Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Comparing Two AM5 Mounts (actually only their PE Test Reports)

  • Please log in to reply
13 replies to this topic

#1 jerrygerber

jerrygerber

    Mariner 2

  • *****
  • topic starter
  • Posts: 259
  • Joined: 07 Oct 2022
  • Loc: San Francisco, California, USA

Posted 10 December 2023 - 12:09 AM

Which of these two AM5 mounts, based on their respective data sheets do you think is most likely to produce the best autoguiding results?  As you can see, not only are the maximum and minimum PE quite different, but the close up of the curves are very different as well, one is very jagged and the other quite smooth.

Attached Thumbnails

  • AM5-2.jpg
  • AM5.jpg


#2 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 34,880
  • Joined: 27 Oct 2014

Posted 10 December 2023 - 12:38 AM

The newer one has better specs.  Both will clearly need autoguiding, there may be little difference between them, guided.

 

They're both jagged, raw ("full") data. 



#3 arbit

arbit

    Apollo

  • -----
  • Posts: 1,401
  • Joined: 19 Feb 2012

Posted 10 December 2023 - 12:42 AM

Can't say.

In theory the one with fewer harmonics and lower PE slope should guide better.

In practice, well.. Just got a second AM5, the new version whose graph makes your first chart look like a pure sine wave :-)

Guides at 0.4 to 0.5 rms total close to Dec 0 both sides w/o counterweight. It's actually a little better than the first one. The first one is the old model, so exactly the situation in your post.

I should add that the maximum PE slope is similar for both, and so the same PHD2 guide settings are used.

Ideally, you would customise the PHD2 settings to each mount, which would set off any innate difference between the two.

Sent from my SM-S908E using Tapatalk

Edited by arbit, 10 December 2023 - 12:50 AM.


#4 jerrygerber

jerrygerber

    Mariner 2

  • *****
  • topic starter
  • Posts: 259
  • Joined: 07 Oct 2022
  • Loc: San Francisco, California, USA

Posted 10 December 2023 - 01:17 AM

Hi Arbit,

 

Hmm, but doesn't the one with the far better min and max PE error look much less "smooth" to you?  The small jagged lines within each slope concerns me on AM5-2, which is the newest one.  The numbers are a lot better but I'm trying to understand how those jagged lines might affect guiding, if they do.The older AM5's PE error numbers are certainly not nearly as low, but the waveform is far smoother.  It probably takes an expert in strain-wave engineering to help me understand this!


Edited by jerrygerber, 10 December 2023 - 01:17 AM.


#5 OzAndrewJ

OzAndrewJ

    Cosmos

  • *****
  • Posts: 8,417
  • Joined: 30 Nov 2010

Posted 10 December 2023 - 01:39 AM

Gday Jerry

Pretty much no units so far really look like their "bench" graphs

when plotted under load in real world situations.

Looking at that data, the first one looks like it has much lower rates of change

so my guess is it may guide better.

That said, the data is only a snapshot of one point in the 360deg

so as per above, it may look very different at another spot.

 

Andrew Johansen Melbourne Australia



#6 bobzeq25

bobzeq25

    ISS

  • *****
  • Posts: 34,880
  • Joined: 27 Oct 2014

Posted 10 December 2023 - 01:56 AM

Hi Arbit,

 

Hmm, but doesn't the one with the far better min and max PE error look much less "smooth" to you?  The small jagged lines within each slope concerns me on AM5-2, which is the newest one.  The numbers are a lot better but I'm trying to understand how those jagged lines might affect guiding, if they do.The older AM5's PE error numbers are certainly not nearly as low, but the waveform is far smoother.  It probably takes an expert in strain-wave engineering to help me understand this!

More an expert in guiding.



#7 arbit

arbit

    Apollo

  • -----
  • Posts: 1,401
  • Joined: 19 Feb 2012

Posted 10 December 2023 - 08:43 AM

Hi Arbit,

 

Hmm, but doesn't the one with the far better min and max PE error look much less "smooth" to you?  The small jagged lines within each slope concerns me on AM5-2, which is the newest one.  The numbers are a lot better but I'm trying to understand how those jagged lines might affect guiding, if they do.The older AM5's PE error numbers are certainly not nearly as low, but the waveform is far smoother.  It probably takes an expert in strain-wave engineering to help me understand this!

