT. Are you seeing the normal "boot" sequence occurring with the LEDs when you power up the system? All LEDs should briefly turn blue, followed by alternating red/green. Then if a distance is measured inside the wake distance, the LEDs would briefly show the appropriate LEDs for that range before going into sleep mode. I you aren't seeing this boot sequence, then something else is going on before it ever gets to the TF-Mini measurements. Simply for testing, I'd also try setting the wake distance to a very short value, say maybe 48" or so. Then set active, parked and backup distances appropriately between 12-48". Assure you have a completely clear field-of-view of at least 5-6 feet in front of the TF-Mini... this includes to both sides, above and below. The TF-Mini has a pretty narrow FOV, but it can still detect items off-center... especially the further away you get. Once you have that setup, try a hard power cycle of the system. After the boot sequence completes, give it another 15-30 seconds, then try to move an object somewhere within the 48" to see if there is any change in the LEDs. Let me know about all the above. I'll add the development of some sort of TF-Mini test app to my list... but again, I can't promise when I might get something like that written and tested. |
Beta Was this translation helpful? Give feedback.
-
|
OK... another follow up. I have created a very basic, quick little test for the TFMini. There's nothing fancy here... no wifi, no MQTT, etc. Unfortunately, you will need an ESP32 for this test, as two serial ports are needed... one for the TFMini and one to output data to the console. I'm using an ESP32 Mini, but any standard ESP32 Dev (WROOM-32) board should work.
I just used Dupont jumpers to connect the TFMini directly to the ESP32 Pins (from left to right when looking at front of sensor): You then connect to USB and flash the test .bin file. Restart the board and display the serial monitor. If all is working, distances should be reported once per second and should vary as you move an object closer and farther away.
If a connection is not established or invalid readings are received, the serial monitor should show an error code and dump from the TFMini. This should at least let you verify that the TFMini is working as expected and communicating using the same libary and methods as the Parking Assistant. Download and extract this .zip file, which contains the .bin to upload to the ESP32: |
Beta Was this translation helpful? Give feedback.
-
|
First, thanks so much for taking the time to help me with this! I will
follow your helpful instructions today and report back to you!
Unfortunately, I will need to order an ESP32, and will do so today! Will
make it a point to give my feedback, it’s the least I can do. Thanks again!
…On Mon, Jul 8, 2024 at 12:58 PM Resinchem ***@***.***> wrote:
OK... another follow up. I have created a very basic, quick little test
for the TFMini. There's nothing fancy here... no wifi, no MQTT, etc.
Unfortunately, you will need an ESP32 for this test, as two serial ports
are needed... one for the TFMini and one to output data to the console. I'm
using an ESP32 Mini, but any standard ESP32 Dev (WROOM-32) board should
work.
image.png (view on web)
<https://github.com/Resinchem/ESP-Parking-Assistant/assets/55962781/73a903a5-1946-4236-b866-1d61ddd4625a>
I just used Dupont jumpers to connect the TFMini directly to the ESP32
Pins (from left to right when looking at front of sensor):
Black to ESP32 ground
Red to ESP32 5V Pin
White (RX) to GPIO17 (TX)
Green (TX) to GPIO16 (RX)
You then connect to USB and flash the test .bin file. Restart the board
and display the serial monitor. If all is working, distances should be
reported once per second and should vary as you move an object closer and
farther away.
image.png (view on web)
<https://github.com/Resinchem/ESP-Parking-Assistant/assets/55962781/6330ba9a-a410-4690-9770-2775813f61c1>
If a connection is not established or invalid readings are received, the
serial monitor should show an error code and dump from the TFMini. This
should at least let you verify that the TFMini is working as expected and
communicating using the same libary and methods as the Parking Assistant.
Download and extract this .zip file, which contains the .bin to upload to
the ESP32:
tfmini_test_esp32.zip
<https://github.com/user-attachments/files/16133243/tfmini_test_esp32.zip>
—
Reply to this email directly, view it on GitHub
<#26 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADWOAR2TZEA73DGXPYIPR23ZLLOMFAVCNFSM6AAAAABC5CXT72VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TSOJQG43TG>
.
You are receiving this because you commented.Message ID:
<Resinchem/ESP-Parking-Assistant/repo-discussions/26/comments/9990773@
github.com>
|
Beta Was this translation helpful? Give feedback.
-
|
No problem. When something like this gets stuck in my head, I can't seem to let it go! When I got the TFMini last night and verified that it was v2.0, I was chomping at the bit to try it first thing this morning. I expected to have to do a lot of troubleshooting with my code. But since it worked without modification, I took that time and put it towards that little test app. Sorry that you have to order an ESP32... but I don't think you can use the USB port on an ESP8266 and connect a serial device at the same time. If it helps, here is an Amazon link to the ESP32 Mini that I used above: https://amzn.to/3XVPPkH |
Beta Was this translation helpful? Give feedback.
