ESP8266 Community Forum

- Sun Mar 11, 2018 3:48 pm #74532 I have a project which needs to be powered from a moving vehicle so I made the assumption that I can simply use a 7805 regulator to the 5v pin of the Wemos D1 mini. I went ahead and made a circuit board based in this assumption and used a LM78M05 0.5amp regulator with a couple of caps as per the data sheet. I guessed the whole circuit would draw a couple of hundred milliamps so I should be OK.
When I powered it up it took just a couple of minutes for the regulator to reach thermal shutdown! That's weird.
This is the schematic for that board.

Wemos Seeder_schem.jpg

So I thought I must have under estimated the power consumption and so I upgraded the regulator to a TO220 L7805CV of 1.5amp max output. Again it got hot and eventually went into thermal shutdown.

So back to the breadboard I went and sure enough, with an input of 18vdc the reg would shut down after a few minutes. If I went back to 12v input it would stay cool enough to not shut down but was still hotter than I would like. The regs I used are rated for up to 35v input which I needed because the project may be fitted to 24v vehicles.

I measured the current draw of each item. The wemos was 80-85mA, the GPS was 8mA, the stepper driver was 70mA, I did not have the linkage sensor connected. I then measured the current going into the regulator at around 170mA. Well within limits of the reg.

The other odd thing is, while powered from the regulator to the 5v pin, the wemos would be active and I could log on via browser but the stepper driver would not work. By simply disconnecting the power to the 5v pin and then powering via the usb port, everything works fine. I thought this might be caused by dirty power from my power supply so I breadboarded a circuit like this minus the 9v regulator.

Wemos Seeder R3_schem.jpg

When powered via the reg to 5v pin I still had no working stepper driver.

I must be missing something. I cannot understand why the regulators are getting so hot, or why the stepper driver works on usb power but not on 5v pin power.

Any clues anyone?

You do not have the required permissions to view the files attached to this post.

- Sun Mar 11, 2018 5:48 pm #74535 Let's say that your current measurements are correct.

(12v-5v) * 0.180 = 1.25 Watts.
(18 - 5) * 0.180 = 2.34 Watts.
(28 - 5) * 0.180 = 4.14 Watts.

I assume you are using a 7805 in a TO-220 package. Did you notice the metal tab with the hole in the center? It actually has a purpose. So that you can bolt the device to a heatsink. Yes the regulator is good to 35 volts, and can supply up to 1.5 Amps. But you must take into account power dissipation.

You should be using a switching power supply. They are way more efficient so you don't have as much heat do dissipate.

You said this will be on a moving vehicle so I assume you will be getting the power for your circuit from that vehicle's charging system. Based on what you have shown I don't expect that your circuit will live too long. Vehicle electrical power is very dirty. Lots of voltage spikes and other crap. You need to protect your circuit from all that. Designing reliable electronics for vehicle applications is not simple.

http://www.st.com/content/ccc/resource/ ... 181783.pdf

http://www.ti.com/lit/an/snva681a/snva681a.pdf

- Mon Mar 12, 2018 6:12 am #74549

I assume you are using a 7805 in a TO-220 package. Did you notice the metal tab with the hole in the center? It actually has a purpose. So that you can bolt the device to a heatsink.

The original circuit included a surface mount LM78M05.
Yes I did move to a TO-220 L7805CV package later and after it also overheated, I fitted a large heat sink to continue evaluation of the circuit.

Let's say that your current measurements are correct.

(12v-5v) * 0.180 = 1.25 Watts.
(18 - 5) * 0.180 = 2.34 Watts.
(28 - 5) * 0.180 = 4.14 Watts.
Yes the regulator is good to 35 volts, and can supply up to 1.5 Amps. But you must take into account power dissipation.
You should be using a switching power supply. They are way more efficient so you don't have as much heat do dissipate.
You said this will be on a moving vehicle so I assume you will be getting the power for your circuit from that vehicle's charging system. Based on what you have shown I don't expect that your circuit will live too long. Vehicle electrical power is very dirty. Lots of voltage spikes and other crap. You need to protect your circuit from all that. Designing reliable electronics for vehicle applications is not simple.

I take your point. I did make the assumption that supplying power from a vehicle would be a relatively simple task but I am now painfully aware that is not the case.

http://www.st.com/content/ccc/resource/ ... 181783.pdf

http://www.ti.com/lit/an/snva681a/snva681a.pdf

Wow, heavy reading there. I have had a look and am beginning to understand what direction I need to move in as far as supplying power from a vehicle.

Thanks for your reply Rudy, I appreciate it.
I am still wondering why I got different results when powering the wemos on the bench from the 5v pin or the usb port though.

- Mon Mar 12, 2018 9:00 am #74553 The first thing I would do is to put a diode in series with the power supply. This should go a long way in protecting your circuit when someone connects the power backwards.

I designed a speed controller for electric motor on a salt and sand spreaders just over a year ago. One 3/4 hp motor on the conveyor and 1/3 hp on the spinner. It is for 12 volt battery systems and that meant the currents were pretty high. At the rated load the motor currents were 55 Amps and 27 Amps. Overload was higher than that. We used a variable current limit that allowed for short term overloads and motor start up. Protection of the CPU core wasn't too hard but the power section was a little more difficult (costly).