M31 and M81/82 in Hydrogen Alpha

My objective for this session was to capture M31 and M81/M82 in Ha. This was my first Ha-only session since discovering that filter offset focusing doesn’t work for Ha and broadband in the same session, because correct focusing of the main camera throws the guide camera too far out of focus. Ironically, filter offset focusing caused much difficulty in getting the session started, even though I was only using one filter for the entire session. In the same line of thinking that led me to change RGB exposures from 120s to 60s, I wanted to try changing Ha exposures from 7m to 5m.

The sequence that I assembled during the early afternoon was straightforward. All imaging would be 5m exposures through the Ha filter. The default autofocus filter was changed from L to Ha, and only one autofocus run was called for, and that was just before the first light frame was captured. Any refocus for the remainder of the session would be as a result of HFR increasing more than 5%. The rotation for this session was set at 30 degrees. M31 was the first target and it would be starting just before the meridian and continue until 0000. After a park/unpark pair to get the scope ready for return to the eastern hemisphere, the next target was M81/M82, and that target would continue until dawn.

Sunset was at 1646, and the end of astronomical dusk was 1820. The moon is close enough to New that it will be rising in the dawn and the waning crescent will not be a factor. Astrospheric was calling for below average seeing improving to average seeing.

I ventured outside at 1520 to uncover the scope, attach the imaging train and flats panel, and connect the power and data cables. This only took a few minutes. At 1610, I was back at the scope to power up. I came back inside to establish ASCOM connections to all of the devices in NINA.

With regards to startup activities, I elected to retain the previous session’s polar alignment. Because of the Ha filter’s much lower transmissivity,  slightly modified approach was used to focus the main and guide cameras. I first acquired and zoomed in on Polaris using the L filter and .5s exposures. Then I switched to the Ha filter with 10s exposures and went outside to focus the main camera. This stage of focusing took about a minute of outside time. I came back in side to prepare for focusing the guide camera by slewing to meridian/equator and waiting for the sky to become dark enough. Once dark enough the trip outside to focus the guide camera took about another minute. I came back inside to initiate the PHD2 calibration and GA runs.

Calibration completed nominally. GA returned hi-freq star motion of .26” and PAE of .5’. I observed the Dec plot line during the GA run, and noticed that the reported PAE varied with each plot’s variance from 0, or no error. At 0 error, PAE was 0, and the PAE was larger or smaller depending on the error of each individual plot.

I initiated the sequence well before the sky was dark enough to begin imaging. As the sequence approached the end of the built in the wait time, I monitored the physical progress of the flats panel and the initial slew.

The panel performed flawlessly, but the slew was putting the scope on the east side of the pier, which was a reach back into the eastern hemisphere. The scope was coming close to the pier, so I stopped the slew and reset the M31 target portion of the sequence, and restarted it with the scope already pointing toward the eastern hemisphere. This time it slewed correctly.

Next the sequence attempted to center the target, and that failed. It then stepped right into the first autofocus run, which also failed. I suspected the focuser, so I reset the M31 target in the sequence, and I reset the focuser ASCOM connection, both to no avail. I continued to fight the problem until I about 1930 when I noticed that the focuser was making an adjustment after the autofocus run, then I realized what was going on.

Although I was smart enough to set the default autofocus filter from L to Ha for this Ha-only session, it didn’t occur to me to turn off offset focusing. Because it was left on, NINA would add the offset steps after every movement of the filter. This would throw the main camera too far out of focus to complete plate solving to center the target or to complete an auto focus run. After turning offset focusing off and restarting, I monitored the sequence start up, which went flawlessly. At 1945 I left the session on its own for the balance of the evening.

I checked the session at 0550 and found it capturing M81/M83 as expected, and the session was performing well. Guiding was 0.75” RMS, star HFRs were 1.81. Even though I had forgotten to check the minimum ADU value at session start up to ensure separation from the left side of the histogram, I found it at a comfortable 370 ADUs. The session was set to complete at 0612.

After the session completed, the scope parked, the camera cooled, and the flats panel closed.


M31: 49 Ha x 5m for 4:05
M81/M82: 69 Ha x 5m for 5:45
Session integration total: 9:50
Session total: 1945 to 0615 = 10.30
Session Efficiency:  9.9/10.5 = 90.5%


The initial slew of the scope to the wrong hemisphere requires additional investigation. Perhaps the mount safety eastern safety limit is set too wide. It probably make more sense to be very conservative here rather than have a permissive limit. Solution: Reset mount safety limits and used a more conservative eastern safety limit.

Session efficiency was great, but this was with just one filter. I detected only five autofocus runs. More sessions with seeing and star motion information are required before any conclusions can be drawn.


The observed relationship in GA between the variance of each Dec plot and the PAE tells me that the overall trend of the Dec plots is the only important polar alignment indicator. A flat trend is great. One that slopes up or down just a little is good. One that slopes up or down a lot is bad. Simple. Observe the trend and go with one that is great or even a good one.

To-Do List

Check the eastern safety limit.

Overall and in spite of the rough start up, this was a good session. Learning how to properly set up focusing for an Ha-only session was a significant advancement in my craft. It may be a while before I am able to process this data.

Andromeda Galaxy (M31) - 2023-12-11 - Hydrogen Alpha
M31 in Hydrogen Alpha

Leave a Reply