Getting the best performance from my AM5: Page 14

Getting the best performance from my AM5

w7ay

@Kevin_A

Practicing what I preach...

I just started to write the code to support my second iteration of an All Sky camera.

The first version used an ASIAIR, with a pancake version (uncooled) of the ASI178MC and a 3.2mm CCTV lens (higher quality that the stuff ZWO gives away for free, but still just a CCTV quality lens). WiFi. Power through a 12V source.

This version is using INDIGO Sky on a Raspberry Pi 4. The camera is an uncooled ASI294MC. The lens is a Sony-E mount version of the Samyang 8mm Fish-eye. Interface will be Ethernet with power from PoE through the same cable. The ASIAIR just is too crippled.

The lens went from this

https://www.amazon.com/gp/product/B08GLYR572

to this

https://www.amazon.com/Samyang-Fisheye-Mount-Cameras-SY8MBK28

I have just started modifying the (macOS) program I had written to work with INDIGO so that it is specialized for All-sky and Meteor captures (my plan is to do frame subtraction to only record frames with meteors -- avoiding aiplanes is going to be challanging).

The "Preview" screen from the original program will be used to capture and saving images. So far, it still has the stuff to control focusers and filters, and a histogram. Those will be removed and replaced by all-sky and meteor specific control in a different window -- I want an All-Sky display to be no-nonsense, with nothing but a view of the sky (and trees, in my case :-).

Hey, notice the two parameter histogram adjustment (the little green arrows) to get proper histogram balancing (not like the single parameter on on the ASIAIR app-- again, another "too easy," thing in ASIAIR).

Now try to do all this on an ASIAIR. Even when you are not my 75 years of age :-).

(Right over my head is my FSQ-85 stored on the second RST-135 that is on a Gitzo tripod.)

Chen

Kevin_A

w7ay awesome! It is a shame that ZWO does not utilize your knowledge to improve their products. Practicality and understanding is something that is lacking in product and programming these days. I too am getting old my friend… I try to pass on the little I know but people want it easy and ready to run without caring about understanding it…. They only care when it does not work, while not trying to learn. Cheers!

w7ay

Kevin_A It is a shame that ZWO does not utilize your knowledge to improve their products.

Actually they ask me a lot of questions through email. But they seldom take any recommendations. I knew about the ZWO mount development more than a year ago. Just recently had a question on an announced feature in an announced proct, but they have no idea how to implement it.

Their aim is very different from mine. Mine is to spread knowledge, theirs is to make money from the naive.

Chen

Kevin_A

w7ay i am definitely not as knowledgeable as yourself in these field but do share the same views and similar sharing goals. My background is more of a multipotentialite. I have many hobbies and background skills ranging from photography , mechanical engineering, an artist, naturalist, psychology and what I learn and discover, I try to pass on to others to help them learn. These days it seems most people do not feel the need for understanding, they just want instant gratification. The world as we know is changing and we can thank greed and corporate for that. I just released my latest artwork to a gallery last week and it sold in 3 minutes… it is called “ Perspective”, it is about finding beauty, happiness and good things in life that matter in a world so cold, greedy and influenced. By using your own mind and spirit you can change anything. Lots of cloudy nights to keep myself busy these days… so I do many things… just not in all the same ways.
Have a nice night!

Kevin_A

w7ay i did manage to get 2 hours last night on M13. Almost a full moon and guiding was at 0.7 overall during the night. I am not sure if the lower 200ms pulses affects the calibration process but I did notice that during the cal process, the stars seemed a bit offset when switching axis… kind of like it was strugling to get the star back on axis causing a big initial error correction when finally switching from calibration to guiding. It did 12 steps in each axis… It then caught up when guiding started and then held the star in the 12 green circles. When cal switched from e to w to n to s the main star was above the dotted line not on it. Not sure why, even when backlash clears 2-4 times.
I may even remove the covers and check for belt backlash soon. I have noticed this issue a few times now… but not everytime. This is not a issue to me but more of a lesson to be looked at.

w7ay

Kevin_A i did manage to get 2 hours last night on M13

Nice.

M13 was (back in about year 2001) the first time I realized how much Richardson-Lucy (blind deconvolution) can sharpen up even defocused stars.

Back then, the Hartman mask was worth the cardboard it was cut from. Pavel Bahtinov changed it all (and the ASIAIR "auto focus" took us back a step again -- it is sad to see humans devolving, just to make a few bucks). But focusing a star to a pinpoint was really tough without diffraction masks. Today's focusing the Sun is equally hard, since there is no point sources at infinity during the day to use diffraction masks like the Bahtinov.

