DIY E-Cig 200W with regulated wattage

[Oczek]

It's almost correct. To have a better approximation you should calculate with the nominal battery voltage, which is 3.7V for our cells. So "full cells" means you have ca. 7.4V under load (until the battery gets used a while, then less). As soon as you hit the fire button your 4.2V batteries won't have 4.2V any more.
So the fully charged calculation should be with 148W, so on the firing side with 90% efficiency you'd have 133.2W left as max. power if you don't want to draw more that 20A from your batteries.

Now let's check what happens at the lowest voltage, let's say 3.2V cutoff again (on battery side):
148W/2=74W per battery
74W/3.2V = 23.125A <- maybe a problem for your component, you will need to regulate power down

That component you linked looks good, but it can only go up to 110W. Now we're almost at the 100W I suggested in my 1st post
Also keep in mind that your batteries will need to power your atmega + your voltmeter (+ your Ω-meter) and lose heat there as well, so the efficiency will go down further.

Oh, I thought they will have 4.2v under load. I think atmega, oled and stuff won’t take more than 2W. What you think about this one? Will it do the job? http://www.ti.com/lit/ds/slusci9a/slusci9a.pdf

 

[mcclintock]

It's really more a hardware challenge than a software one. I considered building my own TC mod, it would have analog knobs and no software.

 

[untar]

This looks rather complicated, don't have the time right now to check it through. It looks like it can get rather hot, so you'll need to pay attention not to put it anywhere near the batteries.
Also iirc the atmega analogue output that you'll need to control the Vin (I guess Vin is the control and Vdd the supply voltage) isn't a true analogue output but PWM, maybe you need to filter/buffer that.

 

[Oczek]

This looks rather complicated, don't have the time right now to check it through. It looks like it can get rather hot, so you'll need to pay attention not to put it anywhere near the batteries.
Also iirc the atmega analogue output that you'll need to control the Vin (I guess Vin is the control and Vdd the supply voltage) isn't a true analogue output but PWM, maybe you need to filter/buffer that.

Think I’m gonna stick with this 110W one. Could you recommend a way to measure ohms? Circuit/ic?

 

[Oczek]

It's really more a hardware challenge than a software one. I considered building my own TC mod, it would have analog knobs and no software.

I know, I won’t really have a problem with programming, just a hardware stuff

 

[untar]

Sorry, can't seem to find my bookmark to the low resistance meter atm. There are some if you search "arduino low resistance meter" but they're incomplete, you'll have to work from there (or I find my link sometime in the future).

What I can say though is that very high quality resistors with little error margin are required for this or the measurements will be completely useless.

 

[Alexander Mundy]

I would suggest the GE UDT020A0X series digital microdlynx dc dc buck converter. It is digial controlled and you can poll parameters such as input & output voltages and output current so no need for any additional hardware to read those. It doesn't meet your 200W wants but will output up to 5.5V and 20A.
Datasheet

 

[Alexander Mundy]

You can check out the dlynx-II series but I have not personally used this series as my vaping style is modest.

DLynx II* | GE Industrial Solutions

Edit: those appear to be lower voltage output than needed by looking at a couple of the datasheets.