I had some time last night to update to the latest INDI code and did some testing of the lx200ap_experimental driver with my Mach1. I discovered that for pulse guide commands of > 999ms that the mount would start to move but not stop. Looking into it today I see some of the code from the generic lx200 driver which the AP driver relies on seems to have changed and this behavior seems to be a side effect.
I have submitted a PR with the changes required to hopefully fix this problem. It has worked in testing in my office and I'll give it a shot under the stars next clear night.
This problem would have most likely prevented someone from properly calibrating their guide setup and most likely dithering as well.
Thanks for fixing this! I had a similar problem with pulse guiding the last time I used it some months ago. I didn't have a chance to do any further testing then and just went back to using my ST4 cable. I have an SD card for my Raspberry Pi3B+ with a nightly build version that I keep updated with the latest builds. If the weather ever opens up here I will give this a proper go with logging on and let you know what I find.
I had a good evening run to test these changes and they appear to work correctly. I was able to calibrate my guide setup using PHD2 as well as do dithers that required > 999ms pulse events and these worked. Previously these event would start the mount moving and not terminate. Fortunately the mount was moving at guide rate but still an unfortunately regression.
I also verified the mount tracking status will properly show the mount is tracking after the FIRST unpark since power on.
I discovered a problem however if you park the mount and then unpark without first powering down. The mount will not start tracking even though the tracking state indicates it is. I am in contact with AP to understand the proper unpark sequence if the mount has been parked and not powered down.
My suggestion for now would be to power cycle the mount after parking to make sure it starts from a known state that the driver expects.
Mike,
I finally did get some time to test the pulse guiding with my AP900 GTOCP3. I had one failure on testing some weeks back, however I have since tested two more times with the INDI nightly (last test was using the nightly from build 4-1-2019) and it worked fine on each test. These were just short tests running several minutes each, however I saw no failures during those runs. On my next outing (maybe tomorrow) I will try to test for a longer period of time. The problem I need to address now is one I am having with bluetooth connections.