I started using my own M3 images from the f/6.3 8" Meade images for use at work to experiment with image sharpening. Apple's Accelerate group (one floor down from my group in the same building :-) got wind of it and was impressed enough with R-L that they hard coded Richardson-Lucy into their framework. All because of a fuzzy M13 image :-).

You can read about the Accelerate framework here:

https://developer.apple.com/accelerate/

I still swear by the FFT algorithms in that library today. Faster than a speeding bullet :-).

By the way, Meade made the f/6.3 for only a limited time. With a 0.63x reducer, it made that 8" an f/4. Then they got sold to a Chinese company, which then got sued out of business a couple of years ago for shady business practice by Orion. The Meade part was subsequently taken over by Orion. MEAD used to be listed on the NASDAQ even.

Chen

Kevin_A

w7ay i still use a bahtinov mask today for focussing initially and use that as a reference number on the asiair eaf during the later part of the night. I was impressed last time imaging with my askar fra300, as I fixed its overly tight focusser that was affecting the eaf severely and my eyes said 3760 and eaf came in at 3760 too, so my eyes are still working fine with a mask! Haha My Meade 6000 series 115mm triplet is still my favorite scope from 6-7 years ago. Only FK61 but a good optical design for its price. I am wondering if my fan on the 2600mc pro has a bit of vibration as it is a bit more rougher than my 533 and 183 smaller cameras…. I need to test that out as I have a slight bit of elongation and bloat. More things to check and there has been widespread thoughts on how zwo mounts it fans and if it is isolating any vibration or contributing more to it. I could try adding a better sorbothane o-ring washer duro 40 but that is still up for thought.

Gernot

w7ay

For now, thanks for your efforts! Unfortunately, due to the persistently bad weather, I still haven't had the opportunity to test the mount. As soon as I have experience I will publish them here.

CS Gernot

w7ay

Kevin_A I am wondering if my fan on the 2600mc pro has a bit of vibration as it is a bit more rougher than my 533 and 183 smaller cameras….

Just change the setpoint temperature so that the fan is just off, to measure a reference star size. Then change the setpoint so the fan just turns on (this way, off and on only changes dark current noise minimally just in case it affects the FWHM calculations).

Take a few exposures with the fan off and a few with it on and take averages.

Send the images to some program (like AstroPixelProcessor) that analyses the FWHM.

Do not trust the HFD numbers from ASIAIR - it is not accurate, and depends on the background light level -- you may have noticed that ASIAIR HFD (when "auto-focusing" and with Star Detect tool) gives better numbers before it gets full dark out of astronomical twilight. Well, just another of ASIAIR's bugs.

See if the FWHM changes from fan off to fan on.

Unfortunately, Noctua does not make the larger (thicker) fans used in the larger ZWO cameras (2600, 6200) etc. They do make one that will fit the older cooled ASI183. Otherwise, it is a high quality European fan, with integrated anti-vibration pads instead of external grommets.

For now, the best you can do is to add some rubber grommets (which should have been there in the first place), or switch camera manufacturer (I recently bought the QHY version of the cooled IMX533 sensor instead of the ZWO cooled ASI533).

Chen

Kevin_A

w7ay i was planning shooting a few with the fan on and off. The camera came with orange grommets but i will change them to a better material and repeat my tests.

w7ay

Kevin_A The camera came with orange grommets but i will change them to a better material and repeat my tests.

The orange could indicate that they are silicone instead of rubber. Silicone can remain pliant with lower temperatures; so you may want to keep the silicone ones, or get thicker silicone ones.

Rubber O-rings tends to be spec'ed to something around -18ºC, and silicone can go lower (I have seen them spec'ed to -50ºC, gulp).

But if you won't go outdoors below -18ºC (I won't even take my telescope out at 0ºC :-), then the rubber ones should be fine, especially if they are more vibration absorbing. You just need to do some "before/after" tests. BTW, you may be able to do this with artificial stars indoors on a cloudy night -- that's why everybody who does any serious astrophotography need to own a good artificial star.

However, if the current grommet is no good, it only gets worse with lower temperatures. I'll bet ZWO will release some new fans that have the proper damping, and make everyone pay again. They did this with leaky filter trays etc. You end up paying more in the long term than buying a higher quality product at the beginning.

