Unfortunately, the thrust you calculate from your thrust coefficient vs advance ratio curve often won’t be the exact amount your propeller was making.
One major factor is your aircraft’s configuration: if your propeller is mounted at the back, there will be some interference from the fuselage that changes the true thrust output. Your prop isn’t seeing “clean” air, it’s getting airflow that’s already been disturbed by the rest of the airplane in front of it.
For this reason, it’s important to remember what reference data your propeller models used.
You may have been lucky and gotten to do some wind tunnel testing of your full airframe and propeller, so your prop model already has those fuselage effects baked in.
More often though, the data will have come from testing the propeller alone, especially if you got this data from the prop manufacturer. You’ll need to consider if, and how, to take any fuselage effects into account.
And even more often, you won’t have a nice CT vs J data curve at all. Testing propellers in wind tunnels is expensive, and it’s just not practical for manufacturers to provide this data for the few dozen (or more) propellers they have on the shelf.
If you have prop data, it’ll probably be from testing on a static stand, with near-zero “forward” speed, or it’ll just be analytical data. You’ll have to figure out a way to make a rough propeller model from just this static data (which I have done before—let me know if you’d like me to share).
What is more time-consuming, but could arguably be the most accurate, is backing your propeller coefficients out of telemetry files from actual flights. If you trust your aerodynamic model enough, you can use that as a base to figure out your propeller’s contributions. But that explanation is way beyond the scope of this email.
It takes more effort, and more thought, to make and implement thrust and power curves based on the propeller’s advance ratio.
Remember though, it’s both RPM and forward airspeed that contribute to your performance. If you hit the easy button and just drop static data into your performance codes, you’re effectively ignoring one of the biggest factors.
No matter what data you actually have, or how much you have to eyeball and guesstimate thrust and power curves made from static data—if you do your best to have models that incorporate advance ratio, at the very least you can say your model is based in physics. And at the end of the day, that’s what we’re aiming for.