Massimo replied to the topic 'Linear 1 Pass gives wrong solution' in the forum. 2 weeks ago

Concerning the focuser motion, believe me, up to now I built three of those things and tested them extensively. I've never seen anything strange happen after a series of forward and backward movements, like the ones you suggest, by issuing direct "go to" commands.
Indeed other focus algorithm in Ekos do work, such us polynomial and linear. It's just the one pass linear which gives this weird results... it must be something really stupid but devilish.

Read More...

Massimo replied to the topic 'Linear 1 Pass gives wrong solution' in the forum. 3 weeks ago

Question: is there a way to test that the software side of the focusing operation is working without waiting for a starry night? In the end, at least on a first basis, I have just to check that the last move issued to the focuser is the same number as that calculated from the curve fitting process.

Read More...

Massimo replied to the topic 'Linear 1 Pass gives wrong solution' in the forum. 3 weeks ago

Hi John, I red about the "set backlash=1" thing in previous posts... what I still do not understand is why in the focus module of my Ekos installation the backlash compensation is greyed out. It looks like the driver is not recognizing the myfocuser2 as backlash capable... the more I dwell into Ekos the more mysteries I found, I think it's part of the game!

Read More...

Massimo replied to the topic 'Linear 1 Pass gives wrong solution' in the forum. 3 weeks ago

Thanks a lot John. I'll take a log as soon (two weeks apparently) as the "monsoon" pass by.

However, my guess is that I'm facing the same problem faced by others of a fight between the backlash compensation of the driver and Ekos. I tried all the possible settings in the Indi driver but I got only "scattered results": some times the algorithm moves to the solution, some times it sticks to the last measurement (as in the attached pictures). I'll keep trying and record a log.

For what I understand in the nightly and in 3.6.2 it is possible to overcome this by muting the backlash compensation in Ekos... right?

Read More...

Massimo replied to the topic 'Linear 1 Pass gives wrong solution' in the forum. 3 weeks ago

I have Ekos running a myfocuser2 with the latest .321 firmware on Ubuntu 22.04 and, despite all my efforts, I have the problem with linear 1 pass algorithm moving to the wrong solution.
For the moment I'm stuck using polynomial algorithm, which by the way works ok since my focuser apparently has little backlash. But the V-curves I get with linear1 are superb... so, since for what I understand there are no other options, I'm really looking forward to 3.6.2 hoping it will solve the problem.

Read More...

Massimo replied to the topic 'SX Filterwheel one-way rotation' in the forum. 4 months ago

That would be awesome!!!

Read More...

Massimo replied to the topic 'SX Filterwheel one-way rotation' in the forum. 4 months ago

Hi, thanks a lot for the info.

I wrote to Terry Platt of SX asking for the source code of the Windows program, his answer was: "I don’t have the source to hand but the rotation direction is easily changed. All you need to do is to send the filter position numbers with 128 added for reverse rotation, so if you never do this, the wheel will always rotate anti-clockwise."

So I checked the sxwheel.cpp code and at line 149 I found the instance: "SendWheelMessage(f + 0x80, 0);"
I naively removed the "+ 0x80" which should be + 128 in HEX, recompiled and installed the driver.
And it magically works, rotation is always anti-clockwise.

There is a but, though. For whatever reason, at start-up, after the wheel finishes indexing (i.e. after it completes a full rotation at power on) the first filter switching is random, clockwise or anti-clockwise following the shortest route. Only the initial connection after indexing experiences this. If I disconnect and reconnect from the INDI Control Panel the rotation is always anti-clockwise, but if I disconnect, power down the wheel, power up it, wait for indexing, then the first movement is random... I'm not a coder, I tried a few random things (add + or - 0x80 here and there... but it didn't work.

My solution is that, since after indexing the wheel always (at least for as much as I tried) start at position number 1, I try to always disconnect the wheel with the filter set at position number 1. Then on connection the filter wheel doesn't move and if I then I make the first movement to position 2, it turns anti-clockwise and from there it keeps going that direction.

Surely a INDI wide implementation would remove all this hassle.

Read More...

Massimo replied to the topic 'SX Filterwheel one-way rotation' in the forum. 4 months ago

I'm facing the same problem... my 36mm filters are not large enough for my QHY268 due to the overshooting. In particular flat correction becomes nearly impossible.

Hope someone is able to put the option for 1-direction in the driver.

Read More...

Massimo replied to the topic 'Kstars crash when live view open' in the forum. 5 months ago

I've seen this type of behaviour when my system was under-powered. Live view suck up a lot of current, so if your power source is underrated or suboptimal under live view high load all type of erratic behaviour can happen. But maybe is not your case.

Read More...

I'm happy it worked... sorry if my explanation was a little bit sketchy.

Best

Read More...

Mmmm...as a last resort, if I were you, I would try exactly what you did, i.e. generate a clean profile with only the camera, but first I would delete all previously generated configuration files and create a new one based on that profile. Maybe you are lucky that way. In my case this worked with KStars 3.5.9 and also 3.6.0 but I have a Mars-M.

Read More...

It happened the same to me. The solution, however, was very simple.

Disable "Auto Connect" in the EKOS configuration panel. Then save a configuration for the camera in the INDI control panel, only after that press the connect button.
Apparently if a configuration file is not present KStars gets mad.

Read More...