Voltage spikes caused by ESP8266

rith87
Posts: 32
Joined: Tue Mar 03, 2015 6:04 pm

Voltage spikes caused by ESP8266

Postby rith87 » Sat Jul 30, 2016 12:14 pm

Hello folks,

I have the following circuit powering my ESP8266 on my PCB (the BP1702 supports 500 mA continuous):

esp8266-supply.png
esp8266-supply.png (9.96 KiB) Viewed 2967 times


When wifi is enabled on my ESP8266, I'm hitting the following voltage spike after my LDO:

webwxgetmsgimg (4).jpg


And I'm seeing this before the LDO (I'm getting 5V using a 5V 1A adapter):

webwxgetmsgimg (5).jpg


This voltage spike does not prevent the ESP from functioning normally but it is causing some failures in other parts of my circuit. How is everyone else building their power supply on their ESP? I can use larger MLCCs but I can't fit electrolytic caps into my PCB (although not sure their high ESR will help that much anyway). What are the voltage spikes that the Espressif team is seeing with a 10uF capacitor anyway? Or does my LDO have an unusually poor transient response?

PS: I have a bug with the layout where the decoupling cap is a bit too far away from the ESP-12F. But I want to check if I need additional spots for more decoupling caps or if the 10uF is large enough.

User avatar
pratik
Posts: 416
Joined: Wed Jun 29, 2016 7:17 pm
Location: India
Contact:

Re: Voltage spikes caused by ESP8266

Postby pratik » Sun Jul 31, 2016 11:23 pm

Hello,

Reading your post, I understand that you know that decoupling caps must be as close to the module as possible. When using around 20uF regular electrolytic caps with the ESP-12F module, I personally had no noticeable issues yet. So I think a tantalum is not absolutely necessary.
Assuming you have wired ESP-12 right (I mean, you should not be putting out a logic '1' on a GPIO that is grounded, etc), the only other culprit could be the LDO.

I could not find the datasheet to it so please link it in. Please make sure the drop-out voltage is not more than 1V for the regulator. An AMS1117-3.3 will work good enough typically.
A low ESR cap (SMD caps also) will usually set up an oscillation with LDOs. Try using a regular electrolytic one if you have not tried it.
Regards,
Pratik Panda
Website: http://www.PratikPanda.com

Custom firmware, Knowledge base and freelancing (ESP8266/ESP32):
http://www.iot-bits.com

rith87
Posts: 32
Joined: Tue Mar 03, 2015 6:04 pm

Re: Voltage spikes caused by ESP8266

Postby rith87 » Mon Aug 01, 2016 2:08 pm

Hello Pratik,

Thanks for the response. I have attached the datasheet to this post.

I have verified that the dropout voltage is <500 mV for 400 mA, so I don't think the dropout voltage is the problem. I can probably try to switch over to the AMS1117 if I'm still having trouble after re-laying out the decoupling capacitor.
Attachments
BP1702A40.pdf
(468.85 KiB) Downloaded 54 times

User avatar
pratik
Posts: 416
Joined: Wed Jun 29, 2016 7:17 pm
Location: India
Contact:

Re: Voltage spikes caused by ESP8266

Postby pratik » Tue Aug 02, 2016 12:23 am

The chip looks pretty good to me. Assuming you have put the two 10uF caps close to the regulator and one 10uF close to the ESP-12F, things should work good.
On second thought, there seems to be something wrong - the voltage accuracy is claimed to be 2%. What you are getting is way off!
I'd suggest dropping in an LMS1117 for now and see how it goes. Having 1A out is not bad.

But if this still happens, it is probably more than just the regulator. Let me know what happens...
Regards,
Pratik Panda
Website: http://www.PratikPanda.com

Custom firmware, Knowledge base and freelancing (ESP8266/ESP32):
http://www.iot-bits.com

Who is online

Users browsing this forum: No registered users and 1 guest