ZIP Function - Need Additional Controls To Mitigate Occasional Looping

Question

I have an extremely large volume of files that need to be individually ZIPed (in blocks of 200-400k files).  While using the function below, sometimes the file does not get added to the ZIP file (initialized/created as part of 'new-object') and the
function gets stuck in an infinite loop checking for the pre-created ZIP file to grow in size greater than its initialized size of 24 bytes.  All files I'm compressing are less than 200-300k in size and compress immediately.  If the addition to the
ZIP file takes longer than 50ms, I see an error pop up once and then it moves onto the next file because the second check for file growth becomes true and the file add has successfully completed.
Personally I just would like to add a counter that checks to see how many times the while statement gets executed so that after 5-10 times, it re-issues the the command to add the file to the zipfile (which appears to resolve the issue when executed manually
from a seperate powershell window).
One option that came to mind was to wrap line for adding the file (into the ZIP file) into the while clause and then try to incorporate the counter in there.  I just haven't tried it yet.
Any additional thoughts on how to accomplish the above modification (to the script below) would be appreciated as my biggest issue is the inability to re-create the problem at-will.  
function zipit { 
  Param([string]$path) 
  if (-not $path.EndsWith('.zip')) {$path += '.zip'} 
  if (-not (test-path $path)) { 
    set-content $path ("PK" + [char]5 + [char]6 + ("$([char]0)" * 18)) 
  $ZipFile = (new-object -com shell.application).NameSpace($path) 
  $input | foreach {$zipfile.CopyHere($_.fullname); while ((Get-Item $path).length -le 25) {Start-Sleep -m 50}}
Any help is much appreciated.
- Rob

Answer 1

Thanks to all for the input.  
If I understand JRV's first comment, it's in relation to using the 'CopyHere' method and the fact without auditing for file completion (i.e. if not exists clause), it was cumbersome to use and slowed down by needing to 'sleep' the script between checks for
existance to prevent excessive CPU utilization by the job along with it being an ASYNC call with no guarentee that your file would be compressed properly without the aforementioned compensating controls.  My personal experience with this method is that
it seems to work, just not consistent when being called repeatedly (i.e. script compressing 200k files individually).  We ended up with the script getting stuck in a loop once and a while if the file didn't get added to the pre-created ZIP file.  This
was the leading reason why I continued to explore other methods as this inconsistency meant 'babysitting' our conversion process more.
I went with what I found (related to .NET Framework 4.5 ZIP functionality) as the implementation was very easy to just re-use and script against (which is effectively what Mike said to use).
The speed improvement in our case with leveraging the .NET Framework 4.5 ZIP functionality was welcomed as its performance is 4x that of using the less desirable 'CopyHere' method.  Its synchronous nature of .NET 4.5 call  also meant not putting
the script to sleep for 50ms at a time regardless if it was necesssary or not which I'm sure was also part of the boost in throughput.
- Rob