v939 very short range

First thanks a lot for the great work!
I did the flysky mod a while ago (RF module was taken from a v911 remote). Everything works great, with 30mW I get plenty of range to do outdoor flying. then I got a v939 quad and to my surprise the range is very short. using 10mW power, the v939 loses signal beyond 5 yards. even with 150mW, the range is no more than 10 yards. my first thought is that the RF modules are different, but they look to be the same (silk screen: RF-20DBM-002 V11). then I suspect I might somehow damage the module during the soldering or heat shrinking (even it works really well with v911). I carefully replaced the module and the range pretty much stays the same. while with the v939 remote, I can easily get a range of 30 yards.
Anyone has clue what could be the issue?
Thanks again for the work!

Problem solved!
The problem is that I was using a system generated 6-digit fix id (e.g., 595880), which works but gives a very short range. I change the id to 4-digit (1113) then everything works perfectly.

I don't understand this if it is true, if anyone could elaborate please, does the code length have bearing on the range of the tx, is it protocol dependant dsm/walkera/j6 etc?

I hope someone could give a definite answer, too.
summary of my experience:
1. DEVO protocol (mini cp, ladybird): 6-digit fix-id works
2. WL toys v911: 4 digit fix-id works, auto-bind.
3. BNF v911: 4-digit fix-id works. you have to re-bind each time when rx power is cycled. 6-digit and no-id does not work.
4. WL toys v939: 4 digit fix-id works. 6-digit and no-id works in some sense: it flies but the range is very short (less than 10 yards @ 150mW).

my 959 got very short range on auto ID, it is ok now after setting the ID to 1111

What a strange bug (is it a bug ?).

It's not a critical one, but did someone file a ticket on bitbucket ?

So PB could check it when he has some time.