-
|
No problem, I’m sure that I can figure out how to use an ESP32 somewhere in
my HA environment! Appreciate the link as well! Didn’t want to piecemeal
info to you, but definitely wanted to reveal the software serial port that
allowed me to connect the Uno as described earlier:
https://lastminuteengineers.com/tfmini-s-lidar-sensor-arduino-tutorial/
This project is exactly how I tested my TFmini-s. It gives real-time
distance results; pretty neat piece of code! I’m sure that the issue I’m
facing is user error, but I feel this project is well worth the effort!
…On Mon, Jul 8, 2024 at 1:20 PM Resinchem ***@***.***> wrote:
No problem. When something like this gets stuck in my head, I can't seem
to let it go! When I got the TFMini last night and verified that it was
v2.0, I was chomping at the bit to try it first thing this morning. I
expected to have to do a lot of troubleshooting with my code. But since it
worked without modification, I took that time and put it towards that
little test app. Sorry that you have to order an ESP32... but I don't think
you can use the USB port on an ESP8266 and connect a serial device at the
same time.
If it helps, here is an Amazon link to the ESP32 Mini that I used above:
https://amzn.to/3XVPPkH
—
Reply to this email directly, view it on GitHub
<#26 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADWOARYT2EAB5QZXNGALUHDZLLRAVAVCNFSM6AAAAABC5CXT72VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TSOJQHE4TC>
.
You are receiving this because you commented.Message ID:
<Resinchem/ESP-Parking-Assistant/repo-discussions/26/comments/9990991@
github.com>
|
Beta Was this translation helpful? Give feedback.
-
|
Yep... that's very similar to what I'm doing with the ESP32. I'm using hardware instead of software serial... but still creating the two serial ports. The TFMini library I'm using is basically doing the same heavy lifting in terms of reading the serial values from the TFMini and converting to a returned distance value. By the way, do you have a Discord account? The discussion here isn't ideal for back-and-forth troubleshooting. But if you do have a Discord account and are interested, shoot me over your Discord user name and I'll reach out to you. I also have a small by-invite-only Resinchem Tech Discord server as well. If you don't want to post your Discord here, you can email it to me... and you can find my email address on my YouTube channel under the "About" section. |
Beta Was this translation helpful? Give feedback.
-
|
I don’t have a discord, but will create one and figure it out! I will email
my new discord user id!
Sent from Gmail Mobile
…On Mon, Jul 8, 2024 at 1:44 PM Resinchem ***@***.***> wrote:
Yep... that's very similar to what I'm doing with the ESP32. I'm using
hardware instead of software serial... but still creating the two serial
ports. The TFMini library I'm using is basically doing the same heavy
lifting in terms of reading the serial values from the TFMini and
converting to a returned distance value.
By the way, do you have a Discord account? The discussion here isn't ideal
for back-and-forth troubleshooting. But if you do have a Discord account
and are interested, shoot me over your Discord user name and I'll reach out
to you. I also have a small by-invite-only Resinchem Tech Discord server as
well. If you don't want to post your Discord here, you can email it to
me... and you can find my email address on my YouTube channel under the
"About" section.
—
Reply to this email directly, view it on GitHub
<#26 (reply in thread)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADWOARZPQP44CGLKKFANM53ZLLTYPAVCNFSM6AAAAABC5CXT72VHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM4TSOJRGE4DK>
.
You are receiving this because you commented.Message ID:
<Resinchem/ESP-Parking-Assistant/repo-discussions/26/comments/9991185@
github.com>
|
Beta Was this translation helpful? Give feedback.
-
|
It has been confirmed via testing that there is a problem with the newer D1 Mini v4 (with USB-C) and issue #31 has been opened. See the linked issue for details. |
Beta Was this translation helpful? Give feedback.
-
|
It's been a while since this topic had any updates so I'll understand if no one sees this. I discovered this project a month or so ago and thought it would be something I would like to try. This despite my lack of any electrical experience. I've always been on the software side of things. At first I was going to do as recommended and bench test using jumpers and breadboards but there was really only one picture in the blog article and it was confusing to me when viewing. So went all in on soldering this together. Spent a few days learning to solder and gained my confidence. Sadly, once I had finished and applied power I saw the blue smoke escape from the D1 Mini. I was hopeful that was the only damage but I needed to test the TFmin-s I had. So I had an ESP32 and downloaded your mini test bin file above. NodeMCU Pyflasher successfully flashed your file and I moved to the Arduino app to connect the serial console. After some struggles getting the port seen (I'm on a Mac) I got the port and board loaded and started the serial console. What I see is a constant stream of these errors - rst:0x10 (RTCWDT_RTC_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT) Not helpful to me. Does this mean the sensor is fried? If so, I may try again with a new board but since that was the one of the more expensive items to buy I want to be sure and verify it is dead. Here's a picture of how I had it hooked up on the breakout board.
|
Beta Was this translation helpful? Give feedback.
