diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py
index 463f995c7..fc1210079 100644
--- a/youtube_dl/extractor/crunchyroll.py
+++ b/youtube_dl/extractor/crunchyroll.py
@@ -52,28 +52,51 @@ class CrunchyrollBaseIE(InfoExtractor):
         username, password = self._get_login_info()
         if username is None:
             return
+        '''
+        import cfscrape
 
-        self._download_webpage(
+        proxies = {"http": self._downloader.params.get('proxy'), "https": self._downloader.params.get('proxy')}
+        tokens, user_agent = cfscrape.get_tokens(self._LOGIN_URL, proxies=proxies, user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:61.0) Gecko/20100101 Firefox/61.0")
+
+        self._set_cookie( '.crunchyroll.com', 'cf_clearance',tokens['cf_clearance'])
+        self._set_cookie( '.crunchyroll.com', '__cfduid',tokens['__cfduid'])
+        '''
+        login_page = self._download_webpage(
             'https://www.crunchyroll.com/?a=formhandler',
             None, 'Logging in', 'Wrong login info',
             data=urlencode_postdata({
                 'formname': 'RpcApiUser_Login',
                 'next_url': 'https://www.crunchyroll.com/acct/membership',
+                'fail_url': self._LOGIN_URL,
                 'name': username,
                 'password': password,
-            }))
-
-        '''
-        login_page = self._download_webpage(
-            self._LOGIN_URL, None, 'Downloading login page')
+            }), expected_status=503)
 
         def is_logged(webpage):
-            return '<title>Redirecting' in webpage
+            return '<title>Redirecting' in webpage or '/logout' in webpage
 
         # Already logged in
         if is_logged(login_page):
             return
 
+
+        '''
+        print [tokens, user_agent]
+
+
+        form_data = self._form_hidden_inputs('challenge-form', login_page)
+        form_data['jschl_answer'] = self.solve_challenge(login_page, 'www.crunchyroll.com')
+        print form_data
+        self._sleep(6, None, 'Solving CloudFlare Challenge')
+        login_page = self._download_webpage('https://www.crunchyroll.com/cdn-cgi/l/chk_jschl', None, 'Login Form', data=urlencode_postdata(form_data), headers={
+                    'Referer': self._LOGIN_URL,
+                }, expected_status= 503)
+
+        import codecs
+        with codecs.open("yop", "w", encoding="utf-8") as f:
+            f.write(login_page)
+        '''
+
         login_form_str = self._search_regex(
             r'(?P<form><form[^>]+?id=(["\'])%s\2[^>]*>)' % self._LOGIN_FORM,
             login_page, 'login form', group='form')
@@ -107,7 +130,7 @@ class CrunchyrollBaseIE(InfoExtractor):
             raise ExtractorError('Unable to login: %s' % error, expected=True)
 
         raise ExtractorError('Unable to log in')
-        '''
+        
 
     def _real_initialize(self):
         self._login()
@@ -123,6 +146,7 @@ class CrunchyrollBaseIE(InfoExtractor):
         # Crunchyroll to not work in georestriction cases in some browsers that don't place
         # the locale lang first in header. However allowing any language seems to workaround the issue.
         request.add_header('Accept-Language', '*')
+        request.add_header('User-Agent', 'Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:61.0) Gecko/20100101 Firefox/61.0')
         return super(CrunchyrollBaseIE, self)._download_webpage(request, *args, **kwargs)
 
     @staticmethod