the browser gives an Internal Server Error

The code I showed isn't a complete CGI example, since it doesn't output the headers, so those would need to be added back in. Since in the original code those are being written by hand, I'd suggest at least upgrading to one of the CGI modules, such as e.g. CGI::Simple, to generate those for you.

Wide character in IO::Compress::Zip::write

That would mean that there's Unicode in your @lines. (Although I don't see an encoding being set on TARGET in the original code, so I think it would have the same issue?) Anyway, although IO::Compress::Zip provides a filehandle-like interface, it looks like it doesn't (yet?) support encoding layers. A manual encoding with Encode does work though:

use warnings;
use strict;
use IO::Compress::Zip qw/$ZipError/;
use Encode qw/encode/;

my @lines = (qw/ Hello World Foo Bar /, "\N{U+1F42A}");
my $eol = "\r\n";
my $encoding = "UTF-8"; # or maybe "CP1252" for Windows

binmode STDOUT;  # just to play it safe
my $z = IO::Compress::Zip->new('-', # STDOUT
    Name => "Filename.txt" )
        or die "zip failed: $ZipError\n";
for my $line (@lines) {
    $z->print( encode($encoding, $line.$eol,
        Encode::FB_CROAK|Encode::LEAVE_SRC) );
}
$z->close();
[download]

Note: For encodings such as UTF-16, it seems encode adds a Byte Order Mark for every string it encodes, and I don't see an option in the module to disable that. One way to get rid of them is to remove them manually, but an alternative might be to replace the for loop with this, at the expense of higher memory usage: $z->print( encode($encoding, join('', map {$_.$eol} @lines), Encode::FB_CROAK|Encode::LEAVE_SRC) ); - or just stick to UTF-8, as that's pretty ubiquitous.

Update:

I remove the $CRLF from the lines

You can leave that in and remove my $eol, as they're the same thing (I missed that on my first read of the original source, sorry).

Thank you, thank you!

I'm now making some progress. I could live with it as-is, I think, but would like, if possible, one more improvement: The Content-Length. I suppose, however, that is not possible when compressing on the fly.

I was already using Encode, so there was little more to adjust. I have now downloaded a zipped file successfully, which can be opened normally, with the following code.

sub exportdatabase { 

fork: {

my ($recnum,$revnum,$book,$chap,$verse,$text) = '';
my @resp = ();
my $timestamp = "$curdate_$curtime";
$timestamp =~ s/[\/:.]/-/g;
my $to_windows = '';
my $CRLF = "\n";
     if ($OS eq "Windows") { 
         $to_windows = '--to-crlf'; # SAME AS -l
         $CRLF = "\r\n";
} 
my $zipfile = "$db_export_file.zip";
my $encoding = "UTF-8";

$statement = qq| SELECT a.RecordNum, a.RevisionNum, a.Book, a.Chapter,
+ a.Verse, a.Text from $table a INNER JOIN (SELECT RecordNum, max(Revi
+sionNum) RevisionNum FROM $table GROUP BY RecordNum) b USING (RecordN
+um,RevisionNum); |;

&connectdb('exportdatabase');

push @resp, "RECORD#\tREVISION#\tBOOK#\tCHAP#\tVERSE#\tTEXT, AS EDITED
+ BY: $curdate $curtime (Pacific Time)$CRLF"; 

while (($recnum,$revnum,$book,$chap,$verse,$text) = $quest->fetchrow_a
+rray()) {
            push @resp, "$recnum\t$revnum\t$book\t$chap\t$verse\t$text
+$CRLF"; 
    }
    
binmode STDOUT;  # just to play it safe
        print qq|Content-Type: application/zip, application/octet-stre
+am$CRLF|;
        print qq|Cache-Control: no-cache, no-store, must-revalidate$CR
+LF|;
    print qq|Accept-Ranges: bytes$CRLF|;
    print qq|Content-Language: utf8$CRLF|;
    #print qq|Content-Length: | . (stat $zipfile)[7] . "$CRLF";
    print qq|Content-Disposition: attachment; filename="$zipfile";$CRL
+F$CRLF|;

my $z = IO::Compress::Zip->new('-', # STDOUT
    Name => "$db_export_file" )
        or die "zip failed: $ZipError\n";
    
for my $line (@resp) {
    $z->print( encode($encoding, $line,
        Encode::FB_CROAK|Encode::LEAVE_SRC) );
}
$z->close();

} #END fork
} # END SUB exportdatabase
[download]

Without the Content-Length header, the client does not know how large the file being downloaded is, nor how long it will take. But, at least the file arrives intact!

The IO::Compress::Zip docs say you can pass a reference to a scalar to get the data written there. This will use more memory, but here's an example that includes the Content-Length header:

#!/usr/bin/perl
use warnings;
use strict;
use CGI::Simple;
use IO::Compress::Zip qw/$ZipError/;
use Encode qw/encode/;

my @lines = (qw/ Hello World Foo Bar /, "\N{U+1F42A}");
my $CRLF = "\r\n";
my $encoding = "UTF-8"; # or maybe "CP1252" for Windows

my $zipdata;
my $z = IO::Compress::Zip->new(\$zipdata,
    Name => "Filename.txt" )
        or die "zip failed: $ZipError\n";
for my $line (@lines) {
    $z->print( encode($encoding, $line.$CRLF,
        Encode::FB_CROAK|Encode::LEAVE_SRC) );
}
$z->close();

my $q = CGI::Simple->new();

print $q->header( -type => 'application/zip',
    -Content_Disposition => qq{attachment; filename="Download.zip"},
    -Content_Length => length($zipdata) );

binmode STDOUT;  # just to play it safe
print $zipdata;
[download]

As for the code you showed, I'd strongly recommend proper indentation (perltidy can help with that). I also have the feeling that you might not be using strict and warnings, or that you're using too many global variables (e.g. $statement). And a nitpick: I wouldn't use $CRLF for outputting the HTTP headers, "\n" is enough, every webserver I've worked with understands that (but really, you should be using a module to write the headers anyway, as I showed above).

Also, as I mentioned, you can save the memory of the @resp array by merging those two loops. For example:

$z->print( encode( $encoding, "RECORD#\tREVISION#\tBOOK#\tCHAP#\tVERSE
+#\tTEXT, AS EDITED BY: $curdate $curtime (Pacific Time)$CRLF", Encode
+::FB_CROAK|Encode::LEAVE_SRC ) );
while ( ( $recnum, $revnum, $book, $chap, $verse, $text ) = $quest->fe
+tchrow_array() ) {
    $z->print( encode( $encoding, "$recnum\t$revnum\t$book\t$chap\t$ve
+rse\t$text$CRLF", Encode::FB_CROAK|Encode::LEAVE_SRC ) );
}
[download]

If any of the fields you're outputting could possibly contain tabs, you might want to switch to Text::CSV for proper handling of all of those complexities.