54 replies to this topic

[KJR]

Recently I had my AVX tuned at Starizona where they adjusted the RA and DEC motor and worm gear backlash.

 

About 10 days ago I decided to set up my rig (see signature below) under a Bortle 3 sky with little to no wind to test out guiding. My goal was to get RA and DEC to be below 1" RMS. I wasn't going to worry about imaging. The moon as half full at the time. I wanted to try the PPEC algorithm in PHD2. I watched Jason Burns youTube video about guiding with an AVX and I set up my PHD2 settings to match what he suggested.

 

I've been trying to image the Leo Triplet recently so I decided to test guiding around this target. I've been using NINA only about a couple of months. I have to set up my rig from scratch every session. I've been trying to balance my AVX RA slightly east heavy (counterweight on the east side) and DEC slightly camera heavy, which is made difficult by the stiction on the axes (DEC is the worse).

 

With index marks set and everything powered up I take a few images around the NCP and check focus and test plate solve.

 

Then use NINA Three Point Polar Alignment. I aim a 10" or better. But sometimes plate solving in the procedure becomes  inconsistent but eventually I get polar aligned.

 

Next test GoTo to a bright star to check focus.

 

Next use PHD2 Calibration Assistant to slew to its default coordinates and calibrate. I think I have had one PHD2 calibration that didn't complain since I started guiding. Most of the time it complains about something I don't understand. Tonight PHD2 complained but the graph showing the RA and DEC alignment was pretty good so I accepted it. See first photo. Then I slewed and centered to the Leo Triplet and to my surprise PHD2 latched onto a guide star and started guiding. Wow, so this is how it's supposed to really work! 

 

I just watched the real time graph and wondered if the PPEC algorithm would work. Seemed like at first it wasn't but the RMS values of the RA and DEC started to come down. The Signal to Noise ratio was a little low so I bumped up the exposure from 1 sec to 1.5 sec and I was in the green at about 15 for SNR.

 

I did run the Guiding Assistant an applied Min/Mo changes.

 

I started taking some 1 minute exposures and to me they looked good. I increased to 2 minutes and 3 minutes.

 

My PHD2 graph was looking like this most of the time in the second photo.  I thought about starting a sequence but decided to take test exposures of 4, 5, 6, 7, 8, 9 and 10 minutes as an experiment.

 

I'm using my Starizona corrector on my C8. According to Starizona it reduces to 7.1 for the C8.

 

The third photo is my 5 minute test exposure. I took a screen shot from ASIFitsView. The file name at the top shows the details of the exposure. 90 gain, Bin=1 300s

 

From there I increased each test exposure in one minute increments until I reached a 10 minute exposure.

 

KJR

Attached Thumbnails

• 
• 
• 

Edited by KJR, 27 April 2024 - 02:16 PM.

[KJR]

This next photo is my 10 minute exposure. 

 

I'm not sure I'll ever see this again. 

 

I'll post screen shots of my PHD2 settings next. Maybe they can help others. I have set up once more since this evening and the guiding was still pretty stable so maybe these changes will allow me to have more consistent sessions.

 

KJR

Attached Thumbnails

• 

[KJR]

Here are my PHD2 settings I used.

Attached Thumbnails

• 
• 
• 

[idclimber]

You do not need 600" subs on broadband targets with any modern CMOS sensor including the 585mc. This is especially true if you running this in high gain mode which is like 250 or so. I would guess you could and should be using no more than about 120" and if you are in a high bottle zone more like 90 or  even 60. 

 

If you want to know, upload a raw light and a raw bias from that camera to a shared drive. I will compare the background to the bias. My guess is your bias/dark is around 500 ADU and your background on your 300" sub is up there around 3 to 5k. 

 

As far as taming the AVX there are many threads on the subject. Some get it to work ok and others don't. I think it is mount dependent and luck of the draw. Are you lucky??

[KJR]

The rest of the settings.

 

If anyone would like to make recommendations for further changes I would appreciate the suggestions.

 

Thanks

 

KJR

Attached Thumbnails

• 
• 

[rgsalinger]

PHD may not like your calibration because the rates in DEC and RA are different. Assuming that you (as the screenshot shows) are calibrating at the celestial equator (DEC 0) then the RA and DEC rates should be the same. They are not. 

