Powering with 3.3v - wifi not working

Hi all,

I'm connecting my MKR1000 to another device which has a serial port for monitoring. The monitoring port of the device has GND, 3.3V, x and Rx. I use this power for th MKR via the Vin pin.

The sketch I wrote communicates with the device on its Serial1 pins and I communicate with the MKR over WiFi. This all works quite perfectly, except that now the arduino does not respond to http requests. It DOES still work when connected with USB and it sometimes seem to work on the device alone.

Therefore I am wondering if this can be caused by the fact that I provide only 3.3V on Vin? I read that I should actually be providing 5V. But than why did it work before/sometimes?
Is it possible that lean scripts still work on a lower voltage and heavier ones dont?

One other thing to mention is that I can see the MKR as connected in my router, with RSSI etc. So the WiFi chip does still do something.

If the power is not ok, is there any other way to use the 3.3V? (since I doubt I'll be able to get 5V from the device)

Thanks for any feedback!

Its not so much the "script" that causes issues but actual power requirements vs that supplied.

A phone charger with micro USB often supply's enough juice.

I powered mine at 3.7v on the 3.3 line at first because there is a slight voltage drop when the WiFi is actually in use caused by the TX power consumption.

If you can measure the voltage during TX you should notice a slight swing.

Not all PSU's will deliver a constant 3.3 at a rated current.

Routers will often keep connected items in thier logs for a little while so you might see it as connected when it actually isn't or is struggling to keep a connection (some routers do keep an accurate log but not all). If I am not sure if anything is connected then my router has an option to delete the connected item and if its still alive it will come right back up in the list again with the same IP

Thanks for your answer. I'll measure the voltage tomorrow to see if it drops.

But if I understand correctly, 3.3v would be sufficient and 5v is not the requirement?

Any idea on the allowed voltage range on the Vin pin?

You could look up the MKR specs for more concise info on the requirements.

Also whenever you have a voltage requirement there is ALWAYS a corresponding current requirement.
e.g. Trying to operate something needing 10V @ 1A with something that only supplies 10V @ 100mA would result in very unpredictable operation and probably a power supply failure.

Thanks.

Actually I did check the specs, this is where I read "Board Power Supply (USB/VIN) 5V", which lead me to pose my initial question. But it's still unclear why it did work than for some time on 3.3V?

Note: I also measured the voltage during operation and it is a clean 3.3V under load. Unfortuately I'm unsure what the current limit of the port is.

Knowing this, do I need to find a 5V source or is there any trick to use the 3.3V in any way?

The current minimum for LIPO is 3.7V, 700mAh with a charge rate of around half that current.

The specs are pretty clear about not exceeding 3.3 on any I/O pins.

Its also clear about using 5V if you intend to use LIPO's.

The schem is here

Its shows the VReg capable of 3.3 to 5.5

An interesting note is the ATMEL spec here states a few differing figures from 2.7 up but seems to suggest 3.6

Not seen anything for the current itself but it does seem to indicate that current requirements will change with loading on the pins (pretty normal)

Long and the short of it I would say go with 5V @ 1A (eg most phone chargers) and if you do need to go down the 3 volt range then go with the 3.6V at around 750mA to 1A