NINA / PHD2 warning - help please!

[edarter]

Hi,

I seem to be going through one of those phases where almost nothing goes right and I'm hoping you good people here can give some sage advice that sees me out of this frustrating situation.

Since I started guiding a couple of years ago I've never really had much of a problem, it all worked out of the box. Apart from the odd recalibration I was getting sound results of under 1 arcsec error and more often than not about 0.6-0.7. Happy days as my pixel scale is 1.4ish. However, over the last 3 months or so, on the odd occasions that the clouds have parted, my guiding has been plagued with glitches where either axis can throw a hissy fit before settling down(ish) again. I've tried everything I can think of within my knowledge of PHD2 and NINA. Balance, checking for sticktion, recalibrating, guiding assistant, even cleaning the RA and DEC clutches (which seemed to have a fair bit of grease on them somehow) but still the issues persist. one image can be 0.6 overall error and the next 3.5 or more. Last night I upgraded NINA from an admittedly old version and since then I have been getting the attached error at the start of every new frame. I have no idea how to resolve this having played with the timeouts in NINA, and the guiding is as bad as it was before.

I don't know if its time to strip down the mount (NEQ6 R Pro) and service it or whether there could be some other explanation / fix? The mount has never been serviced and is probs 10 years old now.

So in summary - Rubbish guiding, and now a warning in NINA with every new frame it takes. Any advice gratefully received as its really spoiling the enjoyment for me at the moment. Kit is as follows:
SW 130PDS, EOS600D, Baader MPC MK3. Guiding via SW ST80 and ASI 120MM Mini all piggybacked on the 130 via a dovetail - so very solid. Software is latest NINA and PHD2 as of last night. Happy to attach guide logs if someone can point me in the direction of where they are kept.

Thanks
Ed

[Stuart1971]

Hmmm, why do you have a settle time set in the guider….?

[tomato]

I think NINA puts a guiding settle time in the sequence as a default, but no doubt you can disable it. I see it at the start of a sequence or if guiding has been disabled for a refocus. It runs for a predetermined time to check that oscillations are within set limits before starting the exposure. It looks like the guiding deviation is exceeding the timeout period and hence why NINA is returning the errors. 

I occasionally get a healthy blip in the guiding but it only lasts a couple of guiding cycles then it settles down again. My cable management is almost non existent so I put it down to that or a freak convection current if I am pointing over the neighbour's wood burning fire chimney stack.

[edarter]

So should I set the guiding settle time to zero? I didn't set it in the first place btw! I'm just concerned that this would mean that the start of every frame (I dither every frame) would then be affected by the mount/guiding settling down after the dither

Tomato - thinking about it a bit more, do you see a spike in the guiding every time a dither is done? My graph 9 times out of 10 shows a massive spike in the guiding in both RA and DEC when the dither is done despite guiding being paused. This doesn't make sense to me.

Ed

[Varavall]

2 hours ago, edarter said:

Tomato - thinking about it a bit more, do you see a spike in the guiding every time a dither is done? My graph 9 times out of 10 shows a massive spike in the guiding in both RA and DEC when the dither is done despite guiding being paused. This doesn't make sense to me.

Hi

I see the spike on a dither with my HEQ5 with belt mod. As I set dither to random in PHD2, the size of the spike does vary a little. I thought guiding continues during a dither (maybe I missed the switch) as PHD2 does the dither. Guiding can be disabled during a slew, but a dither is not a slew.

Yes, I get the niggling timout message as well. I just ignore it as I can't find where to turn it off. Doesn't effect the guiding, at least for me. I use the preview version of NINA, something like v2.1 RC0005.

That's my tupenny worth.

Adrian

[ONIKKINEN]

If you set settle time to 0 you will have a trailed frame aftet every dither, so in this case every frame because it takes a while for guiding to lock on again.

I would advice to increase settle time to at least 30s. I use 40s and settling still fails about 70% of the time. Increasing the settling tolerance somehow would probably fix it but have not found where to do that, or have bothered to try and find that really.

