I mean that the focuser may be tilted and may need to be collimated.
The problem with this is that it can be hard to really know using just a sight tube because you can move your eye around to compensate.
The best way to see if your focuser is tilted is to measure the distance from the center of the focuser to top of the tube, then remove the secondary and shine the light across the tube and measure to see if it hits exactly the same distance from the top of the tube.
Here is how I do it. I use a sheet of writing paper and tape it inside the tube over the hole in the focuser and square to the end of the tube (make cuts for the spider vanes as necessary).
With the focuser racked in so that the tube touches the paper inside the tube, shine your laser and mark where it hits the paper.
Now, move the paper to the other side of the tube and tape it square to the end of the tube. Shine the laser on it and see if it hits the same spot. (In most dobs, one of the spider vanes is over the focuser so you can use the opposite vane to check to see that the spot is in the correct left/right position, and the paper spot is used for up/down.)
If you are careful you can simply measure this using a good ruler. This does not have to be perfect, but it should be as close as you can get it. Most focusers have collimation three collimation screw pairs in the base that will let you tilt the focuser as necessary.
Now, I am not saying you have to do this. If you find the focuser is sagging (which is most likely) then correct that and see if you get the same spot on the secondary, but if that fails then my guess is that the focuser is tilted or the tube is deflecting. The next step to determine which is the case would be to check the focuser for tilt, and if the focuser does not have any, then that would indicate that there is tube wall deflection. (I have a doubt that this is the problem, but it could be. Until you check focuser tilt and focuser tube deflection you won't know.).
It is a pain I can assure you, but the very long light path of the bioviewer, power switch and in your case, filter switch (a total light path of about 160mm, if you have tilt, tube wall deflection, or focuser tube sag against the pinion, even a vary small amount of error in any of these places is going to throw your collimation off and it will change as the orientation and load on the scope changes.
So, we have diagnosed the problem as being with one or a combination of these conditions, and now you have to make a decision as to whether you want to first eiliminate focuser tilt, or try to work around it. My advice though is to start with focuser tilt because everything else will be affected by this.
If you decide not to start with focuser tilt, then look carefully at the focuser tube for defection. If you can't access the bearings, just look at the spacing round the end of the focuser tube with and without the bionoviewer. There should be a small concentric gap between the outside wall of the focuser tube and the inside wall of the focuser housing. Check it without the BV and make sure it is even. Next, put the BV in and look at the gap and see if it changes. In particular, the gap that is on the side of the pinion shaft is the one that usually shows the error, The gap will get smaller with the BV. This is because the pinion is compressing the nylon or Teflon bearing that it rides in. It is a very powerful moment having 160mm of moment. Most focusers are not going to handle this without great difficulty, and even ones that do handle it will go through bearings much faster than with even heavy single eyepieces.
You might also be able to probe the bearing race with a long, thin piece of paper or plastic. With no BV, just push the thin material into the focuser housing at the place where the bearing tracks show on the focuser tube, and measure how far it goes in. Next, put in the BF and re-insert the thin material. If it goes in further, this means that the focuser tube has pulled away from the bearing and the pinion and front bearings are suffering the entire load.