12 replies to this topic

[MikeECha]

Hello there,

 

I was wondering if someone with these apps has noticed what I did as described below. And if so, why is there a difference?   

 

I just discovered that there is a difference in the HA of Polaris for my location at a given time on the iOptron app vs the Polar Scope Align Pro (PSAP) app. The difference changes with time. Sometimes is very small (1 minute or 2) to more than 20 minutes at times.

 

First, I have noticed that after performing a very accurate Static Polar Align (SPA) procedure (<1 degree PA error on both, az and alt) in PHD2, I was always off WRT the PSPA app by some amount. I attributed that to the polar scope reticule not being perfectly straight up and aligned with the RA axis. But I had a chance to check it and it was not that bad so I made it as correct as I could to an almost perfectly concentric rotation. Then, last night again, I was off.

 

I checked my location settings on the app and the iOptron Commander ASCOM driver and both match. The time is synced from the internet on both my laptop and my phone so no a chance for the difference there, The location is taken from the app and entered manually on the controller. Same values checked here too.

 

Then, I checked what the iOptron commander shows as Polaris position and that is where I found the difference. For the same location and time, the app gives me different HA value. For a tiebreaker, I purchased the iOptron app and it matches the commander. I guess I should have used the manufacturer app but at the time I purchase the other app I did not know it. 

 

In any case, I like the PSAP app for the information it provides in just one app and I would like to keep/fix it if possible.

 

Clear skies.

 

 

 

 

[jkcolli]

Mike

I just noticed the same problem. I have seen 15 to 20 min difference between the iOptron app and the Polar Scope Align Pro app.

 

Did you resolve this issue?

 

 

Jack Collins

[MikeECha]

JK,

 

No, I could not figure it out. And I am still curious about what the difference is but, since I started to use Sharpcap for polar alignment I use the apps just for rough alignment. SC get is spot on in just a few minutes. As we speak my modest iOptron Smart EQ Pro+ is tracking at 1.8" RMS and I can do 4-5 min subs with no trails.

 

If you ever find out please, let me know.

 

Cheers!

[ecuador]

Hi guys, you could have just asked me

Anyway, one of the reason I made Polar Scope Align was because the iOptron app was doing the calculation wrong. They were redoing their algorithm in various versions, but they seemed to get it wrong every time. I haven't looked at what they have been doing for quite a while now, but I thought I'd check it out since you mentioned it. Pleasant surprise, they now claim to do the full calculation (aberration, nutation, refraction etc) like Polar Scope Align!  Yay?

Eh, no, not at all it seems... 

Let's take it slow. Trying the current version 7.40 on iOS 12. So, I can't change coordinates on the iOptron app to experiment like I can on Polar Scope Align, but at least I can change the phone time. Good news first, for my location, the apps have 10 secs or less difference in the upper/lower culmination times:

They are within a few seconds and agree with USNO, which is giving the 11:28 and 23:26 culmination times.

 

Immediately though we see the first problem: you will note that iOptron gives 39.5 arcmin as the Polaris - NCP distance for both upper and lower culmination. Given that the refraction at 53 degrees latitude is about 0.76 arcmin [R = 1.02/tan(h+10.3/(h+5.11)) - Meeus], you'd expect the difference to be twice that, sort of like Polar Scope Align's 40.3' vs 38.8'. Amusingly, given how it misses the calculation here, the iOptron app claims to adjust for barometric pressure and temperature, so given the iOptron reported 15.2C and 1007.8 hPa, we need to adjust our calculated value by... 3% [P/1010 * 283/(273+T) - Meeus] giving 0.75 arcmin. Which should tell you why I don't try to adjust for that in PS Align - it doesn't matter much.