 

The header on your second screenshot shows "300" rather than 600 for what seems to be the exposure length. However, I can't imagine that anyone with a modern CMOS camera would actually be using longer than 5 minute exposures for broadband DSO imaging. 

 

It's odd to me that you would use a .9 sidereal guide speed. I would expect the mount to be subject to RA backlash if the mesh is even out a tiny bit.  So,where did that recommendation come from?

 

So, all in all, I would say that the Starizona tuning has worked well. I'm a little baffled by your focal length, though. If you are using a C8 at 2 meters and put a .63 reducer on the scope then the focal length should be 1260 not 1454. Did a miss something?

 

There are some things in those settings that I really find odd. First of all the 174 is a 16 bit camera and so saturation should be at 65535 and not 255. It's a mistake to guide with 8 bits even if you can choose that value. Second, I always run with noise reduction turned on but I've never measured how things go with it on and off. 

Edited by rgsalinger, 27 April 2024 - 02:26 PM.

[idclimber]

So, all in all, I would say that the Starizona tuning has worked well. I'm a little baffled by your focal length, though. If you are using a C8 at 2 meters and put a .63 reducer on the scope then the focal length should be 1260 not 1454. Did a miss something?

He is correct in that the Starizona IV is f/7.1 on a C8. 

[imtl]

I hope you do not take this in a negative way because that is not my intention.

 

The PHD2 graph you showed here does not look very impressive. Especially not with the image scale you're trying to deal with.

 

You got RA rms of around 0.84" and DEC rms around 0.45" which totals around 0.96". That's actually not really good when you want to image with a 1442mm focal length. Your image scale is smaller than that at 0.68"/pixel (I assumed you used your ASI071mc). 

 

The graph itself looks quite jumpy. Not smooth on the scale of +-2".

 

You set your dither to do only RA which is bad.

 

And eventually, did you actually look at your image? What is the FWHM you got on your best frame?

 

Just guiding for 10 minutes is not really a challange. Guiding well with good results is. Make sure you actually have that because on the surface it does not look like it.

[Spaceman 56]

I hope you do not take this in a negative way because that is not my intention.

 

The PHD2 graph you showed here does not look very impressive. Especially not with the image scale you're trying to deal with.

 

You got RA rms of around 0.84" and DEC rms around 0.45" which totals around 0.96". That's actually not really good when you want to image with a 1442mm focal length. 

imtl is getting a bit advanced here in my opinion.  

 

KJR is someone learning the basics, and I feel he is doing very well.  I think KJR will go a long way quickly with his obviously smart mind.

I would have been wrapped with Guiding at 0.96 RMS when I first had a crack at it. 

 

itml is right that at 1442mm better guiding will be required, but I think 0.96 as a starting point is quite acceptable for a beginner.

 

I also like the fact the KJR took a series of test shots at 5,6,7. and 10 minutes. thats exactly what I did too, and whilst it does not prove anything

or mean that he WILL image at those exposure times, it does give him an idea of where things are at.

 

Suggestions from me might be to reduce that Guide Speed Sidereal from 0.9 to 0.5 as thats more the recommended settings generally.

as Rosalinger said you might get oscillations at 0.9 as the mount may overcorrect.

 

Spaceman.

[Oort Cloud]

PHD may not like your calibration because the rates in DEC and RA are different. Assuming that you (as the screenshot shows) are calibrating at the celestial equator (DEC 0) then the RA and DEC rates should be the same. They are not.

The header on your second screenshot shows "300" rather than 600 for what seems to be the exposure length. However, I can't imagine that anyone with a modern CMOS camera would actually be using longer than 5 minute exposures for broadband DSO imaging.

It's odd to me that you would use a .9 sidereal guide speed. I would expect the mount to be subject to RA backlash if the mesh is even out a tiny bit. So,where did that recommendation come from?

So, all in all, I would say that the Starizona tuning has worked well. I'm a little baffled by your focal length, though. If you are using a C8 at 2 meters and put a .63 reducer on the scope then the focal length should be 1260 not 1454. Did a miss something?

There are some things in those settings that I really find odd. First of all the 174 is a 16 bit camera and so saturation should be at 65535 and not 255. It's a mistake to guide with 8 bits even if you can choose that value. Second, I always run with noise reduction turned on but I've never measured how things go with it on and off.

