1) Create the two separate single-feathered beam groups.

2) To connect them, make the first group’s ending note a “ziplet”:

1 Thirtysecond in the space of 0 Thirtyseconds

Choose “Nothing” for Number and Shape.

I hope that this is clear?

If not, ask again.

I cannot make it and the beam will merge together. And there is a  bug, when I wanna hide the number and shape (choose nothing), I cannot edit the X-let again as there is not any white square for me to click it when I wanna choose that X-let.

Here are step-by-step instructions:

Thank you very much! But finally I have this...... When I copy to another area with dynamics or other things, this problem emerges....

Did you use Stack Selection?

(See the end of my step-by-step instructions)

To make a Stack Selection of a single staff in a multi-staff document, create a Staff Set of that single staff only.

Was looking for a solution to this myself, so rather than post a new question, I found this old thread. I also discovered an easier method that does not require tuplets (aside from the ziplet) or changing the time signature. I use speedy entry. Enjoy:

Windows 10 on Mac Pro 5,1 (Boot Camp) | Finale v25.4

Now, I want to make a cadenza-like music to piano(like Liszt), this method is very trouble and it always gives this error.........
Can finale make a new plugin to solve it????

… Can finale make a new plugin to solve it???? …

A plug-in for multifeathered beaming is an excellent idea.

And you do not have to wait for makeMusic to create a plug-in.

There are several third party developers, such as Robert Patterson, Tobias Giesen, and Jari Williamson.

You could probably pay one of them to create a Multifeathered Beaming plug-in for you.

Can you introduce the plug-in to me? Thank you so much!

… Can you introduce the plug-in to me? Thank you so much! …

I do not know whether a such plug-in already exists, sorry!

Now, I want to make a cadenza-like music to piano(like Liszt)

Not for nothing, Liszt usually writes cadenzas in what looks like connected 8th grace notes, adding various expressive markings along the way. It is then up to the performer to add the rubato, accel, etc. as they desire based on interpretation. 

I also understand, though, that you may really want that control in deciding how a passage should sound, so it would make sense if you prefer not to use this method.

I also agree that a plugin would greatly simplify things.