Bemf 0.000 v

[WMP]

I have all bemf all time 0.000 v, maybe is problem with my motors?

I use V57 rev 1.48 .

And I see that my board keep restarting, I haven't more that 15 minutes uptime

[nliaudat]

Hi,
Hard to say what's wrong cause you must provide more infos.
Is your connection with hassio good ? If hassio is enabled and no communication, the board reboots after 15 min.

board.yml

# Enable Home Assistant API
api:
  reboot_timeout: 15min

What voltage do you have between GND and CH1-2 and other at running and at endstop ?

If you want help, you must have search yourself before a bit ...

[WMP]

Can I check this voltage on motor not mounted on valve? I haven't added this board to HA yet...

[nliaudat]

If you haven't mout the motor, there will be no resistance nor endstop. The BEMF value of 0.0 is right

[nstadigs]

I guess @WMP has connected the motors to the board, but he hasn't mounted the motors to the valves.

[WMP]

[WMP]

I asked in the context of testing - my floor heating is behind the cabinet and I don't really want to pull the cabinet back to check the voltage. In addition, I use actuators in it after the return, which do work though. I know I have the same situation in the other floor heating, and it is easier to access and there is a new actuator.

In order not to have to test the board in the floor heating, I will buy another new actuator and some cheap m30x1.5 valve and check how the situation is on the 3rd board. I am using a GPIO22 together with a 4.7 kOhm resistor to which I have a DS18B20 connected - maybe this is the problem?

[nliaudat]

You can disconnect 1 actuator and set it up to your board.
You should have a reading of "moving" values and 1 endstop when the screw that acts as the plunger is at the endstop on the motor side. (you have Homematic actuators).

GPIO22 is good as it's in free pins : https://github.com/nliaudat/esp32_8ch_motor_shield/blob/main/extension.md

[nliaudat]

If you have one, check with another esp32-devkit. Some very cheap board have no buildin ADC calibration and may report incorrect values

[WMP]

Hmmm, my voltage metter show 0.000v...

1. Check if my voltage metter works:
   
   So, works. I have good connected GND and 5v.

2. Then i press Open or Close, my motor works, my valve moving but i see all time only 0.000v on ESP webpage or in my voltage metter:
   
   
   I see 0.000v on all BEMF channels.

[WMP]

Here i have max 0.105 before my motor made a clicking noise:

[nliaudat]

I suspect a bad LM324 module .... Your value of 0.105V in CH1-BEMF is good for endstop.

Can you check that the CH1-BEMF value is the same at "12In" ? (or direct in the pin if the connector is too small)

And also check "12Out" which must be 6x "12In".

Regards

[WMP]

I have https://jlcpcb.com/partdetail/Hgsemi-LM324AMTR/C2688471 or https://jlcpcb.com/partdetail/Hgsemi-LM324MTR/C194286

I think i cannot test anymore because... :

Is possible to bought this item or do you have 3d printer project for this?

So my test results can be wrong:
"12In" - from 0.012 to 0.056 before clicking noise
(this foot in LM324) - from 0.012 to
"12Out" - flapping from 0.001 to 0.003

[nliaudat]

[WMP]

This is printed on 3d?

[WMP]

Ahhh, I see... metal screw or something

[WMP]

I have this:

LM324MTR LM324 SOIC-14_L8.7-W3.9-P1.27-LS6.0-BL C194286
1.27-1*8P母 OPAMP-H1 HDR-TH_8P-P1.27-V-F C691120

[nliaudat]

If your opamp module if bad, a quick fix can be to link the BEMF inputs to the board like this :

The previous version with no opamp were like that and were working, but not with all kind of motors. You have homematic motors and I run my board in that configuration for years.

[WMP]

After i connect only 1channel bemf to CH1|2 i have on both pins 0.000:

[nliaudat]

And you can read BEMF values in the ESP32 board ....

[WMP]

Yes, I have 0.000, but maybe this is my broken motor...

[nliaudat]

Drill 1mm diam, insert a nail and cut it the right lenght.

[WMP]

I fix my motor, and i insert it onto my valve. I see how vale moving up and down. But i stil have all time 0.000v.

When i try to change firmware code to: https://github.com/nliaudat/floor-heating-controller/blob/main/sensor_adc.yaml#L61 i see all time 0.008v in ESPhome, but in voltage metter i have only 0.000v.

[nliaudat]

>100 mA and it's a cardiac arrest

https://www.firstaidpro.com.au/electric-shock-first-aid-survive-240-volts

[WMP]

Assuming that the current draw on the 5v side is maximum - 5w, or 1A, and the efficiency of the power supply is 70%, it comes out that at 230v there is 0.031 A of current.

[nstadigs]

This has nothing to do with what's on the rest of the board. L and N on the 230v side is directly connected to the outlet, so it's the same as putting your fingers in the wall outlet.

[nliaudat]

And if you touch 230V pin with your finger, your feets are the ground, what is the A drawn by your heart ?

[nliaudat]

The discussion is closed. If you don't isolate the 230V part, I won't answer any more queries. Send a photo as proof

[nliaudat]

If you have a second board, please try to cut off the resistor as told in https://github.com/nliaudat/esp32_8ch_motor_shield/blob/main/hack.md . You must cut CH1 and CH2

Have you tried another ESP32 devkit ? some are broken.

[WMP]

I have 5 free boards, I can check with other esp first... I have few without external antena, i can check this.

[WMP]

I haven't insulated the AC yet, and for now I'm not touching the board :). However, I think I have found the problem. In the documentation https://docs.espressif.com/projects/esp-idf/en/v4.4.2/esp32/api-reference/peripherals/adc.html#_CPPv425adc1_config_channel_atten14adc1_channel_t11adc_atten_t it says that usually esp32 measures voltage from 100mV, and according to my meter I have a maximum of 105mV. This is close enough to the minimum limit that I think it is not measuring this voltage for some reason. Another esp32-s2 already measures the voltage from 0mV. Do you think that using esp32-s2 can solve the problem?