Currently EKOS arrives at best focus by moving the sensor "in" and "out" with decreasing amplitude, thus arriving at minimum HFR value.
My home built focuser is belt driven, thus backlash in the system becomes an issue: this could be avoided:
First HFR is mapped in one direction (e.g.: "in" movements only) from outside to inside of best focus position. Once the V curve is established the focuser would rewind to its start position, and incrementally, using only "in" commands arrive at HFR minimum. Backlash avoided, that should eliminate most failed autofocus procedures.
Michael, I don’t understand. Belt driven systems are known for their low backlash or even complete absence of it. My home built focuser is belt driven and has 0 backlash. Wouldn’t it be possible to eliminate the backlash completely?
Wouter van Reeven
ASI6200MM and 7 slot 2" filter wheel with a SkyWatcher Esprit 80 ED on a SkyWatcher HEQ5-Pro
ASI1600MM-Pro Cooled and 5 slot 1.25" filter wheel with an 8" TS Ritchey-Chrétien on a SkyWatcher EQ6-R
My Pegasus DMFC fortunately has an internal driver setting to specify backlash.
But I do agree that it would be valuable to have a focus mode in EKOS that can take care of backlash by itself, for those that cannot set it in the driver.
A simple solution would be to have the focusser approach dialed-in values always from the same side, with a definable overshoot. I.e., to go from 10000 to 9950, it first goes down to 9800, and then up again to 9950.
Should of course be optional, so people with driver-internal backlash compensation can switch it off.
openSUSE Tumbleweed KStars git INDI git
GPDX+EQMOD, CEM60EC, ASI2600/1600/290mini+EFW+EAF
Unless your belt is tight... I don't see how there wouldn't be backlash. I have mine loose enough to get off so I can do large adjustment. I just put a 'bar' on one side of the belt to take out some slack, but slide the bar out and I have slack enough to get it off the gears.
Off topic: That's a cool idea for design. I can't do it on mine, but I'm assuming that's a rotator as well?
On Topic: I also use a 3D printed focuser, and I have similar issues with backlash, though usually it's pretty good about it. I'll put it on my list of things to look at doing. But I've been quite involved with a local makerspace lately, and that doesn't show signs of letting up.
I'm also doing the same. Mine is the Deepskydad AF1. So far, it's been working great. The focuser module hasn't worked very well lately, once the module creates the "donut" it gets lost and reverts back to the original position. So I've been using my bahtinov mask. But to the subject of backlash, the bar in the photo below seems to do the job. I'd like to have some sort of pulley setup with a spring, but this works.
I'd happily volunteer to code a unidirectional focuser, if I get one or volunteer testers!
If you're interested, respond on this thread or message me.
I've been thinking the same thing for a while.
I warn you, though, that I believe backlash is only part of the problem.
I believe the HFR calculations (and, of course, seeing) are noisy and contribute to the lack of good v-curves.
Orion Atlas Pro, William Optics ZS105, Moonlight V2 focuser
ZWO ASI1600, OAG & Filterwheel, Astronomik Filters, QHY 5L-IIm guidecam.
KStars/Ekos/Indi on RPi4 w/SSD
Ekos Projects: Terrain Backgrounds, Polar Alignment, Analyze, Linear Focuser, SEP MultiStar & GPG Guiding, FITS autostretch.