Gosh, your primary reflection looks pretty close to centered under the focuser draw tube. That's good, so you're pretty close to collimated at this point. A tiny bit off (short maroon line upper left). You probably picked up that error when you tightened your center bolt pulling the secondary aft and away from the primary. Once the secondary is centered, again, fore and aft there's no need to adjust the center bolt. Loosen one tilt screw to rotate the secondary. We can fix that error. The secondary is obviously rotated as evidenced by the reflected offset pointing somewhere (green arrow) other than toward the primary (red arrow). And the secondary looks tilted toward the "bottom left" of the focuser draw tube. We can fix that, too, so the three collimation signatures (focuser draw tube, secondary mirror, and primary reflection) remain in tact.
I suspect what is happening is two of the secondary adjustment screws are tilting and rotating the secondary into this position as you are attempting to collimate the cross hair with the primary center donut. That is why your are nearly collimated, but with some residual tilt and rotation error. But, I would not use a collimation cap for this, a collimation cap is a primary axis alignment tool. Use your site tube or focuser draw tube, instead. Focus down until you're almost all the way down or the site tube looks just a little larger than the secondary mirror. I see you're using white paper background. Good. So, go back to your centered concentric circles as a starting point then try again. For now, ignore the primary mirror reflection, just get the secondary centered under the site tube (preferably) or draw tube if necessary. Use your eyepiece locking screws to hold the site tube because we're going to collimate to the site tube registration so the eyepieces will register likewise.
I can achieve this (more or less) by playing with both the center bolt of the secondary and the three small collimation screws.
Loosen all the secondary adjustment screws so they are roughly equal and in a more neutral position. let teh secondary be where it will be. If it's not centered, then you can shim the focuser slightly to center the secondary under the site tube. Once the secondary is re-centered, avoid the urge to use the tilt screws at this point. They are tilting and rotating your secondary away from its careful placement winding up with the image you show above. That's not what you want. Use only fore and aft secondary movement, secondary rotation, and any focuser shim to achieve secondary centering. Tilt is not really for secondary placement unless you compensate for it, tilt is primarily a collimation process.
So, once the secondary is centered, now pay attention to the primary reflection. Avoid the urge, for now, to begin tilt collimation using the three adjustment screws. Rotate the secondary so the reflection of the primary is centered on the secondary major axis. Once you put the primary reflection on the major axis, the primary center marker (donut) will also, and necessarily so, be on the secondary major axis, too. When the secondary is well centered under the site tube, very importantly, the secondary major axis also resides along the focuser axis. So, as a result the focuser axis, as defined by the site tube cross hairs, is in line with the primary center mark. Now, locate only one secondary adjustment screw also in line with the focuser axis. You can see all of your collimation can now be done by tilting the secondary along only the one focuser axis (secondary major axis and primary center mark are all in line with the focuser axis and one collimation screw, which is a great starting point).
So, begin tilting only the one secondary adjustment screw that moves the cross hairs along the focuser axis and secondary mirror major axis. Turn the other two orthogonal adjustment screws in unison, in the same direction, and by the same amount to allow that in line secondary tilt. This tilt (in line with the cross hairs and primary center mark) will bring the secondary reflection to the center of the (well centered) secondary mirror. If the secondary rotation is nearly perfect, then the cross hairs will also be on the primary center mark and all three collimation signatures will fall into place. Part of the problem, I believe, you are experiencing is using the two orthogonal secondary adjustment screws independently trying to collimate from three different directions. Those two screws induce some needed tilt to collimate, but unwanted tilt moving the secondary off center. They also induce some unwanted rotation of the secondary resulting as your image shows. It's intuitive to use the three secondary adjustment screws independently, but I find they are best used in unison as described. When everything is on the focuser axis (and secondary major axis) we need tilt in only one direction, the other two screws are used in unison to allow that in line tilt to happen.
Next, focus outward to the apex where the secondary mirror and the primary reflection are the same apparent size. Any residual tilt and rotation can be seen using the mirror clips (which are a proxy for the edge of the primary reflection). Any residual rotation and secondary tilt can be easily seen against the collimated primary reflection (cross hairs on the center mark). If so, loosen that single adjustment screw in line with the focuser axis so the secondary can rotate (only). Recenter the primary reflection on the secondary major axis, then tighten that single adjustment screw to bring the cross hair back into the center mark. If you miss the center mark, back off, rotate (only) the secondary, and re-tighten the adjustment screw until you nail the cross hairs on the center mark. Or, in truth, once you're this close to being collimated, you can use the two orthogonal adjustment screws to finish the cross hairs onto the center mark.
Here's another thread discussing it. https://www.cloudyni...ally/?p=9979091
I know it's a bit confusing, so here's some illustrations. This is the method I used, it's a little different but I find this to be the best way to hold the collimation signatures without excessive tilt and rotation error your image above shows. This is also an easy and quick method to check you collimation at the apex prior to observing.
Edited by Asbytec, 24 February 2020 - 11:29 AM.