First of all, I have both an 0805 and 0806 (which I am currently using). It seems to me that both units use the same GPS base since I have had occasion to use a unit in the base of the other unit successfully.
I have had similar issues as you. In a different post here (started on 2/2), I was complaining about the fact that even after the lock, I had instances that the lock was lost part of the way into the trip and it went back to 0. I was told here to try putting power into the 0806 instead of the base unit. I tried that and SO FAR, there has been no loss of lock once it locked. For me, the jury is still out - I will keep checking for the next couple of weeks.
As far as the initial 'lock' is concerned, I have had varying times to first lock. Here is my data for all my trips since 2/2 (last Tue) when I changed the power to the 0806. This data comes from analysis through the Datakam player.
13 trips (not video segments) -
Trip Time To Lock
1 3m07s
2 0m08s (25 min after end of #1)
3 8m34s (night time made lock take longer?)
4 4m01s (also night time)
5 no lock - 6m58s trip (happened later once - I'm not sure why it never locked)
6 4m2s
7 11m19s (night time)
8 5m5s (night time)
9 4m56s
10 0m12s (22 min after end of #9)
11 4m40s (sunset)
12 no lock (2m10s trip)
13 no lock (13m56s - night time)
So from this small amount of data here are my tentative conclusions -
1. Day time lock takes 4 to 5 min
2. Night time lock takes 8 to 15 min
3. Trip 5 didn't lock across 7 min at 9 am on a cloudless day. Not sure why but it is wasn't far out of my 4 - 5 min range
4. If a trip starts soon after a previous ended, GPS lock is very quick
I take a look at another week or so of data to see if it is congruent with the above.
Have other of you experience data that matches mine?
I'm also in a 2011 Prius and have the base mounted behind the rear view mirror below that funny stuff/tint on the top of the windshield. Will putting it somewhere else improve the 'lock'? Though moving it somewhere else will reduce it's 'stealth' and could change the video picture.