RA drift on the EQ8-R mount with PHD2

Dear star friends,

For over a year, I've been trying to find the cause of the RA drift on my EQ8-R mount by using PHD2 and I need your help.

📷 20260406_220956.jpgMy EQ8-R is equipped with a mounting plate including two clamps for attaching two telescope tubes in parallel. A C14 and a RASA11 are mounted there in parallel. The total weight is certainly close to the maximum load limit, but it doesn't exceed it. Furthermore, the system is so well balanced that it's always in equilibrium and doesn't exert any uneven pressure on the stepper motors.

When I use FireCapture to guide a planet at a long focal length with the simple, uncalibrated tracking tool, the EQ8-R has absolutely no problem keeping the planet centered in the frame for hours.

However, when I use a reducer with a much larger camera and guide with PHD2, a considerable positive RA drift occurs after only 5 minutes (see first picture above), even though the guiding values according to PHD2 seem to be within a good range:

📷 20260406_220655.jpg📷 20260406_221606.jpgIn the declination direction, I noticed a backlash with the SynScan control and corrected this mechanically on the mount.

📷 20260406_221632.jpgHowever, this error had no effect on the guiding and doesn't help me with the RA problem. I haven't detected any backlash in RA, so a mechanical correction on the mount shouldn't help, right?

I'm wondering why PHD2 isn't able to keep the guide star in its starting position and doesn't correct until the starting point is reached again. FireCapture manages this without a complex multi-star method.

Does anyone have any idea what I need to change in PHD2 to prevent the drift? Or is there a problem with the EQ-8R that I've overlooked?

CS Henning

Are you dithering? Does Fire Cap keep the object centered when you turn its tracking tool off? Is your PA perfect? Are your scopes perfectly orthogonal to the mechanical axis of the mount? Are you sure you have zero deferential flexture?

I find that over time my framing will change slightly for a number of reasons. I use the “center after drift” command in NINA’s advanced sequencer to correct that. Its a part of the imaging loop and checks positioning with a plate solve after a number of frames that I specify. If the change in framing shifts beyond a given distance, it recenters.

when you guide with multi-star you are effectively following the frame and if you have field rotation (that is, if your PA isn’t perfect, literally) then you would experience both RA and Dec drift whose magnitude depends on both the PA error and your image scale (which I get is pretty high). The rate depends on the distance from the pole.

Thanks for your feedback, Tony!

I'm not consciously using dithering in PHD2. I hope it's not automatically or by default enabled somewhere in PHD2. But even if dithering were active, shouldn't the movement be circular?

In FireCapture, the object is usually kept centered for up to 30 seconds with tracking disabled.

The polar alignment is almost perfect; the object is nearly centered in the field of view after every GoTo movement. The telescopes should also be perpendicular to the mount's mechanical axis.

To what extent do you think lateral curvature could affect the guiding?

Is there a "Center After Drift" command in PHD2?

My problem is already the drift within a 5-minute exposure.

I would appreciate your feedback.

Clear skies, Henning

andrea tasselli · Apr 11, 2026, 01:55 PM
when you guide with multi-star you are effectively following the frame and if you have field rotation (that is, if your PA isn’t perfect, literally) then you would experience both RA and Dec drift whose magnitude depends on both the PA error and your image scale (which I get is pretty high). The rate depends on the distance from the pole.

Thank you for your feedback, Andrea!

Since the polar alignment is relatively precise and the exposure time relatively short, I don't believe field rotation is the cause here. For me, it only becomes clearly noticeable after several hours.

CS Henning

Whats your guiding setup? I guess you’re using a guidescope instead of an OAG because of the higher number of stars available in PHD2. If you’re using a guidescope it will be differential flexure most probably.

Thank you for your contribution!

You're right, I'm using a Primalucelab 60mm CompactGuide guide scope, which is perfectly aligned parallel to the C14, and an ASI 290mm guiding camera.

Therefore, this error can be ruled out.

Clear skies, Henning

Henning Schmidt · Apr 11, 2026, 02:01 PM
Thanks for your feedback, Tony!
I'm not consciously using dithering in PHD2. I hope it's not automatically or by default enabled somewhere in PHD2. But even if dithering were active, shouldn't the movement be circular?
In FireCapture, the object is usually kept centered for up to 30 seconds with tracking disabled.
The polar alignment is almost perfect; the object is nearly centered in the field of view after every GoTo movement. The telescopes should also be perpendicular to the mount's mechanical axis.
To what extent do you think lateral curvature could affect the guiding?
Is there a "Center After Drift" command in PHD2?
My problem is already the drift within a 5-minute exposure.
I would appreciate your feedback.
Clear skies, Henning

Dithering movement would be random, not circular.

