EKOS scheduler Q

[vineyard]

Hello,

After a little bit of tinkering with settings on the guiding, I managed to get a decent chunk of data last night.  Will try again tonight.  I had a Q on the EKOS scheduler function.  I do PA, and then use guiding, but slowly over time (b/c my PA isn't bang on perfect) the image gradually shifts, so while each light is fine, over 30-50 lights the whole image will have shifted a bit.

So, what I have done in the past is run the EKOS scheduler on a loop of smaller routines (say 1 job of 10 lights run on a loop).  That way, at the start of each loop it re-aligns.  But I noticed last night that that doesn't happen anymore.  It just continues the next batch of imaging, without re-aligning.  Is there a way of getting EKOS to automatically re-align every so often (I haven't been able to find an easy mention of that in their online docs)?

Thank you & stay safe,

Vin

(PS - was having all sorts of frustration initially with the guiding last night where "Exposure failed" messages would come & the guiding stop.  I already had bin 2x2, but the "Swap" and exposure time of 1 sec was on.  I slowed the exposure to 2 sec, and noticed that on at least one target, the "Swap" automatically turned off.  Just sharing that in case anyone else recognises that!  Fingers crossed for tonight)

[MarkAR]

Are you not using the plate solving alignment ?

It won't drift then.

I'm pretty new to this lark but I've pretty much found a start up sequence that works.

1. Plug in and turn on everything.

2. Lauch Ekos, start indi, check everything is connected.

3. Polar align mount (not through Ekos)

4. Goto target.

5. Focus.

6. Align (plate solve)

7. Guide

8. Set up capture sequence and press go.

This gets me close enough over several nights on one target.

 

[wimvb]

Image drift is either poor polar alignment or a mechanical issue. You can check PA with Ekos or PHD guide assistant. If I were you, I would run GA anyway. It seems to me that you are trying to compensate a very basic setup miss with a software fix that isn't intended for the job. What setup do you use? Scope, mount, guide scope, guide camera? Check that everything is bolted down and that there is no mechanical movement. Also check balance and cables.

Edited April 16, 2020 by wimvb

[MarkAR]

Well, last night my rig was already set up, tweaked PA and got started.

Halfway through session I noticed the mount had drifted a little. Paused and re-aligned and everything stayed good until my power tank died.

Maybe my mount needs levelling again, might have sunk on the grass or it might be cable drag.

Edited April 16, 2020 by MarkAR

[vineyard]

Hi both, hope you're all well.  Yes I've noticed that the realignment is needed more when I'm sloppy with my PA.  So have been trying to do it more precisely now.  I use the PA tool in EKOS & last night was the closest I've got it to being bang on.  And it seems to have worked - 3 hr long sessions guided with what looks like v little image shift.  Have to look at the lights properly, but so far keeping fingers crossed.  (And Mark, I definitely use plate solving - would be lost without it, especially in the LP skies here!).

Stay safe,

Vin

 

[MarkAR]

14 minutes ago, vineyard said:

I definitely use plate solving - would be lost without it,

Me too.

Two nights ago I didn't check PA, just left it and guiding kept target pretty much dead centre. Last night I tweaked it (was slightly off)and got drift !

[wimvb]

2 hours ago, MarkAR said:

Maybe my mount needs levelling again, might have sunk on the grass or it might be cable drag.

A way to avoid sinking tripod legs and get a repeatable tripod positioning (faster PA), is to use fence post anchors or something similar. I used these (L shaped one), with the vertical plate removed.

https://www.ebay.co.uk/itm/Zinc-Plated-U-Type-Post-Fence-Foot-Anchors-Packs-/262736097733?redirect=mobile

They are cheap and won't be a problem if you have kids or dogs using the lawn daytime.

Edited April 16, 2020 by wimvb

[JamesF]

I think there's something going on with Ekos in this area, and I can't work out what it is.  (Actually I have one or two outstanding issues with Ekos that I can't solve right now.)

I'm using the internal guider and did my PA some time late last year using Ekos.  The mount is on a permanent pier in the observatory.  I power the mount off and shut down Kstars etc. at the end of every session.

On Tuesday night I captured something like four hours worth of 10m luminance subs of M101 and over the four hours there was noticeable drift over the course of the night.  Last night I set up for RGB.  I was expecting an automated meridian flip between G and B, but it didn't happen so I paused everything, moved the mount to point at a nearby star, moved it back and in the process it flipped sides.  I also checked the R and G images and saw some drift in those.  Once I'd plate solved and centred the image again I just couldn't get guiding to work, even though it had been quite happy a few minutes previously.  In both RA and DEC it was all over the place and unable to calibrate.

After messing around trying to get things to work for ten or fifteen minutes I disconnected and shut down Kstars, powered the mount and everything else off, rebooted the PC, powered the mount up and restarted Kstars.  From then on it worked absolutely fine.  Better than it did for the R and G subs, in fact.

James

[MarkAR]

It sounds like there is something not quite right with Ekos, different modules not always communicating with each other.

One thing I had to do a couple of times last night is click autofocus to get it started again, once was definitely at a filter change.

[vineyard]

I don't have an autofocuser so can't compare on that I'm afraid.

But definitely some weird glitches w the guider module recently.  For me the thing has been that sometimes while guiding, the exposure fails (I use an ASI120MM mini) and so then it drifts a bit more (into the yellow zone) & sometimes cuts out the main imaging.  In the tinkering I've been doing, I found that setting the guide bin to 2x2 helps (I got that from their online docs), but also bizarrely changing the exposure time on the guide cam.  Two nights ago, if I was setting 1sec exposures it kept cutting out, I changed it to 2sec and it was fine.

I haven't had a chance to look at last night's lights in detail yet (will Blink them on PI) so can't comment for sure on last night, but whenever I checked on it, it seemed fine (except for one weird point when the guiding cut out, I noticed the binning had somehow re-set to 1x1 so I had to change that back, and then it restarted & stayed fine).

The guide module definitely seems slightly volatile right now.

I'll probably pass on setting up tonight, so good luck if you all are out there again - hope no glitches w anything!

Cheers

[wimvb]

Have any of you posted this on the indilib forum?

I use phd with indi and haven't had these issues. And the meridian flip works every time. The only problems I've had lately were with the mount; dec backlash and stiction.

Fwiw, I have indi on a rock64 with emmc drive, zwo imaging and guide cameras (asi120mm-s) and eqmod mount. Ekos runs on the same hardware.

Is the drift you experience in ra, dec, or both? 

[MarkAR]

I'm not on that forum, spend enough time on this one. 😆

I have found some threads crop up on google but nothing relevant to this topic.

I think my drift was in RA but not certain.

@vineyardI think changing the binning and exposure for guiding may sometimes be necessary depending on seeing conditions and what stars are available. I had my first image stop last night after a few seconds but there were no error messages in any of the modules.

[JamesF]

I posted about my flip problem today in the INDI forums.  The response suggests to me that perhaps the mount has become confused about its home position.

I am going to try plugging in the handset, telling it to park itself and then if that's not where I expect I'll release the clutches and move it to the right place.

James

[JamesF]

Not sure I'm going to get to test it tonight.  Doesn't look like the sky is going to be clear for very long.

James