Not OP, but I will say that my EQ6-r definitely seems to guide better at .9x than .5x, so it may just depend on the particular copy of mount in question.

Also, the 174 is a 12-bit camera, not 16. If OP is running it in 8-bit mode (PHD2's default settings), then 255 is absolutely correct. However, PHD2 can run the camera in 16-bit mode meaning that it will interpolate the 12-bit values into a 16-bit space. However, doing so results in saturation at a point below 65,535, so that absolutely should not be set as the max. I don't remember the specific value off-hand, but it's easy enough to test for. Find a star that is obviously saturated, and select it manually. I know from using the 533 (14-bit), that the max is 65,532 for 14-to-16 but conversion, but the only time I've seen it for 12-to-16 was when I did what I just said to input the correct value for my own 174, which I had set to 16-bit mode, and then proceeded to have issues with, due to incorrect saturation value setting.

[imtl]

I'll add a further comments.

 

1. Your DEC aggresivness is too high. People have a tendency to go very high with that, thinking that they can fight their mount with it. 

 

2. Your RA MINMO is way too low. Why are you setting it at 0.36 when your RA is at 1px on average? Your mount is at constant work with too many pulses. Guiding is about correcting when needed. Not "correct" all the time. Do you see the sawtooth pattern in both RA and DEC? That means your mount is struggling. Ease on the parameters. The plot should be much smoother than this.

 

And yes, it might mean that your rms will be higher, but at least your cycle will be much smoother. :Look at your peak values, not only rms. They are large, even for something shorter than 1442mm focal length. 

 

Spaceman, these are not advanced stuff, this is basic.

[rgsalinger]

I've been running both QHY and ZWO 174 cameras with the saturation setting at 65536 (or so) for about 6 (or so) years now. I get pretty good results, I think. As I understand it, the ZWO camera driver converts the 12 bit ADC output to 16 bits before PHD sees the pixels. That's based on this "Most guide camera drivers return image data in either 8-bit or 16-bit formats even if thecameraelectronics natively work in 12-bit or 14-bit modes" from the PHD manual on page 26. (Yes, I had to look it up.)

 

So, nothing should be lost in terms of rejecting saturated stars by setting it up as I've been doing. If you set it down to 12 bits (4096) then I would expect that you will find a ton more saturated stars. Looks clear tonight so I'l play with this. Who knows my guiding might get even better if I monkey with this! 

 

Setting it to 255 and using 8 bit mode is really a bad idea regardless unless you are using an 8 bit camera. Otherwise you are throwing away the dynamic range of the guide camera (12 bits). Even worse a small increase in brightness at the top saturates the star. 

 

I've never seen any of my mounts work better (or worse) by monkeying with the guide speed setting. Now, one can never rule out the exception but I tend to go with what the manufacturer recommends for my mounts. I know, though, that the closer you get to 1x, the more likely that a command to slow the tracking will slow the mount sufficiently to create RA backlash which you really don't want. 

Edited by rgsalinger, 27 April 2024 - 07:23 PM.

[ayadai]

The proof is in the pudding. If perfectly round stars are achieved in a 10-minute exposure with that focal length, the settings are exactly where they should be. OP: ignore the noise in this thread and keep on doing exactly what you're doing.

Edited by ayadai, 27 April 2024 - 08:04 PM.

[imtl]

The proof is in the pudding. If perfectly round stars are achieved in a 10-minute exposure with that focal length, the settings are exactly where they should be.

No. Round stars are not the only measure. If they are bloated, aka FWHM is large, then it could be partially because of bad guiding.

[ayadai]

No. Round stars are not the only measure. If they are bloated, aka FWHM is large, then it could be partially because of bad guiding.

No perceivable bloat in those images. Just admit it. The OP has it dialed in.

[idclimber]

No perceivable bloat in those images. Just admit it. The OP has it dialed in.

The guiding alone is double the image scale. I am however super impressed you can measure fwhm from that screenshot. 

[ayadai]

The guiding alone is double the image scale. I am however super impressed you can measure fwhm from that screenshot. 

Who cares? Pinpoint stars and amazing detail. The ultimate arbiter is our eyes.

