+ sign has a semantic meaning in the query string. It is used to represent a space. Another character that has semantic importance in the query string is & which is used to separate the various var=value pairs in the query string.

Most server side scripts would decode the query parameters before using them, so that a + gets properly converted to a space. Now, if you want a literal + to be present in the query string, you need to specify %2B instead.

+ sign in the query string is URL-decoded to a space. %2B in the query string is URL-decoded to a + sign.

See the difference between

http://www.google.com/search?q=foo+bar

and

http://www.google.com/search?q=foo%2Bbar

In the above examples, Google's server script is URL-decoding the query parameters and then using them to do the search.

URL-encoding is nothing but % sign followed by the hex-code of the special character. For example, we know that the hex code of A is 0x41 (decimal: 65). Try this:

http://www.google.com/search?q=%41

Hope this makes URL-encoding clear.

So, if you want the + sign to be preserved when a JavaScript is fetching a URL with + signs in its query parameters and a server side script would process the query parameters after URL-decoding it, you should URL-encode the query parameters in the URL before using issuing the HTTP get request so that all + signs are converted to %2B's when the request reaches the server side script. Now when the server side script URL-decodes the query string, all %2B's gets converted back to + signs which is what you want.

See Encode URL in JavaScript? to learn how to URL-encode the parameters using JavaScript. Short answer from the discussion there:

var encodedURL = http://example.com/foo.php?var= + encodeURIComponent(param);