Starter reluctant to engage flywheel on Honda GXV630 engine

I have a Cub Cadet LXT-L46 Fab with the Honda GXV630 engine and about 9 months ago I replaced the starter and ever since then more often than not, the starter will just click, click, click and nothing happens and finally it kicks in and spins the engine. Upon a close inspection I realized that the starter was hitting the flywheel tooth on tooth and not slipping of in between the teeth and starting the spinning phase and turning the flywheel. Has anyone ever had anything like this happen on your GXV630 engines. It is an aftermarket starter not OEM, I must say but I talked to the manufacturer of the starter and they say they are unaware of any complaints of that nature and say that the starter is most likely fine, since it does work as intended the second it engages the flywheel. Any thoughts on that? I was thinking sharpening the teeth a bit, but that probably wouldn't be possible, because I am sure all that stuff is hardened.

Hi Scott.
Correct me if I am wrong but I think that the solenoid as it pushes up the gear towards the flywheel starts turning the starter gear just a little bit too late. I can see the solenoid kicking up the gear and it lands tooth on tooth under the flywheel and that's it. I assume a correctly calibrated solenoid would start turning the gear at about that point if it doesn't do so already from the moment the solenoid pushes the gear up towards the flywheel. Maybe somebody is an expert on starters and can tell me if the gear starts turning a split second before it gets to the flywheel or if it is already turning when it is on the way up. Mine, when it hits tooth on tooth, it just sits there and is not turning yet. If it were to start turning at that point it would slip into gear and all would be well. When it finally does work, after how many tries, it is because it so happens that the gears just happen to line up right and the starter goes all the way up into the flywheel and spins and starts the engine. I already ordered a new one. That's the easiest way to find out. To shimmy the starter sounds like a great idea, even if mine actually bolts up from the bottom. I will tinker with that if the new one does the same thing.

Yes Scott, the initial contact surface where the starter gear tries to get into the flywheel is beveled. You confirm what I suspected, that the popping up of the gear and the gear spinning should happen at about the same time and mine doesn't. The gear pops up, gets caught under the flywheel because it is not spinning yet. I am pretty sure we have found the problem - defective starter. A new one is on the way and we should know soon. I will update. Thanks

Update!  I received a new starter and to my disappointment it does exactly what the old one does. So that means that the starters are working as designed. If 2 starters of the same model and brand doing the same thing odds are it is probably by design. The starters are not OEM but in no way can be called cheap Chinese junk. The company refers to itself as a manufacturer located in the U.S. and is selling a lot of these, I am sure, and according to them it is the right model for my GXV630 Honda engine. If few or no GXV630 owners bought any of these starters, the manufacturer may not be aware of the problem like he says. I don't think that the problem is with my engine either. The mounting surface on the block where the starter is mounted on is clean, undamaged and the other variable is the flywheel, which looks like it has never been removed or touched in any way and seems to line up where it should line up. I know many other engines probably can use this same starter and it may work flawless with them but it certainly does not work on my GXV630 engine because the spinning of the starter gear and it popping up are not simultaneous. There is a delay. Maybe the spinning starts half way up, or maybe 40 % up, I don't know but one thing is for sure. When the starter kicks in it sits under the flywheel tooth on tooth and is not spinning and therefore NO starting. Monday I did an experiment with the hopes to prove this point (Scott, great idea, you nailed it). I built a 0.10" spacer/shim which I put between starter and mounting surface, which would increase the travel of the starter gear by 1/10 of an inch before it gets to the flywheel in the hopes that it goes over the threshold where the spinning begins. And sure enough - it did! I guess now the spinning begins just 100reds of a second before the gear hits the flywheel and it slips right into gear every time. I must have started the engine now at least 20 times in the last 2 days and zero fail. Well, I don't know what's going on here and I am not going to lose anymore sleep over it. Now it works like it is supposed to and sometimes you just don't have an answer.