I am have a lot of troubles to solder QFN104 (10×10 mm) package. At first I tried by pre-tinning the pads, adding flux on top of it, positioning the chip and heating up to 250C on a hot plate. This sort of works however I observe erratic chip behavior. For instance the voltage at the pin is 2V but at its pullup resistor 3V. I am almost sure this is due to parasitic capacitance caused by the solder flux. So next I tried without it but now the chip behaves even worse - peripherals are dead, probably the pads don't connect with the chip pins anymore.
So my questions is how to do it right? I think a lot of people propose the stencil - I gave it a try too, but first, it is super hard to position it well, second, it is even more difficult to dose the right amount of solder paste. In my case it always ends up with more than needed causing bridges between pins.

Thank you.

 

How closely are you adhering to the recommendations for this package?

https://www.ti.com/lit/an/sloa122/sloa122.pdf

 

. For instance the voltage at the pin is 2V but at its pullup resistor 3V. I am almost sure this is due to parasitic capacitance caused by the solder flux.

Thank you.

it almost certainly isn’t. Capacitance can’t cause a shift in DC voltage. It Is probably due to a lack of solder.
I would suggest that it might be due to too much solder on the centre pad preventing the package seating on the pads

 

I believe it is due to the type of flux you use. Should use no clean flux.Proper flux and heat, and the chip will seat it self properly.

 

Agree with R!f@@
An incorrect flux will leave out a residue which causes a stubborn leakage (been there, done that).

 

It is 100% due to flux or incorrect soldering. But I blame flux because when I heat up the board, flux liquidates and chip behaves different. I have attached the flux that I currently use. Theoretically it is no clean..
https://www.amazon.com/gp/aw/d/B0BX62ZKJ2?psc=1&ref=ppx_pop_mob_b_asin_title