-
|
First, sorry you fried the D1 Mini. But don't feel bad... it happens to all of us. I could fill a barrel with all the parts I've fried over the past few years. Just part of the learning experience and in most cases, it isn't very costly. The TF-Mini is an exception, cost-wise. Unfortunately, the TFMini uses serial/UART for communication (by default). This means that on a D1 Mini (ESP8266), you cannot have serial output to the console when also using the TFMini due to only one UART bus. That's why an ESP32 was used for the testing you see above. But your test results don't necessarily mean the sensor is fried. It is just returning an invalid header. There are a lot of potential causes for this. One, the test code using the ESP32 was quickly written for a very specific scenario and not tested with other ESP32 boards. You connections look correct (from the part I can see), but one of the most common mistakes is reversing the RX/TX. You can certainly try swapping the TX2 and RX2 wires above to verify that's not the case. This won't hurt the sensor if reversed... you just won't receive a valid distance... but it's good to eliminate reversed RX/TX as a cause. Also, what version of the TFMini-s are you using? There were some initial problems with my code when v1.8.x was released (the code was originally written for v1.7) and now v2.0 is out. But I have tested the code with all three versions and it appears to work without issue. It likely won't work properly with other TFMini models, like the Plus or Luna. Assure the TFMini is receiving 5V. It's rated operating range is 4.5-6V, so it likely won't work if just given 3.3V. I'm happy to try to help out, but try/verify the few steps above and let me know the results. I'd hate to have you spend another $40+ on a new TFMini to only have the same results. It is possible that the TFMini got fried (especially if you had polarity reversed somewhere, which is what is sounds like), but let's see what we can or cannot verify first. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the reply. Just to answer your questions. It is the TFmini-s. I pretty much followed your links when purchasing the items I needed. The version is 2.0. I saw that you tested several versions and it seems they are still shipping 2.0. The ESP32 board I am using for testing the sensor is an ESP-Wroom-32 I purchased from Amazon. I know these don't work in the final product since there was an issue with them not waking from sleep. I am only using this board to verify the viability of TFmini-s. rst:0x10 (RTCWDT_RTC_RESET),boot:0x13 (SPI_FAST_FLASH_BOOT) Doing some research on the TFmini-s it seems the manufacturer has a Windows program called WINCC_TF.exe for Windows systems that can connect to the sensor when the sensor is connected to a USB-TTL serial converter. I think I may have one. I'll dig through some old boxes. Then I'll need a Windows machine or see if someone has something similar for the Mac. If not available for the Mac I can borrow on old laptop from one of my sons. They live in the area so I can run over and get one in the next day or so. Thanks for your help. I find my self drawn into things more when they don't work. I believe it's a personality disorder but it served me well back when I worked for a living. Now I use chasing problems as a way to fill time. |
Beta Was this translation helpful? Give feedback.
-
|
Well the fact that you are getting the exact same invalid header even when the RX/TX pins were swapped definitely means that the TF-Mini isn't communicating at all. The code calls the TF-Mini library to get a reading and my guess is that nothing is being returned, so there is no header info. This tends to indicate that the TFMini could be bad... but not 100% for sure. Out of curiosity, does the 'invalid header' message just continue to repeat until you stop the controller? I want to be sure that the controller isn't bootlooping. I see that RTC_RESET at the top which can indicate the board is looping. If you let the serial monitor continue to run, does the invalid message just continue to repeat or is it interspersed with other messages, like the rst: message? You may also be able to watch the ESP32. If you see the onboard LED flashing intermittently (or even somewhat rapidly), it likely means that the ESP32 is resetting. If the board is constantly resetting or bootlooping, then there is a problem with the test code on that particular ESP32 board. This may or may not be related to a faulty TFMini. Again, happy to try to help in any way that I can. |
Beta Was this translation helpful? Give feedback.
-
|
I have confirmed the sensor is dead. Thanks for the assist. I was pretty sure that was the case. |
Beta Was this translation helpful? Give feedback.
-
If you are having an issue or have a question, especially if it does not seem directly firmware related, please start by posting here and not as a new issue. If through troubleshooting, it is determined to be a firmware issue, the discussion will be moved over to an open issue.
I'm trying to maintain the issues for things directly related to bugs in the firmware. But posting troubleshooting and general questions here, they will remain visible for others to review should they have a similar issue. Thanks for your cooperation!
Beta Was this translation helpful? Give feedback.
All reactions