NavList:

Frank, thanks for the advice.  I’m keen to keep at it, but it won’t be tonight, because the low cloud has crept in, which secretly I’m rather pleased about, because I’ve been neglecting my other jobs.  Lying in bed this morning, I thought the first thing I’m going to do is tinker with last night’s values to see how far off my sextant work was, but you’ve done that for me, so thanks for that too.  The Venus height was spot on and the moon wasn’t bad, so all the recording error is in the lunar.  One problem was remembering where the drum was in the dark.  In the end I came up with ‘a bit at a time’, but need more practice on ‘how much gives what’.  One thing I thought about is, if the value is changing, if you can work out if it’s increasing or decreasing, can you set a slightly low or high value and wait for the Moon to drift on?  Also, if the apparent speed of the planets varies cf the stars, are there times when a Moon - planet lunar is likely to be more or less accurate than a moon – star one?  E.g. like comparing the speed of the hour hand of a clock to the speed of the minute hand.  DaveP

David Pike, you wrote:
" This time UTC was only 6 minutes 10 seconds out so it’s improving."

That's a big error. It's roughly equivalent to 3.1 minutes of arc error in the observation, which is 10% of the Moon's diameter. Clearly you could see a gap that large between the Moon and Venus even under terrible conditions. So something is going wrong.

May I suggest you try a Sun-Moon lunar tomorrow. The angle is convenient. Historically, daytime lunars were considerably more common than night lunars. Also, remember this: lunars were not required at a moment's notice. There was no pressure to shoot one unless there had been horrible weather and perhaps an imminent landfall. You have time to pick and choose opportunities when the geometry is not fatiguing. As the old joke goes, "Doctor, Doctor, it hurts when I do this!"... "Well then don't do that..." With lunars, if it's painful, make it un-painful. Arm tired? Then find a fence to lean against. Hard to look at the Moon in the horizon glass? Then try it the other way around, with the Moon in the index mirror's view. Also, try sitting down. That can help a lot, and it was recommended historically, too. And as I have suggested for many years, don't shoot "live". Use both hands to hold the instrument's frame for stability. Look at the gap or overlap between the Moon and the other body, and then lower it to make a small adjustment with the micrometer. Then look again. Rather than trying to hold the instrument stable, and tweak the micrometer simultaneously, as is normally required with altitude observations, you can be much more relaxed with lunars. Lunar observations are much more carefree in that sense. You're not in a race with a rapidly-changing altitude.

The mathematical process of clearing lunars can be an additional source of error when you're new at this. Why not take that part out of the equation while you're just experimenting with the sights themselves? Use the clearing web app on my web site here: http://www.reednavigation.com/lunars/. When I clear your sight using my web app, I get an error of 2.0' which would be approximately equivalent to 4 minutes of time. That's definitely better than the results you've reported, but there's still plenty of room for improvement. 

Frank Reed