Anyway, so far we established the iOptron app gets culmination times correct, so they do include at least nutation/aberration now, but they broke refraction (which was working a few years back). However, when I first noticed issues with the app, it was not at the 0/3/6/9 HA points, it was some weird bug when far from those. There are 11h 58m between each culmination point (half a sidereal day), let's divide this interval by 4 and we get 2h 59.5m, so we expect 1h30m HA (or 3h in the 24-notation) to be covered on each such interval. Refraction would mess it up a bit, but since the iOptron app calculates it at under 0.5', so practically zero, we just disable refraction from the PS Align advanced settings and see what we get:

Yup, the Earth, bless it, rotates at a constant angular speed, so Polaris moves exactly 1/8th of a circle every 1/8th of a sidereal day. Without refraction distance is indeed 39.5' from NCP.

Let's go to the iOptron app (the middle screen is 1.5m earlier to illustrate when 3:00h HA is hit):

And this is the facepalm. When it is not around the 4 cardinal points, it is off by over 30m HA (that's over 1h in the more common 24h notation). In the first 3 hour interval, Polaris seems to manage to cover 60 degrees, then in the next ~3 hours it does only 30 degrees, then another 30 degrees in 3 hours, to speed up to 60 degrees again to reach lower culmination. At those midpoints, the iOptron app gives you the location Polaris would be 1 our before or after...

 

Am I the only one seeing this? I mean it was similar to what I had noted years ago (different app version, device, iOS), I can't believe it is still not fixed if it's like that for everyone...

Edited by ecuador, 31 May 2019 - 06:51 PM.

[Der_Pit]

Ooops....

 

One question in that context:  Do you know if the iOptron app uses the same algorithm(s) as the one on the 8407+ handcontrollers?  'cause that's all I have, and what I am using so far....

[Der_Pit]

Immediately though we see the first problem: you will note that iOptron gives 39.5 arcmin as the Polaris - NCP distance for both upper and lower culmination. Given that the refraction at 53 degrees latitude is about 0.76 arcmin [R = 1.02/tan(h+10.3/(h+5.11)) - Meeus], you'd expect the difference to be twice that, 

Ehm - no?.

Yes, the total refraction is 0.76 arcmin, as you write.  However, you are only interested in the differential refraction between the NCP and polaris position, i.e., the difference for 53⁰ ± 39.5', which is ±0.018' or 1.1 arcsec....

[fmeschia]

 

iOptron2.png

And this is the facepalm. When it is not around the 4 cardinal points, it is off by over 30m HA (that's over 1h in the more common 24h notation). In the first 3 hour interval, Polaris seems to manage to cover 60 degrees, then in the next ~3 hours it does only 30 degrees, then another 30 degrees in 3 hours, to speed up to 60 degrees again to reach lower culmination. At those midpoints, the iOptron app gives you the location Polaris would be 1 our before or after...

 

Am I the only one seeing this? I mean it was similar to what I had noted years ago (different app version, device, iOS), I can't believe it is still not fixed if it's like that for everyone...

I can replicate this only partially. I set my iPhone for the same date/times you listed. At 17:25 Polaris is reported at 3 hours, at 14:25 it was reported at 4 hours 15 minutes, and at 20:25 it was at 1 hour 42 minutes. Looks like there’s some wrong logic but more complicated than just a fixed offset. 

 

 

Francesco

[ecuador]

Ehm - no?.

Yes, the total refraction is 0.76 arcmin, as you write.  However, you are only interested in the differential refraction between the NCP and polaris position, i.e., the difference for 53⁰ ± 39.5', which is ±0.018' or 1.1 arcsec....

No, that's not how it works.

You are trying to get your Polar scope to aim at the true NCP. Atmospheric refraction causes the NCP/Polaris region in your Polar Scope to appear higher than it would if there was no atmosphere, so you have to aim lower by the atmospheric refraction amount to account for that.

 

Technically, if you are imaging something low in the horizon, the refracted NCP would give you better unguided results, but since we tend to select targets high in the sky it is common to align to the real NCP.

 

The difference you mention would be useful if you wanted to aim to the *refracted* NCP while accounting with Polaris' relative refraction. Which you don't need to as you note, it's too small. In the general case we want the true NCP.