[Stuart1971]

I never understood the point of this dithering malarkey, I have never done it and never seen the need to…it just seems like another thing to go wrong, and makes no difference to my images…I know it helps with hot pixels, but my darks sort all that out, and never had any walking noise either….

And besides the OP has no dithers in the guide graph shown above….🤔🤔

Edited February 9, 2023 by Stuart1971

[scotty38]

In my limited experience the guider not settling can be down to lack of calibration and/or poor polar alignment but you say you checked those?

[ONIKKINEN]

2 hours ago, scotty38 said:

In my limited experience the guider not settling can be down to lack of calibration and/or poor polar alignment but you say you checked those?

And a scruffy mount in general. If the axis are sticky, or the system is not balanced (such as because the axis are sticky), or there is excessive backlash or some other mechanical gremlin then settling will take a while. My AZ-EQ6 has some gremlins in it and so the dither spike always appears.

[tomato]

7 hours ago, edarter said:

So should I set the guiding settle time to zero? I didn't set it in the first place btw! I'm just concerned that this would mean that the start of every frame (I dither every frame) would then be affected by the mount/guiding settling down after the dither

Tomato - thinking about it a bit more, do you see a spike in the guiding every time a dither is done? My graph 9 times out of 10 shows a massive spike in the guiding in both RA and DEC when the dither is done despite guiding being paused. This doesn't make sense to me.

Ed

I don’t dither with my dual rig set up, it wastes too much time and I generate a bad pixel map in APP from the calibration frames so I don’t see the benefit. I used to see a significant blip (but not massive) in the guide graph when dithering, it was moving around 3-5 pixels from memory.

[Varavall]

6 hours ago, Varavall said:

Yes, I get the niggling timout message as well. I just ignore it as I can't find where to turn it off.

A PS to this comment. Today I re-adjusted the belt tensions after replacing one of them and the niggling timeout messages have gone. Looks like my cause was down to slightly sloppy belts

[Paul M]

Sounds about right.

I had a spell of APT timing out while waiting for PHD2 to resume guiding after a dither. It would start the next exposure after the warning but some subs were spoiled. I tuned up the mount, reducing back lash as much as possible and that fixed the problem.

[edarter]

22 hours ago, scotty38 said:

In my limited experience the guider not settling can be down to lack of calibration and/or poor polar alignment but you say you checked those?

yep - I polar align using the guide scope and Sharpcap and always make sure its at 'excellent' before continuing.

On 09/02/2023 at 16:26, Varavall said:

Hi

I see the spike on a dither with my HEQ5 with belt mod. As I set dither to random in PHD2, the size of the spike does vary a little. I thought guiding continues during a dither (maybe I missed the switch) as PHD2 does the dither. Guiding can be disabled during a slew, but a dither is not a slew.

Yes, I get the niggling timout message as well. I just ignore it as I can't find where to turn it off. Doesn't effect the guiding, at least for me. I use the preview version of NINA, something like v2.1 RC0005.

That's my tupenny worth.

Adrian

I assumed guiding stopped as I'm sure I've seen 'resuming guiding' come up as a message after a dither. I could be wrong though! I do wish I hadn't upgraded NINA now as it was just fine before!

 

23 hours ago, ONIKKINEN said:

If you set settle time to 0 you will have a trailed frame aftet every dither, so in this case every frame because it takes a while for guiding to lock on again.

I would advice to increase settle time to at least 30s. I use 40s and settling still fails about 70% of the time. Increasing the settling tolerance somehow would probably fix it but have not found where to do that, or have bothered to try and find that really.

Thanks - next time out I'll increase the settle time. Given this is in NINA, should I completely restart the sequence after each change? ie do changes not take effect until next time you initiate a sequence.

The comments about sticky axis are interesting. back in the summer I very slightly backed off the bearings in the mount as its always been a bit sticky from new. This resulted in lovely smooth rotation mostly but I still have a little stickyness rotating to the right from park on the RA. I couldn't dial this out completely without the rest of the movement being VERY free and being a bit worried that I had introduced play in to it. This has been the reasoning behind me wondering if I should bite the bullet and strip the mount, clean it all up, re-grease and sort any backlash out. Dreading the thought of doing that though in case I mess it up!