 

Punter 1: "Hey that horse just won the race by 4 lengths!"

Punter 2: "Yeah, but his legs were .0001% shorter than the other horses and the leg frequency was .00025% faster"

 

The horse still won.

Edited by ayadai, 27 April 2024 - 08:53 PM.

[rgsalinger]

There's a point here that can get lost in the "sarcasm". You will get round stars even if, for example, you are out of focus. So, what you're after is both round and small stars because those indicate the level of detail that you have available. Another similar point is that poor tracking in both axes will give you round stars but again, they won't be small and you're not getting all the detail. 

 

Maybe this is less important these days as we now use AI based sharpening tools, but the difference between excellent tracking and well focused data is not small at all and will be noticed in the quality of your images. You just can't judge data quality beyond roundness usng "our eyes". You need to measure and then think about how to get better data next time out. (Or at least I do.)

Edited by rgsalinger, 27 April 2024 - 09:43 PM.

[ayadai]

You just can't judge data quality beyond roundness usng "our eyes". You need to measure and then think about how to get better data next time out. (Or at least I do.)

And that's the crux of the biscuit, 'aint it? If the difference in the final image is so subtle that even an experienced hand has difficulty in discerning it, does all the effort represent a good ROI?

 

I understand following the FWHM and other numbers, because, science. I do find it disingenuous however, when said practice takes the form of hyper-critical and/or unsolicited advise when assessing someone else's results.
 

[imtl]

The guiding alone is double the image scale. I am however super impressed you can measure fwhm from that screenshot.

It's AI that took control over their eyes.

Edited by imtl, 27 April 2024 - 10:53 PM.

[KJR]

PHD may not like your calibration because the rates in DEC and RA are different. Assuming that you (as the screenshot shows) are calibrating at the celestial equator (DEC 0) then the RA and DEC rates should be the same. They are not. 

 

The header on your second screenshot shows "300" rather than 600 for what seems to be the exposure length. However, I can't imagine that anyone with a modern CMOS camera would actually be using longer than 5 minute exposures for broadband DSO imaging. 

 

It's odd to me that you would use a .9 sidereal guide speed. I would expect the mount to be subject to RA backlash if the mesh is even out a tiny bit.  So,where did that recommendation come from?

 

So, all in all, I would say that the Starizona tuning has worked well. I'm a little baffled by your focal length, though. If you are using a C8 at 2 meters and put a .63 reducer on the scope then the focal length should be 1260 not 1454. Did a miss something?

 

There are some things in those settings that I really find odd. First of all the 174 is a 16 bit camera and so saturation should be at 65535 and not 255. It's a mistake to guide with 8 bits even if you can choose that value. Second, I always run with noise reduction turned on but I've never measured how things go with it on and off. 

Thanks for the heads up on the wrong photo. Here is the 10 minute exposure.

 

I use the PHD2 Calibration tool which slews the mount close to the celestial equator. I may try using Stellarium on my minPC to set another location for the NINA framing assistant to slew the mount.

 

The Starizona website indicates the focal length on the C8 would be 1450, mathematically 7.1 would be 1454 but I used their value.

 

The sidereal rate of .9x is what the CPWI mount driver uses. According to the PHD2 manual, PHD2 doesn't set it. My ASIAir, which I don't use anymore due to AVX interworking issues, also used .9x. 

 

The guide drivers I use in PHD2 are ZWO drivers which are 8 bit. I tried the ASCOM drivers which show 16 bit but PHD2 always seems to get confused and it chooses the wrong camera even though I have NINA connect to the ASI071MC first before starting up PHD2. If I remember correctly the ASCOM driver doesn't allow me to see the two camera in PHD2. The only way I knew what camera PHD2 connected to was to look at the pixel value in the settings and often it chose the ASI071 and I couldn't get it to switch so I just use the ZWO drivers. 

 

Thanks for the comments. 

Attached Thumbnails

• 

Edited by KJR, 28 April 2024 - 10:04 AM.

[KJR]

I hope you do not take this in a negative way because that is not my intention.

 

The PHD2 graph you showed here does not look very impressive. Especially not with the image scale you're trying to deal with.