Chen

Kevin_A

w7ay yes, the orange ones are silicone but very thin. I have shock absorbing sorbathane sheet material that is 2.5mm thick and a soft duro 40 that I may try out making donuts from. As far as the cold goes, I don’t work below -10c so my equipment won’t have to endure that either! Haha Some people say that a direct mount to the body is good if the body of the camera is affixed to a heavy scope, others swear by isolating the fan with springs to float it in the housing. It just seems a bit weird that the zwo design is just recently being scrutinized and questions on only the 2600 cameras! I would have thought it would have been an issue a long time ago unless their fan suppliers are shipping out duds due to profit over quality. Quality is getting worse, day by day and I know about light leaks in zwo and bad William Optics flatterners, scopes etc as I have returned a lot of good name brand products over the last 2 years…. Sigma too!

AstroBen60

w7ay

So, I investigated a bit and realized there are two types of mounts here
Some have a 432s period, some 288. (433/1.5)
Mine, shown above, and the one of @xmj0001, show a shorter period than other ones.
But on zwo Website, the Am5 is stated for a 288 seconds period only ( and the am3 also by the way)
Did zwo change the design? Did they change of manufacturer?

By the way, I used webplotdigitizer as advised above, and this is fantastic and easy tool to measure that curve and get relevant data about it.

Thanks a lot if anyone has any information about this main period difference!

Ben

xmj0001

So I've started to jump into this issue. Seems quite intensive from what I've read in most of the forums. But if I understand correctly I should forgo Auto-Calibration in PHD2 in the ASIAIR app as it will get the wrong numbers. I should manually set the Calibration Step for about 1900ms Set Max DEC 150ms Set Max RA 150ms, Guide Scope should be set to .5 sec to 2 sec exposure. Guide Camera to Max Gain Aggressiveness starting at 5% on both working up till i get good RMS values.

Guide Scope: WO 50mm UniGuide 50mm
Guide Camera: ZWO ASI120MM Mini

Kevin_A

xmj0001 based on your guide scope and camera combo calibration steps should be around 1100ms. Let it calibrate on auto. The max durations and aggressiveness settings do not influence calibration, moreso affecting guiding once calibration is complete. Start aggressiveness at 30-40% and see how the lower numbers work first before bumping things up. Max Dec and Rabdurations should start at about 200ms on each but you will need to calculate your PE slope to get accurate numbers. These are just balpark numbers. Guide exposure length will be either 0.5 to 1s and this too will be based on how sharp your PE slope calculates out. Read some of the earlier posts regarding how to calculate guiding parameters based on your mount as others numbers are based on their own PE slopes…. Cheers.

xmj0001

Kevin_A Thank you

Galaxium

w7ay

I took my ZWO provided PE graph and digitized it via WebPlotDigitizer. I took that data and put it in a spreadsheet and calculated the local slope between each row of data. That didn't work out so well but I was able to visually identify the max slop area and choose the points to give me my maximum slope.

I chose these points to calculate my max slope:

(250.92, 6.588)
(251.03, 2.5158)

=> 4.0722 / ( 0.11 * (24 hr * 3600 sec / 360 degrees )
=> 4.0722 / (0.11 * 240)

I am unsure about the 240 as it relates to my AM5. Graph below of a November 2023 mount.

Max Slope (arc-sec / sec) = 0.154

Am I doing this correctly? My guiding results were not great with the ASIAIR+ defaults. I have been testing RA and DEC Aggression percentages. I have found this thread and everything is making more sense. It appears ZWO has made some improvements to the AM5 mounts over time? My graph below for inquiring minds.

I have not read how to properly calculate the "Calibration Step" size. Can someone clear this up for me? Kevin_A

ASIAIR+
William Optics 32mm f/3.75 Guidescope Focal Length 120mm
ASI120MC Mini (I just tossed my ASI 174MM in the garbage which I knew was a mistake but now I know it is confirmed as a mistake)
AM5
Redcat 71
I assume 0.5x sidereal since I haven't changed anything in my ASIAAIR+.

Many thanks to everyone in this thread with special appreciate to Chen. Thank you!

MrAstro

Hi,

Can anyone please take a look to my AM5 PE report and suggest Max Dec., Max Ra., and calibration step ..
Other settings I can tuning during the imaging session I suppose according to environmental factors, Like Exp. time (usually 1-1.5 Sec.) and aggression (usually 50-70 )
May be Chen can do with thanks 🙂 ..
I am using PHD2, ASI290MM Mini and Skywatcher 9x50 finder FL=175mm.
My imaging scale for the guider is 3.42"/pixel
All my imaging sessions are in Bortle 2-3 ( Travelling ).

