The below is a false statement and reveals that your images have no zero point calibration.
The ellipsoidal variation was first noted in the visual observations in the BAAVSS dataset by John Isles over half a century ago, so eyeballs can detect it. Over the period of time you have been observing it you should have seen a one tenth to two tenth magnitude variation. Thus it is likely you will never notice the first stage of the outburst by simple visual inspection of the images until the thing has brightened half a magnitude at least when the skies are good, may up to one magnitude if they are not.
What's zeropoint? Well, you don't necessarily need to do photometry, but if you look at your images closely you will probably find that the stars in the field, the constant brightness stars, don't all look the same brightness = 'size' from image to image, and the zeropoint needed is the level of the sky background. You even see this in professional surveys if you look at only the initial fits images before the derived magnitudes are generated after photometric calibration, for no image has the same zeropiont as any other (it can be near, possibly even coincidentally the same, but not deliberately the same).
And THAT is why you consider the system's mechanics. You have failed to note a variation that exists, a borderline non-negligible variation.
A naked eye observer giving a negative result pre-outburst is as valid as your images in that context.
I am not knocking your images, you are having fun, keep going, but don't get snooty about it and assume far too much based on your images. It's only recently you've started marking T CrB's position on your images and it is still not immediately obvious from the images themselves what inversions, orientations, whatever your system provides, are, nor where North lies.
Summary : if you think the brightness of T CrB has been rock solid over your observing runs you are incorrect, it varies by very roughly two tenths of a mag every 123 ish days (all from memory, 123 ish is max to min, min to max, as the orbital period is 227 ish days. Actually, I think that may be wrong, by definition ellipsoidal variabel have two maxima and minima per orbit, due to the mechanics of it all, and thus it should really be 62 ish days for max to min, and 123 is days max to max, min to min, and 227 days for two consecutive maxima or minima).
For the moment, T Crb is still glued at its level, as on yesterday evening the 24th.