Edited by ecuador, 01 June 2019 - 04:45 PM.

[ecuador]

I can replicate this only partially. I set my iPhone for the same date/times you listed. At 17:25 Polaris is reported at 3 hours, at 14:25 it was reported at 4 hours 15 minutes, and at 20:25 it was at 1 hour 42 minutes. Looks like there’s some wrong logic but more complicated than just a fixed offset. 

 

iOptron-PolarApp1.jpg

 

Francesco

Yeah, different results for you, but still quite off from what they should be... I'll have to check my hand controller to see what it does...

[Der_Pit]

No, that's not how it works.

You are trying to get your Polar scope to aim at the true NCP. Atmospheric refraction causes the NCP/Polaris region in your Polar Scope to appear higher than it would if there was no atmosphere, so you have to aim lower by the atmospheric refraction amount to account for that.

 

Technically, if you are imaging something low in the horizon, the refracted NCP would give you better unguided results, but since we tend to select targets high in the sky it is common to align to the real NCP.

 

The difference you mention would be useful if you wanted to aim to the *refracted* NCP while accounting with Polaris' relative refraction. Which you don't need to as you note, it's too small. In the general case we want the true NCP.

Thanks for the explanation ecuador.

It was a good day - I've learned something new

[ecuador]

Huh, well, I'd never actually checked my CEM25P's polaris display before and it's not looking good. Different from the iOptron app, possibly because I have not updated it (version 160523), but still weirdness:

At upper/lower culmination times I get: 

0h 1.5m / r=38.5m

6h 2.0m / r=39.9m

Which means that it does not get the culmination times as precisely as the latest app, but does add refraction, at around 0.7', which shounds OK, BUT it gets the r before refraction at 39.2' which is a bit off. So, it gives for the 17:27 timestamp:

3h 1.2m / r=39.2m

Checking the intermediate times, 14:27, 20:26:

4h 2.8m / r=39.7m

1h 59.8m / r=38.7m

So long way off, similar to what I get we the app

This is very weird, I love the mount and don't really have issues, but why do they get something as simple as that wrong? And wrong in various different ways as well?

[gmiller123456]

I know this is an old thread, but there seems to be a lot of confusion about adjusting for refraction and other variables.  It should become obvious that it is impossible to use a fixed reticle polar eyepiece to polar align to within a higher accuracy than the sum of the variables introduced by refraction, precession, nutation, aberration, gravitational deflection, polar motion, parallax, proper motion, and polar motion.

 

Lets say whoever manufactured your eyepiece went out on Jan 1 midnight UTC and measured the declination of Polaris as 89d 15' 50.8" and manufactured your eyepiece with that offset.  The problem is, today Polaris is closer to 89d 21' just due to precession.  Atmospheric refraction will add a few more minutes to that error, and nutation and aberration likely a noticeable number of arc seconds.  So, if you center Polaris in the crosshairs of your eyepiece for a given hour angle, your alignment will likely be off by several minutes.  Sometimes you'll get lucky and the errors significantly cancel each other.

 

Your scope is physically rotating about the Earth's center of mass.  And it should be obvious that effect is independent of any of Earth's atmospheric effects.  In order to counteract that effect, you want to align your scope's RA axis with the Earth's.  But, what you're really interested in, is the path the object you want to observe's path through the sky, which will be altered significantly due to refraction.

 

Due to refraction, a given object will appear above the horizon visually before it rises above the geometric horizon, and likewise still appear above the horizon even after it has set below the geometric horizon.  And the effect of refraction will change as the object's elevation above the horizon changes.  So, it should be clear that the object actually appears to move at different speeds throughout the night, and the only way to adjust for that is to have your RA axis adjust is speed to match.  So, it is not possible to adjust for refraction by imagining a fictitious refracted NCP in the sky somewhere.  And even if you could, you'd want to be adjusting for the refracted elevation of the object, not Polaris or the imagined refraction of the NCP position.