 

You got RA rms of around 0.84" and DEC rms around 0.45" which totals around 0.96". That's actually not really good when you want to image with a 1442mm focal length. Your image scale is smaller than that at 0.68"/pixel (I assumed you used your ASI071mc). 

 

The graph itself looks quite jumpy. Not smooth on the scale of +-2".

 

You set your dither to do only RA which is bad.

 

And eventually, did you actually look at your image? What is the FWHM you got on your best frame?

 

Just guiding for 10 minutes is not really a challange. Guiding well with good results is. Make sure you actually have that because on the surface it does not look like it.

I agree the RMS values are not impressive but I was surprised by the results of my individual photos. I've been working to try to get my guiding under 1". I'm hoping with some further changes and recommend suggestions I may achieve that goal.

 

 I've had some test images when guiding was 1.2" to 1.8" RMS and I could see oblong stars.  I only was testing the guiding changes that night and I only took individual test images to test the guiding with each image being one minute longer in exposure. My total guiding time according to PHDLogView was 1 hour and 22 minutes with the 10 minute exposure at the end of the session. I expected the guiding to get worse as time progressed but it seemed stable but I was surprised with my results.

 

I haven't done much dithering when taking a sequence and don't know much about it. I know the mount will move a little at a set interval and that the stacking software will remove some noise that it won't do in without a dithered stack. 

 

The HFR in the file name of the 10 minute exposure from NINA shows 4.39. Is that the same as FWHM?

 

I'll be looking at upgrading my AVX in the future but this is what I have to work with, "stone knives and bear skins" to quote Mr Spock.

 

Thank you for your comments.

 

KJR

[KJR]

imtl is getting a bit advanced here in my opinion.  

 

KJR is someone learning the basics, and I feel he is doing very well.  I think KJR will go a long way quickly with his obviously smart mind.

I would have been wrapped with Guiding at 0.96 RMS when I first had a crack at it. 

 

itml is right that at 1442mm better guiding will be required, but I think 0.96 as a starting point is quite acceptable for a beginner.

 

I also like the fact the KJR took a series of test shots at 5,6,7. and 10 minutes. thats exactly what I did too, and whilst it does not prove anything

or mean that he WILL image at those exposure times, it does give him an idea of where things are at.

 

Suggestions from me might be to reduce that Guide Speed Sidereal from 0.9 to 0.5 as thats more the recommended settings generally.

as Rosalinger said you might get oscillations at 0.9 as the mount may overcorrect.

 

Spaceman.

It appears PHD2 just takes the CPWI settings it doesn't calculate. I will try this recommendation. I don't want to change too much in one sitting as I will not know what change was affective. Thank you

[idclimber]

The HFR in the file name of the 10 minute exposure from NINA shows 4.39. Is that the same as FWHM?

 

I'll be looking at upgrading my AVX in the future but this is what I have to work with, "stone knives and bear skins" to quote Mr Spock.

 

Thank you for your comments.

 

KJR

Voyager also gives me HFR. My understanding is FWHM is roughly double HFR. The units are pixels so we would need to know which of your cameras you are using to calculate pixel scale. Then it is a simple method to compare by converting to "/px. 

 

If you care to upload the file, we could measure and compare in a more objective way. 

[KJR]

I'll add a further comments.

 

1. Your DEC aggresivness is too high. People have a tendency to go very high with that, thinking that they can fight their mount with it. 

 

2. Your RA MINMO is way too low. Why are you setting it at 0.36 when your RA is at 1px on average? Your mount is at constant work with too many pulses. Guiding is about correcting when needed. Not "correct" all the time. Do you see the sawtooth pattern in both RA and DEC? That means your mount is struggling. Ease on the parameters. The plot should be much smoother than this.

 

And yes, it might mean that your rms will be higher, but at least your cycle will be much smoother. :Look at your peak values, not only rms. They are large, even for something shorter than 1442mm focal length. 

 

Spaceman, these are not advanced stuff, this is basic.

I forgot to add I used PHD2 Guide Assistant that night and used the recommendations for the min/mo.

I usually don't run the Guide Assistant. I had been running min/mo at a higher level like 1.2 to 1.4. But 

 

The DEC aggressiveness of 100 was what PHD2 defaulted. I will make a note to try to make a change to lower that value.

 

Thank you for your advice.

 

KJR