CS all !

Kevin_A

MrAstro you should be using between 0.5-1.0s long guide exposures. Guide calibration step size will be around 900ms based on your guide scope and 290mm mini camera. Aggressions should always start at the lower range of around 30-40% on each axis. These are good ballpark settings. If you follow some of the previous post you can figure out what your max durations need to be when using either 0.5s or 1.0s guide exposures. You want to bump up the gain on your guide camera so multistar guiding selects at least 10-12 stars. I set mine now at 300 gain in asiair pro.
Your dec and ra max durations should be around 200ms on each using 1s guide exposures but that is just a ballpark but never go higher ever than 750ms on 1s exposures. You need to calculate your slope yourself like we all had to to get the exact correct number for 0.5 or 1s and the duration pulse length in ms. I would start off using 1s guide exposures and if it needs faster corrections use 0.5s. Cheers.

MrAstro

Kevin_A
Hi,

Long time huh!!..🙂
I already calculate it, Its = 0.141"/sec.
So, Could you tell me please what to do with this number? Rather than I search all this long thread 🙂
I am using PHD2. WO scope FL=200, ASI290MM mini.
Thanks

AstroBen60

Hello Folks

As I was waiting for my AM5 to arrive, I spent a lot of time reading all that extraordinary stuff and the links posted here.
I tought I was ready to measure the steepest slopes as soon as I would get it out of the box , and get ready for a perfect setting of phd2 ...

What a surprise I got when watching at my Test Report... !
It doesn't look like others posted here. It 's way more detailled and chaotic.
Looking at the amplitude, it looks like it's actually quite reduced commpared to others posted here. But the curve is'nt smooth at all.

So I wonder why this curve is like this, and how to deal with it.

Is it so detailled because of the relative small Total amplitude , and so the zooming effect on that part of the curve gives like a higher resolution of the measurement ? How to get back to a reasonnably smoother curve that I could work with ?
Is it so rough because that particular mount is just a piece of crap ?
Or finally, did Zwo change it's measurement method, with a higer resolution of measurement ?

Thanks a lot for your help .

Ben

Kevin_A

AstroBen60 yes, they have changed their supplier! It should guide great but definitely looks like you will need fast guiding. Let us know how it works out for you. Mine runs between 0.5 to 0.7 rms everytime no matter if its 0.5-2.0s guide exposures at max durations between 150-350ms. And thats fine for my image scales.

w7ay

AstroBen60 Mine, shown above, and the one of @xmj0001, show a shorter period than other ones.

Yes, ZWO had changed it to a shorter period, ignoring the astrophotography rule of thumb that mounts with longer period are easier to guide (which is mathematically explained using slopes -- the slope of the fundamental of a periodic waveform is inversely proportional to the period). If you have problems guiding, part of your problem is from the shorter period.

Increase your guiding frame rate -- use a 0.5 second guide exposure with a camera that can do 2 FPS, and couple that with a good guide scope to compensate for the short exposure time. Attempt to reduce the guide rate to as low as you can get away with (0.25x sidereal rate will flatten out the guide sawtooth compared with 0.5x sidereal rate). Use a guide camera that gives you a wide FOV since you will need that to get as many stars as possible in multistar guiding, which you will need to reduce "seeing" problems with such short exposure times.

Make sure that the max pulse duration is comfortably shorter than the guide exposure time, otherwise the longer guide pulses will cause the guiding to skip frames.

FWIW, the RST-135 and the NYX-101 (both strain wave geared mounts) have gearings that give 430 second period (sidereal day divided by 200).

Chen

AstroBen60

Kevin_A

Thanks Kevin
Before reading your reply, I just sent an email to zwo, because they have a large FAQ about how to deal with the PE Test Report (it looks like they read that forum before writing it 😁), but state a 432 main period in their examples.
But the specs on the product page do state 288 sec for main period.
So they should confirm as you say they change supplier.

My image scale is 1,94"/pixel, so I thing it will be OK.

I'm still in the process of getting valuables results of my data with webplot and excel.
Actually, we have to take into account the slope in between points, but also the amplitude.
I mean a really steep slope giving 0,5arcsec move, then going the opposite way, even steeper for 0,4arcesc, is not realy a big deal... As long as your resolution is OK.

Ben

« Previous Page Next Page »