I experieced what I think was severe torque roll on the maiden flight of my pilatus porter.... see the short video.
Fortunately by quickly shutting off power and correcting the roll (and a bit of luck) only minor undercarriage damage occurred which is now fixed.

Out of curiosity to try and analyse the problem I sat the port wheel on a scale and revved the motor to about half throttle... it showed a momentary force of about 100 grams (1 newton) on the scale when I did this.
The force was acting about 145mm from the centreline of the model so the rolling force was about 0.145 newton /metres which in a model weighing 1.2 kg was about 1/10 of the models mass.

Since the rolling torque occurs due to change of throttle, if the plane was held still while the throttle increased and only released at a constant RPM for takeoff it should not induce torque roll. I know that there will be propwash hitting the tail but that will yaw it but not roll it.

Is this assumption correct?

Hobbyking Pilatus PC-6 Porter EP Balsa ARF (1300mm) maiden (0 min 13 sec)

There will be torque reaction as long as the prop has to fight against drag to turn. Essentially, the motor has to push against something to get the prop to turn in the otehr direction, and that something is the fuselage. You can only trim for it at a set speed and rpm, though since the plane has dihedral it can also be partly compensated by adding side thrust to the motor.

Hi morgans
If you wait a little longer before you pull up the take off will be better. It looks to me that your plane stalled at take off. Try giving it just a little up elevator when you believe you are at take off speed. Let the plane rise on its own rather than ripping it off the ground.

I'd go with the stalled wing theory as well. That and perhaps a trifle of a cross wind takeoff.

I like to hold a little up as the throttle is first applied so the drag of the wheels doesn't try to bury the nose on the grass field that I typicaly fly from. As the speed comes up I ease off the up elevator so it comes back to darn near neutral well before the model reaches takeoff speed. In your case it seriously looks like you simply held too much up so when it lifted off the tail stayed low and the model stalled the wing. The only roll the torque would have in such a case is to aid a little in deciding which way the model is going to then roll towards. But the primary issue is that the wing was stalled.

The other option, silly though it seems, is to ensure that the ailerons are going the right way. In this age of easily switched programmable transmitters it's all too easy to just wiggle the controls and not remember to double check that they are moving the correct way. Not saying you did this for sure but it's worth checking before EVERY flight

I vote on it being a roll from aileron deflection
also I saw no rudder correction at all.
On this kind of model - getting a straight line FIRST then setup trim for take off (slight up) is always best
I saw no sign of 'torque roll There is a fair bit of torque on any of these motors but the idea is to get plane going at flying speed before increasing AOA more than absolutely necessary
The torque and or precession then becomes of little importance .
Takes practice but really not that difficult to learn.

I agree with these other posts. The airplane was pulled into the air going too slow and nowhere near straight. There appeared to be no rudder correction in the take-off roll and this correction is mandatory for a smooth take-off.

Having the rudder of ailerons reversed would also cause this. Torque roll is a factor but not at the level seen in the video.

Thanks to all who replied.
I did check the ailerons, they were operating correctly.

I hadnt considered it was just a stall, as this is a heavier model than I am used to it is probable that I didnt allow a long enough takeoff run to build up sufficient airspeed prior to lifting off.
Next flight I will try the technique that BMathews described and see how that goes.

I forgot to add in my post that the idea behind easing back to just a hair of up being held is that the model will then simply lift off when it has the proper flying speed.

I should also note that I've seen a lot of guys that horse the model off prematurely and get away with it. But that's due more to our ability to GROSSLY overpower our models compared to the more modest power to weight ratios found in full size aircraft. But while such flying style doesn't result in a crash, at least not EVERY time, those folks are getting away with it by good luck more than good flying skills. And typically the results from these premature leaps into the air seldom looks good either.

So now you know the "how" and "why". Hopefully the next video has a happier ending.