So this is only from one part of the complete curve. Unlike worm gears, the plot may look very different at a different part of the cycle / loading / position. So these charts are really just for getting some idea of the max / min pe. Since there is so much variation over the full cycle, the general approach is to guide fast. There is hardly any backlash, so the recommendation is to avoid aggressive moves to avoid oscillation.

 

The approach that seems to have worked generally is eyeball the graph, do some measurements and calculate the max slope of the curve. For most people, this seems to be in the region of 0.20 - 0.30 arc-sec/ sec. Mine is 0.22 on one and 0.25 on the other, for example.

 

Then, based on your imaging and guiding scales, you can figure out your guide exposure (1-1.5 sec seems to work for most people, some report 0.5 or even 2), max RA / Dec moves (200-300ms should be fine to start with), and min-move. Aggression needs to be lower than for traditional mounts. Most people find 50% RA and 60% DEC a good starting point. A guide rate of 0.3x works well for me.

 

There are a couple of excellent threads over on the zwo support forum which take you through all this step by step.



#8 Kevin_A

Kevin_A

    Surveyor 1

  • *****
  • Posts: 1,737
  • Joined: 22 Aug 2018
  • Loc: Belmont, Ontario Canada

Posted 10 December 2023 - 10:03 AM

The newer one is a 288s period length and the older one is 432s. The partial chart is really useless as it only represents a small area of gear tooth engagement at the largest PE area. Large PE is not the priority when autoguiding, the slope is more important of a factor to determine your guiding results. Rainbow mounts are an example of that theory as their PE is large but the PE is smooth. 
Both look fine but both will require different aggression, duration and exposure setups.
The newer one is more sinusoidal in shape.
I would prefer guiding the mount shown above with the smaller overall PE for astro, which is the 288s period as it requires less to get it back on track even if it needs it done faster to get there due to steeper slopes.
The only real way to check however is to try them both out.

My mount is 0.16 arcsec/sec so technically a good one but it behaves only in the 0.5-0.7rms guiding range.
Rainbow mounts have much higher 60-90 PE but they are much smoother sinusoidal curves with less additional harmonic distortion so very very small 40-80ms rapid corrections at 0.5s intervals are needed to get great results. I have seem a lot of 288s period charts that were very knarly looking yet guided very well, so having a very low PE and a fairly sinusoidal yet a bit zaggy partial should do ok. Test, test test is the best way to find out. 


Edited by Kevin_A, 10 December 2023 - 10:37 AM.


#9 jerrygerber

jerrygerber

    Mariner 2

  • *****
  • topic starter
  • Posts: 259
  • Joined: 07 Oct 2022
  • Loc: San Francisco, California, USA

Posted 10 December 2023 - 01:38 PM

So this is only from one part of the complete curve. Unlike worm gears, the plot may look very different at a different part of the cycle / loading / position. So these charts are really just for getting some idea of the max / min pe. Since there is so much variation over the full cycle, the general approach is to guide fast. There is hardly any backlash, so the recommendation is to avoid aggressive moves to avoid oscillation.

 

The approach that seems to have worked generally is eyeball the graph, do some measurements and calculate the max slope of the curve. For most people, this seems to be in the region of 0.20 - 0.30 arc-sec/ sec. Mine is 0.22 on one and 0.25 on the other, for example.

 

Then, based on your imaging and guiding scales, you can figure out your guide exposure (1-1.5 sec seems to work for most people, some report 0.5 or even 2), max RA / Dec moves (200-300ms should be fine to start with), and min-move. Aggression needs to be lower than for traditional mounts. Most people find 50% RA and 60% DEC a good starting point. A guide rate of 0.3x works well for me.

 

There are a couple of excellent threads over on the zwo support forum which take you through all this step by step.

