astrosatch How many stars must be minimum on screen for multistar guiding to kick in? I have OAG and there are not many stars to choose from.
As usual, the answer is "it depends."
ASIAIR will obviously not pick any stars that are saturated (I did confirm this with their folks) since saturated stars are pretty useless for guiding (centroid changes with how much it is saturated as the star scintillates).
When stars are too dim, their signal to noise ratio will also make it not useful, since at some point, the noise is greater than useful contribution of the star centroid. I do not know what ASIAIR is using as the SNR (or whatever criteria for dim stars).
So, there is only a range of stars that are usable. If too many stars are saturated, you are wasting those stars (and the camera gain will just add noise to the dimmer stars). As usual, don't just crank up the gain, as many people do, but to adjust the gain to find the "goldilocks" region where you have the maximum number of guidable stars.
One way you can do it is to start with a log gain. Ask ASIAIR to guide and look at how many stars it chooses. Then go increase the gain by say 3 dB (30 units in the gain slider) and restart guiding again. Keep doing this to see if the number of stars ASIAIR picks increases.
As of the v1.6 Beta 5, ASIAIR autoguides with a maximum of 12 stars.
If the best you can get is say, 4 stars, then your guide system does not have a high enough dynamic range. You may beed a camera with a lower read noise, or a guide scope with a larger objective.
There may be regions of the sky where you might not get 12 stars. While moving the mout by just a little will give again more stars.
Remember that multiple stars just reduces the centroid estimation error. And if probability theory holds, the variance half each time you double the amount of stars; i.e., the RMS changes by 0.707 each time you doule the amount of stars.
The biggest improvement is of course doubling from 1 star to 2 stars, where the RMS of the measurement (not the RMS of the guiding, since that depends on the quality of the mount) drops by 0.707. Adding a third star won;t have as much an effect, unil with 4 stars, you again improve by another factor 0f 0.707.
Similarly, you won't gain the full 0.707 factor until you hit 8 stars, etc. So, 10 stars and 8 stars are probably not very noticeable.
Most of all, more stars does not mean you have better guiding, as I discussed very early in this thread. If the mechanics of your mount may be a limiting factor well before the limitation of accuracy the centroid. The mounts that benefits most from multi-star guiding are mounts that have very small backlash. (The PHD2 log should give some idea of how much backlash is measure during the calibration phase.)
I had already mentioned to ZWO folks some time back that I expect people with OAG not to benefit as much as people with large guide scopes, together guide cameras with a deep well. OAG typically have a 8x8 mm prisom, which limits how much light reaches the sensor. I have myself stayed far, far away from OAG for years now.
Focus as well as you could and you should have the most guidable stars.
I also use a Baader Semi-APO filter on my guide scope (55mm aperture, fluorite doublet) to minimize UV (it starts cutting off at 480nm) and IR (it cuts IR past 1200nm, beyond most IR cut filters). That also maximixes SNR, but also cuts about half of the light, so probably not suitable for OAG either. I use a ASI290MM (not the mini, but the USB3 version). I wish I could use a ASI178MM for better dynamic range, but the frames are too large and the FPS is too slow for guiding (I wish you could select the guide ROI), so I am stuck with the 290. I have a ASI462, but that is a color camera and not useful as a guide camera.
Chen
