UserWarning:Discarded range with reserved name & Corrupted saved Workbook

Created originally on Bitbucket by ben_ceeme (Benoit)

I am using Anaconda 64x (Python v.2.7) and openpyxl==2.2.4 I open an xlsx file with openpyxl read_wb = openpyxl.load_workbook(z)

Playing with it, cells values and formula get out just fine: so far so good.

Then when I save the file I get the 'classic' discarded range with reserved name prompt (looked at the user forum of openpyxl)

UserWarning: Discarded range with reserved name
warnings.warn("Discarded range with reserved name")

Looking at the saved file, it lost some 300k in weight compared to the original.

Opening up the saved file, I get an error message that some content needs to be repaired

Excel does what it can and of course the file ends up badly messed up

Now I traced back the calculations down to a missing name range called 'tR' which is an empty single cell

Curiously enough, I call on openpyxl to get me 'tR' properties with
print read_wb.get_named_range('tR') I get:

<Worksheet "Depreciation">!$A

Uhm, no figure, no number, just !$A

So then I unzipped the .xlsx, went into \xl\workbook.xml to check it out:

Depreciation!$A1

Then the name manager in Excel tells me that tR refers to Depreciation!$A7 !!!!

So it seems that something low level overrides the instruction of workbook.xml, but despite looking at all the xml files one by one I cannot figure out what...