Max Vin voltage

Schematics, Example Code and Q&A

Max Vin voltage

Postby traviswatson » Sat Feb 16, 2019 11:33 pm

If I'm reading correctly, the product page says the max Vin is 12V: https://www.ezsbc.com/index.php/products/wifi01-33.html

Based on the schematic, that seems to be due to the HT7333A's 12V maximum limit. Is that where the limit comes from? I'm asking because I have an application with voltage closer to 14V, and I'm guessing I'm going to need an additional regulator.

Thanks!
traviswatson
 
Posts: 1
Joined: Wed Feb 13, 2019 4:25 pm

Re: Max Vin voltage

Postby Daniel » Thu Mar 28, 2019 12:17 pm

Yes, the regulator is the limiting device. However, for higher voltages heat dissipation will enter into the picture as well.
Daniel
 
Posts: 117
Joined: Tue Nov 13, 2012 5:10 pm

Re: Max Vin voltage

Postby gritnix » Fri May 15, 2020 3:02 pm

Hello,

Sorry to revive an old thread, but I'm wondering what your solution was. I'm thinking very strongly about hitting the board with some hot air and removing the regulator and replacing it with an MCP1703, 16V max operating input (the absolute max is 18V). I have a 15V project where I'd prefer not to have an extra component to deal with.

Edit: The cap on the Vin size sure looks like a 1206, and filtering mouser for a 47uF cap has a single hit with a voltage rating of 25V (a few at 16V, but no way am I running 15V in a 16V MLCC cap) at a little over $1 each. The voltage regulator is 54 cents, but the cap to use it is over $1 :D I may still try this and replace the cap and the regulator, but it made me chuckle.
gritnix
 
Posts: 3
Joined: Fri Apr 17, 2020 9:32 am

Re: Max Vin voltage

Postby Daniel » Fri May 15, 2020 5:07 pm

The Cap case size is 0805 and it is an expensive device, I struggle to buy them a reasonable prices. If the input voltage is higher than 7V you can safely change the cap to be 22uF 0805 25V which is much more reasonably priced.

I tried using the MCP1703A and it is not a universal solution. I found that some units will cause resets in the ESP32 to reset during load spikes. Its a combination of two factors, the MCP1703A current limit and ESP32 peak current draw. The ESP32 will draw 280mA peaks which exceeds the limit of the MCP1703A so some units will have a voltage drop during these peaks. The ESP32 typically draws 280mA peaks during some transmit operations but the duration varies depending on the application and the peak varies from device to device. With the MCP1703A you are definitely in the grey area of the specifications.

Also, when you supply the regulator with 15V and it generates 3.3V output your are dissipating (15-3.3)*0.08W = 0.93W and that is an issue. The regulator will rapidly overheat under those conditions. 80mA is the current the ESP32 draws when executing code that does not involve transmitting at 2.4GHz. The only way you can operate at 15V is if you spend the majority of the time in a sleep mode of some description.

The MCP1703A is a 'slow' LDO. When the part is supplying small output currents and there's a load step the output voltage droops very badly. This has to do with the bias arrangement of the regulator. In order to have a low standby current when the load is low the design employs some kind of circuit that limits the quiescent current when the load is low and increases it when the load current is higher. This makes a part with nice headline specifications but poor transient response. Look at Figure 2-32 in the datasheet and you will see why the ESP32 resets. The 1V drop in the output voltage is a show stopper and the overshoot exceeds the Vmax for the module. With bigger output capacitors the situation is better but never good.
Daniel
 
Posts: 117
Joined: Tue Nov 13, 2012 5:10 pm


Return to ESP32 Related

Who is online

Users browsing this forum: No registered users and 1 guest

cron