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

CEM60EC problems

  • Please log in to reply
263 replies to this topic

#1 OzAndrewJ

OzAndrewJ

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 30 Nov 2010

Posted 04 August 2020 - 09:07 PM

Gday All

I have been working with Ben Dan on his 60EC problems.

He was going to tag onto the current 120EC thread but now we have his results, it is probably better to start a new thread.

Basically, he was getting eggy stars at times, and on analysing his data, it showed a really odd error at about 17x the worm period.

This didnt line up with anything known, but then i noticed his frame rate was close to the SDE rate for this mount

and my simulator indicated that for his given frame rate, SDE would show at about 17x  grin.gif

I then suggested he needed to get a high framerate dataset to confirm or discard the theory.

He then practiced doing slow calibrations followed by high speed unguided data grabbing

and his results show some of the best/clearest SDE to date.frown.gif

Have attached a plot of his RA and DEC via PHD viewer so you can compare the DEC trace ( which indicates seeing )

and the RA trace, that shows "visible" SDE of about 2 arcsec pk-pk.

The PEMPro plot merely confirms this ripple is 54x the worm period, ie the known SDE frequency.

 

IIRC, other 60EC users with clear SDE tried different firmwares over time to try and reduce it,

but there are so many threads now i cant make sense of all the replies confused1.gif

Sooooo, is anyone out there using a 60EC that had bad SDE, that is now "better"

and if so, what firmware are you running.

Andrew Johansen Melbourne Australia

SDE Raw test.jpg SDE FFT.jpg


Edited by OzAndrewJ, 04 August 2020 - 09:38 PM.

  • Dynan likes this

#2 gotak

gotak

    Vanguard

  • *****
  • Posts: 2,418
  • Joined: 18 Sep 2016
  • Loc: Toronto, CA

Posted 05 August 2020 - 07:41 AM

The firmware never helped. There was one which disabled guiding and encoders which did helped but it wasn't what anyone wants. Simply put iOptron has zero software solution.

2" peak is more than what I have. This one seems a clear RMA to me. For my mount it's just around 1" peaks but works out to something like .50" ish RMS so most of the time the stars are round enough.

Edited by gotak, 05 August 2020 - 07:42 AM.

  • RossW likes this

#3 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 05 August 2020 - 10:49 AM

My mount was purchased here on CN, so RMA is out of the question.

 

 

The firmware never helped. There was one which disabled guiding and encoders which did helped but it wasn't what anyone wants. Simply put iOptron has zero software solution.

Was there a firmware that disabled the encoders but still allowed guiding? Did this disable the guiding of PHD2 completely?



#4 gotak

gotak

    Vanguard

  • *****
  • Posts: 2,418
  • Joined: 18 Sep 2016
  • Loc: Toronto, CA

Posted 05 August 2020 - 11:40 AM

No they never released a fw with the encoder disabled. We found in the cem120ec examples of copies with very large native PE. Likely why iOptron is unwilling to allow disabling the encoders.

On the cem120 three are solder bridges you can undo that will make the FW thinks it's a regular CEM. You can try and check for that.

Sorry to hear you have such a bad result. The seller didn't know about this excessive SDE?

You should contact iOptron to see what options you might have if still under warranty or even out of warranty. Likely you are up the creek without means but doesn't hurt to ask.

One reason why I find it hard to recommend this brand. If they took care of people and stood behind their kit it's fine. But we have had an impossible situation with the after sales support. For me I even asked about SDE and was assured there was no longer any issue by iOptron support before I buy, so once burnt twice shy.

Edited by gotak, 05 August 2020 - 11:43 AM.

  • Dynan and RossW like this

#5 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 05 August 2020 - 06:22 PM

Does anyone know (or have a guess) as to hat would happen if the encoder was simply unplugged from the stepper motor?

 

Probable board failure I would think.

 

This is really heartbreaking...and $2K.5 wallet breaking. I was thinking of getting a CEM70, but with this kind of abandonment, I'm done with iOptron.



#6 RossW

RossW

    Mariner 2

  • -----
  • Posts: 258
  • Joined: 15 Jun 2018
  • Loc: Lake Biwa, Japan

Posted 05 August 2020 - 06:54 PM

Dynan's mount's  RA trace is exactly what I saw in my original CEM120EC2 (although 3.3s period). Looking at Dynan's RA trace I don't even need to do an FFT to know that the cause will be SDE; it is so close to perfect periodicity :-)

 

I've analysed quite a few PhD2 logs from various users of EC mounts and I have not seen a single case where firmware reduced the SDE magnitude (random variations aside). My own experience after trialling around 6 different firmwares in my broken 120EC2 was the same: SDE in iOptron encodered mounts cannot be solved by firmware. If anyone has before and after PhD2 logs with proof of SDE reduction I'd love to see them. The effort iOptron put into firmware upgrades in 2019 was to do with high-frequency oscillations, a.k.a. resonance, of around 3 to 5 oscillations per second, not SDE oscillation. They were successful in reducing resonance via firmware, according to users.

 

The only solution to such out-of-spec SDE worth trying is to get a replacement mount *within the dealer return period*, because good luck with receiving a replacement from iOptron after the return window has expired. For some reason they can't see this SDE on a test bed, and under the terms of their warranty they will only cover problems that iOptron can confirm exist.

 

An alternative is to disable the encoder, but as Gotak explained it would appear that the encoder mounts were equipped with poorer quality worms, so you may be replacing a fast (unguidable) SDE with a guidable but large periodic error. I know of one owner who measured his two CEM120EC2s as having around 40 arc-sec peak-to-peak PE when he disabled the encoders.

 

I'm sorry this happened to you Dynan; it is hard to believe the previous owner was not aware of the problem because it is so obvious when watching the PhD trace and star movement that it cannot be not noticed, IMHO.

 

Cheers,

 

Ross



#7 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 05 August 2020 - 07:00 PM

This is the reality check. I got this performance on Monday night after calibrating and running short exposure looping for Andrew's SDE test data.

 

Andrew informed me that since I was 60° away from the Equator I was reducing the SDE mechanically by about 50%:

 

That said, your bubble neb piccy ( nice )  was taken at DEC = 60deg and this effectively halves the SDE effects "on the sensor". Try imaging something closer to the equator and see what it comes out like.

 

Bubble pic thread to which he was referring: https://www.cloudyni.../#entry10395094

 

To begin the SDE test, I checked PA with SharpCap. I was pleasantly surprised since my last PA was 6/16/2020. It held rock steady...apparently any settling of my pier has completed. (I'd been watching it for over a year steadily get better.)

 

2020_0803_SC_PA_CHECK_S.jpg

 

After the calibration mentioned above, I got this guiding for almost five full hours:

 

2020_0803_GUIDING_S.jpg

 

As Andrew informed me, the test will be to imagine AT the Equator next outing. I guess I could be perfect if I imaged POLARIS for the rest of my hobby days...

 

iOptron...shakecane.gif

 

Just heartbreaking...  sigh2.gifbawling.gif

 

 



#8 OldManSky

OldManSky

    Skylab

  • *****
  • Posts: 4,418
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 05 August 2020 - 07:07 PM

Daniel, I'm a little confused -- your RA RMS in the guiding snapshot above is 0.27".  I can't imagine you're not happy with that...?

The "big" error is in DEC, which isn't an SDE issue...so...?



#9 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 05 August 2020 - 07:28 PM

Paul, Thanks for the encouragement!

 

I AM happy with the 0.27 RMS (0.48 px) result. But look at the RA Oscillation at 0.38 and total at 0.47 RMS. (I've asked before but never got an answer: Does dithering affect the final readings posted for Guide Stats?)

 

As Andrew stated, the test of quality will be my next image at the Equator.

 

During my last outing 6/16/2020, I imaged M13 and got eggs for stars. That was 28° closer to the Equator and look at the footballs for stars I got:

 

2020_0616_M13_PI_ST_1_SM.jpg

 

My heartbreak is that iOptron would obfuscate this imperfection at all. AND the previous owner, a CN member, might have known and didn't reveal. (The CN sale was listed as a CEM60 NON-EC. I was stunned to see that I got the EC model for the price of a used NON-EC. Yippee! Now I suspect there might have been some karma built up.)

 

Also, if you look back through my threads, you'll see the previous torture of fighting a BAD CGX for more than a year. I guess it primed my disappointment.

 

I think it was Siddhartha Gautama that said: "Three things come out every day, The Sun, The Moon, and The Truth."

 

Thanks again for kind words.


Edited by Dynan, 05 August 2020 - 07:36 PM.


#10 OzAndrewJ

OzAndrewJ

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 30 Nov 2010

Posted 05 August 2020 - 08:16 PM

Gday Paul

your RA RMS in the guiding snapshot above is 0.27".

This is a hidden nuance of his guide rate.

His bubble neb piccy was taken at DEC = 60deg, so effectively halved the "mechanical" axle error due to SDE.

His frame rate of around 4 seconds means the high freq ripple is still there

but you dont "see" it in the stats due to the effects of undersampling.

As a simple example, if you made up a 10 arcsec pk-pk sine wave with a period of exactly 5 seconds

then sampled its position every 5 seconds, you will get a flat line in the guider

but the main sensor will still be averaging the underlying sinewave.

 

Andrew Johansen Melbourne Australia


  • RossW likes this

#11 OldManSky

OldManSky

    Skylab

  • *****
  • Posts: 4,418
  • Joined: 03 Jan 2019
  • Loc: Valley Center, CA USA

Posted 05 August 2020 - 08:58 PM

Gday Paul

This is a hidden nuance of his guide rate.

His bubble neb piccy was taken at DEC = 60deg, so effectively halved the "mechanical" axle error due to SDE.

His frame rate of around 4 seconds means the high freq ripple is still there

but you dont "see" it in the stats due to the effects of undersampling.

As a simple example, if you made up a 10 arcsec pk-pk sine wave with a period of exactly 5 seconds

then sampled its position every 5 seconds, you will get a flat line in the guider

but the main sensor will still be averaging the underlying sinewave.

 

Andrew Johansen Melbourne Australia

Thanks for the explanation, that makes perfect sense.

Just FYI, I downloaded the M13 pic above, and tried to plate solve in ASTAP -- no joy.  My reason for doing so was to get the image angle, and see if the egg shapes were indeed in RA...or DEC.

Might try astrometry.net.



#12 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 05 August 2020 - 10:05 PM

Paul, if you need better data to do the RA/DEC search, I can upload data to Dropbox if you want.

 

Next New Moon (clouds willing) I am going to image NGC6962 for the test. It's on 19" off the Equator and my pier gets an 8 hour shot at it. I should have plenty of data.

 

I'm trying to process a larger M13 file on Astrometry at the moment. Been having trouble with Astrometry all day. I'll post the angle here if successful. Would you please then explain how you figure RA/DEC from the given angle? Much appreciated.


Edited by Dynan, 05 August 2020 - 10:09 PM.


#13 RossW

RossW

    Mariner 2

  • -----
  • Posts: 258
  • Joined: 15 Jun 2018
  • Loc: Lake Biwa, Japan

Posted 05 August 2020 - 11:27 PM

 

This is a hidden nuance of his guide rate.

 

Furthermore, the pixel scale is quite large at close to 2 arc-sec per pixel, so that helps hide a good proportion of whatever SDE remains due to the undersampling. But as Andrew mentioned the imaging sensor is still seeing the full 2 arc-sec peak-to-peak SDE. With your current scope and imaging train the SDE equates to about 1 pixel of movement so stars will not be badly affected, but if you move to a scope of longer focal length (like an f/10 SCT) you'll easily see the SDE as star elongation in the RA direction. 


  • OzAndrewJ and Dynan like this

#14 OzAndrewJ

OzAndrewJ

    Soyuz

  • *****
  • topic starter
  • Posts: 3,818
  • Joined: 30 Nov 2010

Posted 06 August 2020 - 01:16 AM

Gday Ross

 

Furthermore, the pixel scale is quite large at close to 2 arc-sec per pixel,

Good extra info, but people also need to remember the pixel scale of the guider is what may or may not hide the effects in the PHD log, but it is the pixel scale of the imaging camera that will decide if the SDE appears in the final images.

Hence why when doing "testing" of a mount, i always suggest using the smallest arcsec/pix you can get.

 

Andrew Johansen Melbourne Australia


  • rgsalinger and Dynan like this

#15 RossW

RossW

    Mariner 2

  • -----
  • Posts: 258
  • Joined: 15 Jun 2018
  • Loc: Lake Biwa, Japan

Posted 06 August 2020 - 04:27 AM

it is the pixel scale of the imaging camera that will decide if the SDE appears in the final images.

Ahh, fell for that trick again :-) Yes, you are correct of course. Dynan, what is the pixel scale of your imaging rig? If it is on the small side your mount's SDE may well be the cause of your elongated stars.


Edited by RossW, 06 August 2020 - 05:43 PM.

  • Dynan likes this

#16 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 06 August 2020 - 04:55 AM

My pixel scale for the ASI533MC-Pro is 0.814, calculated. (Astrometry.net yields 0.816).


Edited by Dynan, 06 August 2020 - 07:08 AM.


#17 cytan299

cytan299

    Viking 1

  • *****
  • Posts: 803
  • Joined: 24 Nov 2014

Posted 06 August 2020 - 05:35 AM

My pixel scale for the ASI533MC-Pro is 0.814, calculated. (Astrometry.net yields .816).

 

In another thread I will propose a cure for this, if one is possible by changing the code and NOT physically/mechanically limited. I will invite iOptron to step up. This website is becoming quite well regarded in the astronomy community, so using its popularity may be to CEM60-EC owners' advantage. It will be a leverage of buying power...no threat of any kind. (I'm ready to replace this mount, possibly with a CEM-70). And just might be what is needed to cure a (relatively) wide-spread feeling of abandonment.

Good luck with what you plan to do. Why would iOptron care? Even with hundreds of threads on buyer beware when buying iOptron mounts, whether it is EC or not, I don’t believe any of the cautions have sunk in. Exhibit A: even after this experience you are still thinking of going the iOptron route!

 

And just for the uninitiated, here’s my advice: Buying iOptron is a crapshoot. Star test as soon as you can. If it doesn’t work, send it back. Once it’s out of the no questions asked return period, you are SOL.

 

As the saying goes, “fool me once shame on you, fool me twice shame on me”. And this is coming from me, who got burnt twice with brand new iOptron mounts (zeq25 and cem60)

 

cytan


Edited by cytan299, 06 August 2020 - 05:44 AM.

  • Dynan likes this

#18 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 06 August 2020 - 07:13 AM

And just for the uninitiated, here’s my advice: Buying iOptron is a crapshoot. 

Thanks, Cytan. You have helped solidify my commitment to a different mount manufacturer. There will be no more shooting of crap by me.



#19 chrisastro8

chrisastro8

    Ranger 4

  • *****
  • Posts: 323
  • Joined: 25 Sep 2010
  • Loc: Texas

Posted 06 August 2020 - 09:36 AM

Gday All

Have attached a plot of his RA and DEC via PHD viewer so you can compare the DEC trace ( which indicates seeing )

and the RA trace, that shows "visible" SDE of about 2 arcsec pk-pk.

The PEMPro plot merely confirms this ripple is 54x the worm period, ie the known SDE frequency.

 

Isn't this just confirming what iOptron already provide to every owner of EC model (40 60,120)?    It is what iOptron describe as Real Time PEC and it works as intended for me. Perhaps I'm missing something since I'm not seeing what the problem is.  The analysis is most certainly interesting. 

 

Attached was provided by iOptron, I hope you like the sticky note covering up the Serial# of my mount:-)   

Attached Thumbnails

  • 20200806_091657.jpg


#20 chrisastro8

chrisastro8

    Ranger 4

  • *****
  • Posts: 323
  • Joined: 25 Sep 2010
  • Loc: Texas

Posted 06 August 2020 - 10:04 AM

 The effort iOptron put into firmware upgrades in 2019 was to do with high-frequency oscillations, a.k.a. resonance, of around 3 to 5 oscillations per second, not SDE oscillation. They were successful in reducing resonance via firmware, according to users.

I wanted to lay support to this comment and can confirm that with my CEM40 EC there was sometimes a resonance problem that is now fixed through the March 2020 released firmware.  The resonance was not there with smaller tubes but became apparent with larger loads especially when the counterweight was further down the shaft.  I suspect this is why the newly released CEM40G and the CGEM45 prior, have thicker stubbier counterweight shafts that are the same as the CEM60.  



#21 gotak

gotak

    Vanguard

  • *****
  • Posts: 2,418
  • Joined: 18 Sep 2016
  • Loc: Toronto, CA

Posted 06 August 2020 - 11:34 AM

Isn't this just confirming what iOptron already provide to every owner of EC model (40 60,120)? It is what iOptron describe as Real Time PEC and it works as intended for me. Perhaps I'm missing something since I'm not seeing what the problem is. The analysis is most certainly interesting.

Attached was provided by iOptron, I hope you like the sticky note covering up the Serial# of my mount:-)

We all got perfectly ok plots from the factory. If they can detect the issue they should have resolved it. No they seem to believe their own alternate truth on SDE for the most part.

Btw there are copies of cem40ec with SDE as well. And it looks pretty much the same as the 60 when they do have it.

Edited by gotak, 06 August 2020 - 11:36 AM.

  • RossW likes this

#22 gotak

gotak

    Vanguard

  • *****
  • Posts: 2,418
  • Joined: 18 Sep 2016
  • Loc: Toronto, CA

Posted 06 August 2020 - 11:38 AM

About the only solution for SDE is get a better copy, get another mount without encoder, or get AO and try to correct the SDE using AO.
  • RossW likes this

#23 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 06 August 2020 - 12:58 PM

I just asked iOptron if they would take the CEM60-EC in trade for a CEM70. We'll see what they think of their gear.hmm.gif


Edited by Dynan, 06 August 2020 - 02:19 PM.


#24 rgsalinger

rgsalinger

    Fly Me to the Moon

  • *****
  • Moderators
  • Posts: 6,991
  • Joined: 19 Feb 2007
  • Loc: Carlsbad Ca

Posted 06 August 2020 - 01:55 PM

Keep on iOptron until they take it back. It took me quite a while but I have an RMA in hand from them for my CEM120EC2 even though the warranty has expired. I was finally able to remove the mount last week and put it on a different pier with one of my refractors. I plan one last test of it this weekend with a borrowed Edge 11HD to see if I want to keep it or send it back.

 

Ozandrew graciously helped me determine that I have .3 arc seconds of SDE. That sounds OK but at the image scale I use it isn't. I get good and bad results - stacking/dithering, etc cures a lot of problems. When I measure the stars in PI, I get around .6 on average. Not good enough for my taste in a 7k mount. By putting the same telescope on my other "premium" mount, suddenly the stars measure around .4 eccentricity in PI. Much better but guiding is worse - around .4 arc seconds in each axis. 

 

I would point out that using all three of my refractors (even the AP155 with about 40 pounds on the mount) the mount performs as well as anything on the market. I see .2 to .3 arc seconds of RMS guiding error night after night. I think that's because the image scale completely hides the SDE but I've never been sure. When I image at 1"/pixel that .3" of SDE is small compared to the size of the stars. When I image at .3"/pixel, it's an entire ***** pixel! Still, the 3.3 second resonance does not explain why short exposures (1 second or less) still show oblong stars with the mount.  

 

One other thing -----

 

As an well established (at least in my own home) skeptic I did not find the evidence (that I saw) of a poor worm very convincing in the "proof" that was presented to me. What was presented purported to be a video of a CEM120EC2 with the encoders disabled by unsoldering the bridge that connects them to the RA board. There are a number of things wrong with testing that way. First, no one actually knows what disabling the board does when running the firmware for the mount. Second, if you want to document something then use software like PEMPRO so that there's little or no chance of user error with the test. So, until I see something better than this, I'm not a buyer.

 

Having said that my number one wish is that the iOptron tech guys would give me a way to turn off the encoders. In line with my point of view, I think that they don't not because the worms are bad but rather because the firmware can't cope with that use case. I have no proof of this but it just makes more sense than sending out a worm with 40" of PE and then lying about it to a customer. 

 

Rgrds-Ross


  • chrisastro8 likes this

#25 Dynan

Dynan

    Skylab

  • *****
  • Posts: 4,267
  • Joined: 11 Mar 2018
  • Loc: NOLA

Posted 06 August 2020 - 03:11 PM

Thanks for the support, Ross. Unfortunately I bought it used here on CN. With 2 seconds pk to pk SDE, this CEM60-EC is a low grade paper weight!

 

iOptron hasn't replied to my trade-in email request, so I think I know where this is going.

 

You said:

Not good enough for my taste in a 7k mount.

If you have an RMA, why not return it and get an AP or Paramount? Mount it and never look back!

 

Having said that my number one wish is that the iOptron tech guys would give me a way to turn off the encoders.

I would hope that would cure the problem for CEM60-EC owners. CEM60 NON-EC owners have no problem with SDE ruining their imaging.

 

I gave my CGX away after fighting it for a year...and throwing $1k worth of ceramic bearing into it. Looks like I might chuck this heartbreaker into a nondescript dumpster too. I cannot sell this with a clear conscience. If I do, it'll be total honesty, leading with the defect Andrew has so skillfully revealed. What can I get...$500? $1k? $1.5k?

 

I'm not into anger, revenge, or obsession. But I will at every opportunity tell the truth of what iOptron has done to many loyal customers. Caveat Emptor! (Buyer Beware)

 

Any decent alternatives to iOptron CEM60-70) jump to the $6k range (with no $$$ help from a rig I can't sell).

 

Just heart breaking...again. iOptron...this WILL affect the company, eventually, monetarily, ethically.




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