 

It is possible for the electronic polar scopes like iPolar to account for these variables.  Whether or not they actually do, I don't know, I always end up drift aligning anyway.

[ecuador]

I know this is an old thread, but there seems to be a lot of confusion about adjusting for refraction and other variables.  It should become obvious that it is impossible to use a fixed reticle polar eyepiece to polar align to within a higher accuracy than the sum of the variables introduced by refraction, precession, nutation, aberration, gravitational deflection, polar motion, parallax, proper motion, and polar motion.

 

Lets say whoever manufactured your eyepiece went out on Jan 1 midnight UTC and measured the declination of Polaris as 89d 15' 50.8" and manufactured your eyepiece with that offset.  The problem is, today Polaris is closer to 89d 21' just due to precession.  Atmospheric refraction will add a few more minutes to that error, and nutation and aberration likely a noticeable number of arc seconds.  So, if you center Polaris in the crosshairs of your eyepiece for a given hour angle, your alignment will likely be off by several minutes.  Sometimes you'll get lucky and the errors significantly cancel each other.

 

Your scope is physically rotating about the Earth's center of mass.  And it should be obvious that effect is independent of any of Earth's atmospheric effects.  In order to counteract that effect, you want to align your scope's RA axis with the Earth's.  But, what you're really interested in, is the path the object you want to observe's path through the sky, which will be altered significantly due to refraction.

 

Due to refraction, a given object will appear above the horizon visually before it rises above the geometric horizon, and likewise still appear above the horizon even after it has set below the geometric horizon.  And the effect of refraction will change as the object's elevation above the horizon changes.  So, it should be clear that the object actually appears to move at different speeds throughout the night, and the only way to adjust for that is to have your RA axis adjust is speed to match.  So, it is not possible to adjust for refraction by imagining a fictitious refracted NCP in the sky somewhere.  And even if you could, you'd want to be adjusting for the refracted elevation of the object, not Polaris or the imagined refraction of the NCP position.

 

It is possible for the electronic polar scopes like iPolar to account for these variables.  Whether or not they actually do, I don't know, I always end up drift aligning anyway.

You are mixing some things together, allow me to separate them a bit.

- Precession is not an issue with most modern polar scopes, like the one discussed on this thread. The apps calculate precession and the reticle has a grid to put polaris on - you can get within a couple of arcmins or so with the iOptron reticle (which is why I don't drift align - it takes me about 30 secs to get that close). The app that I develop, Polar Scope Align, is the only one I know that also calculates stellar Aberration and nutation, but I am fully aware (and it is noted in the extensive in-app help) that they have little impact to the actual polar alignment (for many setups even flexure has a larger effect etc), so the main reason they were added was because people would compare the results with other apps like the official iOptron that was off by 30+ arcmins at times and would doubt that PS Align was correct. So I thought if I matched the US Naval Observatory, or Cartes du Ciel exactly (well within a couple of arcsecs or so) - that would be proof enough.

- Refraction is a whole different issue and while you may not be near the equator yourself, it affects people in lower latitudes significantly. It is true that all celestial objects are affected by refraction. However, we tend to image objects that are not low in the horizon, in which case we get better tracking if we have polar aligned with the true celestial pole vs what we'd get if we ignored refraction for Polaris. Again, in the app help I explain that if you want to track objects nearer the horizon, and you are not at a low enough latitude that refraction pushes Polaris way up, it is actually better to polar align with the refracted NCP - which is why PS Align gives you that option. Importantly, drift alignment is not "magic" in this respect either, depending on how you choose the drift stars you will similarly end up with a polar alignment that is more or less affected by refraction and will be better for tracking at some elevations than others.

- There is actually one way to account for the effects of atmospheric refraction on the object you want to track. Some mounts offer special variable rate of tracking (depending on elevation) e.g. several iOptron mounts have the "King's" rate for this task. For the "King's" rate to work, you have to polar align with the true celestial pole, so you make sure you use an app that includes the refraction calculation. 

Edited by ecuador, 06 April 2021 - 03:10 PM.