Hi Arbit,

How do I calculate the max slope of the curve?  I've heard that's important for understanding Max and Min RA and DEC duration (which I usually set around 150-250ms) but I don't how to calculate that..



#10 jerrygerber

jerrygerber

    Mariner 2

  • *****
  • topic starter
  • Posts: 259
  • Joined: 07 Oct 2022
  • Loc: San Francisco, California, USA

Posted 10 December 2023 - 01:39 PM

The newer one is a 288s period length and the older one is 432s. The partial chart is really useless as it only represents a small area of gear tooth engagement at the largest PE area. Large PE is not the priority when autoguiding, the slope is more important of a factor to determine your guiding results. Rainbow mounts are an example of that theory as their PE is large but the PE is smooth. 
Both look fine but both will require different aggression, duration and exposure setups.
The newer one is more sinusoidal in shape.
I would prefer guiding the mount shown above with the smaller overall PE for astro, which is the 288s period as it requires less to get it back on track even if it needs it done faster to get there due to steeper slopes.
The only real way to check however is to try them both out.

My mount is 0.16 arcsec/sec so technically a good one but it behaves only in the 0.5-0.7rms guiding range.
Rainbow mounts have much higher 60-90 PE but they are much smoother sinusoidal curves with less additional harmonic distortion so very very small 40-80ms rapid corrections at 0.5s intervals are needed to get great results. I have seem a lot of 288s period charts that were very knarly looking yet guided very well, so having a very low PE and a fairly sinusoidal yet a bit zaggy partial should do ok. Test, test test is the best way to find out. 

Hi Kevin,

How can you tell that the newer one has a period length of 288s and the older one 434s? 

 

Jerry


Edited by jerrygerber, 10 December 2023 - 01:42 PM.


#11 Kevin_A

Kevin_A

    Surveyor 1

  • *****
  • Posts: 1,737
  • Joined: 22 Aug 2018
  • Loc: Belmont, Ontario Canada

Posted 10 December 2023 - 03:32 PM

Hi Kevin,

How can you tell that the newer one has a period length of 288s and the older one 434s? 

 

Jerry

You can tell by the amount of degrees between the repeating peaks on the partial. If it repeats at 1.2 degrees it is a 288s period and every 1.8 degrees on the partial then it is a 432s period. Each lined segment is 0.6 degrees.



#12 jerrygerber

jerrygerber

    Mariner 2

  • *****
  • topic starter
  • Posts: 259
  • Joined: 07 Oct 2022
  • Loc: San Francisco, California, USA

Posted 10 December 2023 - 03:34 PM

You can tell by the amount of degrees between the repeating peaks on the partial. If it repeats at 1.2 degrees it is a 288s period and every 1.8 degrees on the partial then it is a 432s period. Each lined segment is 0.6 degrees.

Thanks Kevin, that helps me understand better.


  • Kevin_A likes this

#13 arbit

arbit

    Apollo

  • -----
  • Posts: 1,401
  • Joined: 19 Feb 2012

Posted 11 December 2023 - 01:05 AM

Hi Arbit,

How do I calculate the max slope of the curve?  I've heard that's important for understanding Max and Min RA and DEC duration (which I usually set around 150-250ms) but I don't how to calculate that..

The actual procedure is explained at https://bbs.astronom...h-in-new-am5/19  Further down there are a few posts discussing how to use software instead of a manual measurement.

 

Strongly recommend you go through that thread, as also https://bbs.astronom...nce-from-my-am5


Edited by arbit, 11 December 2023 - 01:10 AM.

  • cougarone likes this

#14 jerrygerber

jerrygerber

    Mariner 2

  • *****
  • topic starter
  • Posts: 259
  • Joined: 07 Oct 2022
  • Loc: San Francisco, California, USA

Posted 11 December 2023 - 10:41 PM

The actual procedure is explained at https://bbs.astronom...h-in-new-am5/19  Further down there are a few posts discussing how to use software instead of a manual measurement.

 

Strongly recommend you go through that thread, as also https://bbs.astronom...nce-from-my-am5

Thank you Arbit!




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics