Back to forums

CEM70 guiding curve

iOptron CEM70 Open PHD Guiding Project PHD2 iOptron CEM70-NUC Equatorial Mount Guiding
20 replies•1.6k views
Götz Golla avatar
Götz Golla avatar
Hello there,
I have asked this question already in another post, put the session got sort of hijacked, so I try it again here.
I am using PHD2 for guiding my CEM70 mount.  phdlogview is able to calculate a nice raw guiding curve without corrections except for drift. For my telescope this curve has an amplitude of more then 10arcsec. On the other hand, the PE curve provided by IOptron for my mount has an amplitude of about 3 arcsec. I would think that the two curves should be at least similar. Does anyone has an explanation why this is not the case ? I contacted IOptron support but did not get an answer - again.

Included are the two curves, first the one provided by IOptron:



Second the one measured with PHD2/phdlogviewer:



Any idea or an explanation is welcome.  Thanks a lot !
Götz
Engaging
Yuxuan avatar
Yuxuan avatar
I know that iOptron measures the PE on an encoder without any actual payload. I'm not sure whether they directly measure it on the motor, or on the gear, but at any rate, I don't think the reported PE necessarily imply real world performance.

I would also inquire with John at iOptron support. They are very responsive and helpful from my experience.
andrea tasselli avatar
andrea tasselli avatar
1. What encoder were they using (i.e., what is its resolution)? What was the setup? I suppose no answers to these questions. I suspect there is a lot more of what they DON'T SAY than of what they say. If they say anything at all…

2. Yours is the real thing, with an actual telescope and kit, which is rather different from a (unknown precision) test on the mount head in a room. BTW, your period is around 20". The only issue I have is that the period isn't 348 sec. Are they using intermediate gears, planetary gearbox or such?
GergoB avatar
GergoB avatar
Götz Golla:
Hello there,
I have asked this question already in another post, put the session got sort of hijacked, so I try it again here.
I am using PHD2 for guiding my CEM70 mount.  phdlogview is able to calculate a nice raw guiding curve without corrections except for drift. For my telescope this curve has an amplitude of more then 10arcsec. On the other hand, the PE curve provided by IOptron for my mount has an amplitude of about 3 arcsec. I would think that the two curves should be at least similar. Does anyone has an explanation why this is not the case ? I contacted IOptron support but did not get an answer - again.

Included are the two curves, first the one provided by IOptron:



Second the one measured with PHD2/phdlogviewer:



Any idea or an explanation is welcome.  Thanks a lot !
Götz

Go to the merridian, calibrate then run Guide Assistant. Let’s see that curve as well. What’s your total RMS if you try guiding? iOptron will respond, just analyze the problem better.
Götz Golla avatar
Götz Golla avatar
Thanks for the answers. @Yuxuan has explained the smaller period nicely, its 1/3 of 348sec. There is also a period of 1/2 of 348sec in the frequency analysis/fourier transform. Thats all OK.

I havent seen any reply from [email]support@ioptron.com[/email] yet. This is already the second time this happens. Is there an alternative address I could write to ?

Thank you all.

Götz
Yuxuan avatar
Yuxuan avatar
Götz Golla:
Thanks for the answers. @Yuxuan has explained the smaller period nicely, its 1/3 of 348sec. There is also a period of 1/2 of 348sec in the frequency analysis/fourier transform. Thats all OK.

I havent seen any reply from [email]support@ioptron.com[/email] yet. This is already the second time this happens. Is there an alternative address I could write to ?

Thank you all.

Götz

[email]tech1@ioptron.com[/email], this is the email address John replies from in my case.
Jon Hanson avatar
Jon Hanson avatar
@Götz Golla , I have an iOptron CEM40 and its real world performance is very close to the encoder measured PE curve from the factory.    I've included the PE curve that came with my mount and the PHD2 Log Viewer analysis from a recent imaging session below so you can compare.

If you want to generate a PE curve from your PHD2 log data you can use a free program called EQMOD PECPrep.  PECPrep can import PHD2 log data and generate a smoothed PE curve.  It works best when you have enough log data to cover ~10 cycles of your worm gear.

This first graph shows the factory PE curve overlayed with a PE curve generated by PECPrep based on my PHD2 log from the same imaging session as the analysis below.  The max error for both is ~3 arcsec.

For reference, my overall guiding for this session was 0.77" RMS which is pretty typical for me--although when seeing is good I can get as low as 0.6".  On this night my polar alignment error was ~0.8'.








Well Written Helpful Insightful Engaging Supportive
Götz Golla avatar
Götz Golla avatar
Hi Jon et. al.,

I tried to use PecPrep but failed. It analyzed my PHD2 Logfiles, but uses only a very small window which cannot be resized. Maybe PecPrec can do more than phdlogviewer, but the phdlogviewer can do a spectrum analysis as well.
Last night I did some imaging tests. Again frequency analysis shows a pronounced peak at the 3th harmonic of the worm gear. PHD2 is not quite able to correct it, no matter what initial or fixed period I enter in the setup. On the other hand, the results are not bad, total RMS is 0.6arcsec

What bothers me a bit are the sudden deviations of more than 3arcsec in RA. They always are negative, as if something is slowing down the mount for a short moment. Without them, results would probably be excellent. Could it be due to some dirt or other hardware problem in the gears of the mount ? The telescope is in a small observatory building and the night was calm.

In include an example of the guiding curve here. The peaks I mean are at about 15min, 17min, 32min and 38min;
Helpful Insightful Respectful Engaging
andrea tasselli avatar
andrea tasselli avatar
Possibly something tugging the mount. In my cases I've found to be stiff cables.
Sean van Drogen avatar
Sean van Drogen avatar
andrea tasselli:
Possibly something tugging the mount. In my cases I've found to be stiff cables.

agree with andrea that looks like something snags once in a while, had a clear night last night and just wanted to share the analysis of that guide log with you. Seeing was good but not excellent.





and my factory measurement is somewhere in a box but is not that much different from yours. Think the RA can probably be improved by mesh changes but at this point i dont feel the need.
Götz Golla avatar
Götz Golla avatar
Just got the message from John at IOptron support that the second harmonic in the guiding curve might be due to a loose belt in RA. He send a lot of documentation how to disassemble the whole thing and tuning it. This will be something for some cloudy nights soon. I will let you know about the results.

Also, there is one cable attached to the telescope. Its not stiff. But I also noted once that even the slightest touching of the mount is already visible in the guiding curve. Thats maybe because the TriPier 360 I am using has fairly soft rubber feet. I will replace them as well soon.
Helpful
Nick Ambrose avatar
Nick Ambrose avatar
Götz Golla:
Just got the message from John at IOptron support that the second harmonic in the guiding curve might be due to a loose belt in RA. He send a lot of documentation how to disassemble the whole thing and tuning it. This will be something for some cloudy nights soon. I will let you know about the results.

Also, there is one cable attached to the telescope. Its not stiff. But I also noted once that even the slightest touching of the mount is already visible in the guiding curve. Thats maybe because the TriPier 360 I am using has fairly soft rubber feet. I will replace them as well soon.

Any chance you can share that doc ? I checked my DEC belt and it's tight (could be too tight) but RA looks harder
Götz Golla avatar
Götz Golla avatar
Here is the doc, its public:
https://www.ioptron.com/v/Support/CEM70_Belt_Tension_ADJ.pdf
https://www.ioptron.com/v/Support/CEM70_RAPlay_Fixing.pdf

In case of my mount I found that the belt is much too tight. John said that they recommend the belt can be pushed down 3-4mm - which is a lot. Its difficult to measure, but I think I have moved from 1mm to about 3mm now. The result is that the sound of the mount moving is now much different because the vibrations of the stepper motor are no longer transmitted via the belt.
Loosening the belt should lead to more hysteresis. This is not relevant in RA but in DEC. I will see what the PHD2 guiding assistant will say on the next clear night.
Nick Ambrose avatar
Nick Ambrose avatar
Götz Golla:
Here is the doc, its public:
https://www.ioptron.com/v/Support/CEM70_Belt_Tension_ADJ.pdf
https://www.ioptron.com/v/Support/CEM70_RAPlay_Fixing.pdf

In case of my mount I found that the belt is much too tight. John said that they recommend the belt can be pushed down 3-4mm - which is a lot. Its difficult to measure, but I think I have moved from 1mm to about 3mm now. The result is that the sound of the mount moving is now much different because the vibrations of the stepper motor are no longer transmitted via the belt.
Loosening the belt should lead to more hysteresis. This is not relevant in RA but in DEC. I will see what the PHD2 guiding assistant will say on the next clear night.

Thank you. Definitely interested in your results
Götz Golla avatar
Götz Golla avatar
OK, so right now I am having a fantastic clear and dry night. NINA is busy taking images of SH2-124. Last night I tuned PHD2 as much as I can. Tonight the real imaging has started. What I can say:
  • RA guiding improved a lot by easing the belt tension. RMS got down from 0.75" to 0.35". Particularly the big spikes are gone now
  • DEC guiding was excellent as well before meridian flip (RMS 0.2"), so that the total RMS always was below 0.5".
  • Unfortunatelly, DEC guiding got much worse after meridian flip in the first hour, RMS for DEC alone above 0.5"
  • One hour after meridian flip situation has improved, DEC RMS is 0.32", total RMS 0.52"

It would be very interesting to understand why a meridian flip would influence guiding behavior in DEC. The mount is very well balanced, maybe too much ?

I should mention that the guiding assistant determined a huge DEC backlash of 3300ms.

My resumee is:
  • The recommendation of IOptron to have a belt tension of 3mm (whats the right English word for it?) is a good one.
  • The tension in DEC should probably be higher, I would go for a push distance of 1-2mm, as it was before
  • I will try the latter tomorrow night and let you know it the theory was right.

Have fun !
Goetz
Helpful
Nick Ambrose avatar
Nick Ambrose avatar
Götz Golla:
OK, so right now I am having a fantastic clear and dry night. NINA is busy taking images of SH2-124. Last night I tuned PHD2 as much as I can. Tonight the real imaging has started. What I can say:
  • RA guiding improved a lot by easing the belt tension. RMS got down from 0.75" to 0.35". Particularly the big spikes are gone now
  • DEC guiding was excellent as well before meridian flip (RMS 0.2"), so that the total RMS always was below 0.5".
  • Unfortunatelly, DEC guiding got much worse after meridian flip in the first hour, RMS for DEC alone above 0.5"
  • One hour after meridian flip situation has improved, DEC RMS is 0.32", total RMS 0.52"

It would be very interesting to understand why a meridian flip would influence guiding behavior in DEC. The mount is very well balanced, maybe too much ?

I should mention that the guiding assistant determined a huge DEC backlash of 3300ms.

My resumee is:
  • The recommendation of IOptron to have a belt tension of 3mm (whats the right English word for it?) is a good one.
  • The tension in DEC should probably be higher, I would go for a push distance of 1-2mm, as it was before
  • I will try the latter tomorrow night and let you know it the theory was right.

Have fun !
Goetz

Good update. I may bite the bullet and try to mess with the RA belt. Or I may find someone locally I can pay/bribe to help me so I don't make it worse !
Götz Golla avatar
Götz Golla avatar
@Nick Ambrose : The tricky part of the RA belt adjustment is during reassembly. You need to be really careful not to jam the cables. It is very well described in IOptrons manual, and you really need to follow it to the word. Otherwise its easy.
Götz Golla avatar
Götz Golla avatar
Hello ! I forgot to mention one important thing: Adjusting the belt tension in RA, as suggested by IOptron, did not change at all the PE curve with a period of 115sec and amplitude of >8arcsec for my mount. It looks like this is a (comparably) massive effect which will not go away with a little tuning here and there.

However one more thing John of IOptron suggested was a loose worm bearing end cap. I did not check this  last time I opened the mount since it is more difficult to do. Looks like work for another cloudy weekend.
Helpful
D. Jung avatar
D. Jung avatar
What tripod/pier are you using?
Götz Golla avatar
Götz Golla avatar
Some news about this topic:

I found a thread at CloudyNights about this:
https://www.cloudynights.com/topic/821249-ioptron-cem70-tracking-and-guiding-troubles/

Someone had the same weird PE Curve. Let me briefly summarize the discussion and result:
- a PE Curve of that amplitude cannot be a matter of tuning, it must be a more severe problem
- the fact that the curve is very regular and periodic points at a single big problem and not a mixture of small problems

Finally the owner send the mount back to IOptron and got a bearing of the RA worm replaced. The mount was fine afterwards.

Since I am not into sending the mount back to the US, I just bought two new bearings of the right type in a local online shop - they will be delivered tomorrow. I have received an internal manual from IOptron how to replace the bearing. The process is also nicely documented in a video by a russian guy (Ivan) here: https://youtu.be/s91IEXF6tuU

I will go for it asap. Wish me luck.

Goetz
Helpful Insightful Engaging Supportive
Götz Golla avatar
Götz Golla avatar
Hi folks,
 
here comes another update on the guiding situation with my CEM70. Equipped with a new grease and new bearings I disassembled the RA worm and gear completely and intended to replace the bearings of the worm and re-grease everything. 

What I found is that the old worm bearings were in better shape than the new ones. Moving the worm was perfectly easy and smooth with the old bearings, and not so much with the new ones - so I put the old ones back in.

Otherwise I didn't find any problems. It was a good feeling, though, to have seen and touched every part and to have verified that it all is put together perfectly.

The first guiding test tonight revealed that the guiding curve with the 115sec period and about 8arcsec amplitude did not improve at all. This means that the reason for this weird behavior is still unexplained and not related to the worm and the worm bearings.

There is one more possible explanation. The number of teeth of the worm pulley is 45, the number of teeth of the pulley on the shaft of the stepper motor is 15. The worm period is 345sec. This means that problems with the stepper motor shaft and pulley would have a period of 345/3 = 115 sec !!!

So I think that the origin of the guiding curve problem must be the stepper motor. It could be a bent shaft or maybe some internal problem.

So which amount of bent are we looking for ? Lets say 2x the amplitude of the curve is 20arcsec. One revolution of the big RA gear is 360deg in 24h.

One revolution of the worm than is (360*60*60)/(24*60*60) * 345sec = 15 * 345 = 5175arcsec = 1.4375deg on the sky.

One revolution of the stepper motor shaft corresponds to 5175/3 = 1725 arcsec .

20arcsec then corresponds to 0.0116 revolutions of the stepper motor. This is just a fraction of about 6 of one tooth of the pulley, or roughly one tens of a millimeter.

The conclusion is that if the shaft of the stepper motor is bent by only 1/10mm, this would already explain the shape of the guiding curve. This sounds scary to me. 1/10mm is difficult to measure. Do we know the specifications of the motor ? Or maybe I made a mistake in my calculation ? Please let me know what you think.

Goetz
Helpful Insightful Engaging
Related discussions
DISCONNECTION WITH IOPTRON COMMANDER
HELLO EVERYONE I have an ioptron cem 120 EC2. Randomly the ioptron commander application disconnects saying that it can't find any devices anymore. Looking in the USB peripherals I see the mount connected. The only way to reconnect the applicatio...
May 22, 2025
Both posts discuss technical troubleshooting issues with iOptron telescope mounts and their associated software or performance problems.