Thanks
Ed

Edited February 10, 2023 by edarter
spelling

[scotty38]

10 minutes ago, edarter said:

yep - I polar align using the guide scope and Sharpcap and always make sure its at 'excellent' before continuing.

Thanks

Ed

ok great and do you recalibrate every time you setup or do a new polar alignment?

[edarter]

On 10/02/2023 at 17:19, scotty38 said:

ok great and do you recalibrate every time you setup or do a new polar alignment?

No, that's something I don't do very often to be honest. Will give that a go. Supposed to be clear tomorrow night so fingers crossed the forecast stays like that! 

[scotty38]

53 minutes ago, edarter said:

No, that's something I don't do very often to be honest. Will give that a go. Supposed to be clear tomorrow night so fingers crossed the forecast stays like that! 

I am going to suggest that's a possible reason then. I say that as I suffered exactly the same and a new calibration, if PA changed, fixed things for me.

[edarter]

On 12/02/2023 at 08:48, scotty38 said:

I am going to suggest that's a possible reason then. I say that as I suffered exactly the same and a new calibration, if PA changed, fixed things for me.

Thanks Scotty, I will give this a go and report back.....when these clouds b*gger off!

[ONIKKINEN]

 i recalibrate every time, might be worth considering to do the same. Only takes a few minutes and since the scope is still cooling down at this point it doesnt actually bite into my imaging time.

You could open up the mount during the upcoming summer when nights are short so you lose only a little or no imaging time at all when the mount is in pieces. Its not too difficult to do either with a guide if you take pictures of how everything looks before tearing down. New SKF bearings and proper PTFE lube cant hurt and if yours has a lot of mileage already the old bearings could be busted so might give you a boost in performance. Actually would be surprised if everything looks good after 10 years of use.

[edarter]

Quick update on this. Some progress made but still not where I would ultimately like it to be. Last night was clear here in Oxfordshire (though I couldn't see the aurora - gutted!) So I did another session on the Horsehead. I increased the settle time to 30s and recalibrated PHD2 once I had slewed to the target, figuring that calibrating at the orientation I would be imaging at would give the best result. The good news is I'm no longer getting those pesky warnings popping up after every dither. The relatively good news is that guiding was better. I was averaging 1" total error but with some going above my 1.3ish pixel scale and which would therefore need to be ditched.

What I would say though is that my guiding is nowhere near as good as it used to be. Previously I would regularly get guide errors of only 0.5-0.6". Not a chance of that last night, or indeed over recent months. I'm also getting regular glitches still on both axis which PHD2 is needing to be very forceful with its corrections to keep under control. I suspect these two are related! I think I'll do a separate post asking for people to have a look at guide logs and give me their opinions.

Question - if I can feel any backlash on either axis - is that too much? I definitely feel a slight 'knock' on either axis if I try to rotate the mount by hand when the clutches are nipped up. What I don't know is how much is too much and needs tweaking out.

Thanks for all the advice so far!

Ed

[Clarkey]

I would say if you can feel it, it is probably too much. Not a major issue in RA, but definitely for DEC. 

[michael8554]

"recalibrated PHD2 once I had slewed to the target,"

The stars appear to "move" fastest at Dec = 0, which makes it easier for PHD2 to give an accurate RA Cal .

PHD2 then applies a correction to RA for the Dec you actually image at.

Imagine trying to Calibrate near Dec = 90, where the stars hardly seem to move at all.

Michael

[newbie alert]

On 27/02/2023 at 11:59, edarter said:

So I did another session on the Horsehead. I increased the settle time to 30s and recalibrated PHD2 once I had slewed to the target, figuring that calibrating at the orientation I would be imaging at would give the best result.

If using the ST4 connection then fine, if using the pulse guide method then definitely not... Calibrate on the equator/ meridian intersection , and nudge the mount north until you see the star move and then calibrate...  Waiting 30 secs won't help the known backlash