Good GoTo’s arn’t necessarily an indication of a very good P.A. How are you doing your P.A. and what kind of numbers are you getting?

How do you know the OTA’s are perfectly parallel with the R.A. axis and perfectly orthagonal to the Dec. axis?

If you’re talking about field curvature in the guide scope, I don’t think that would be a factor.

Center after drift is a command in NINA, not PHD2. It requires the ability to plate solve to work.

I guess my point in all this that everything isn’t going to be perfect in any rig. PHD2 is doing what it’s designed to do. Keep the guide scope image from drifting. As you can see, there’s a lot that can happen outside of that constraint. I think switching to off-axis guiding would help but the biggest improvement would be to step up to a plate solving approach to mount control and pointing.

Let me briefly summarize the discussion at this point:

EQ8-R with C14 and RASA11 balanced in parallel operation (in equilibrium)
Primalucelab 60mm CompactGuide scope with ASI 290mm parallel to the C14 (no offset)
nearly perfect polar alignment (no alignment error)
exposure length only 5 minutes (no image rotation error)
C14 parallel to the mount's mechanical axis (no offset)
no differential deflection, as RA drift also occurs when operating with just the C14

From my perspective, it remains unclear whether

PHD2 automatically performs dithering and how to control it,
an RA backlash might be present,
it's possible to force PHD2 to continuously correct until the starting point is reached again (in FireCapture, a simple tool can keep a planet centered for hours)

CS Henning

PHD2 has to be commanded to dither. The reason is there are other things that need to happen. You can’t just dither randomly or while you’re taking a sub. The dither frequency also needs to be specified in number of frames. Once you reach that number the camera stops taking subs and the dither command is sent to PHD2. Once that’s complete, PHD2 goes back to guiding and the next sub is started. That’s why working with something like NINA is so important. It controls everything to make things like dithering happen properly.

On the rest, are you making assumptions about the mechanical perfection of your setup and alignments or have you measured it? For instance, the guide scope might be initially aligned with the imaging scope but how do you know that there absolutely isn’t any differential flexure as the system changes orientation. We are talking about deflections in the arc/sec range here. Have you measured it?

My point is that there are errors in every system. You can’t get everything down to zero error but the errors that remain can be compensated for.

Henning Schmidt · Apr 11, 2026, 03:07 PM
Thank you for your contribution!
You're right, I'm using a Primalucelab 60mm CompactGuide guide scope, which is perfectly aligned parallel to the C14, and an ASI 290mm guiding camera.
Therefore, this error can be ruled out.
Clear skies, Henning

Differential flexure isn't about the alignment from the guide scope camera to the main imaging sensor, it’s about mechanical imperfections in the rigidness of your setup which introduces wobble in the system. It does not have to be the guide scope per se, it can be rigid enough and something else can be wobbly.

Henning Schmidt · Apr 11, 2026, 03:23 PM
Let me briefly summarize the discussion at this point:
EQ8-R with C14 and RASA11 balanced in parallel operation (in equilibrium)
Primalucelab 60mm CompactGuide scope with ASI 290mm parallel to the C14 (no offset)
nearly perfect polar alignment (no alignment error)
exposure length only 5 minutes (no image rotation error)
C14 parallel to the mount's mechanical axis (no offset)
no differential deflection, as RA drift also occurs when operating with just the C14
From my perspective, it remains unclear whether
PHD2 automatically performs dithering and how to control it,
an RA backlash might be present,
it's possible to force PHD2 to continuously correct until the starting point is reached again (in FireCapture, a simple tool can keep a planet centered for hours)
CS Henning

Few points here:

Differential flexure needs to be ruled out by pointing at different objects at different altitude and azimuth and checking that the relative position of the image in the guidescope vs. the main imaging scope does not change (assuming no mirror flop/shift). Have you done this check?
One (or the main) scope is a SCT. Mirror flop WILL happen. That’s why most of the guys using such a scope are guiding with an OAG. That is especially true for a large mirror such as the one in a C14. To see that in action, see above test and perform a drift test around the meridian.
PHD2 does what it is told. Just check its settings or the settings for the session in NINA. By default it does NOT perform dithering. In NINA you can also chose how the dithering is performed, mimicking the operation in PHD2.
If you want a similar behavior as FireCapture do not use multi-stars tracking in PHD2.

Tony Gondola · Apr 11, 2026, 03:53 PM
My point is that there are errors in every system.

Thanks again for your suggestions, Tony!

My last post overlapped with yours.

After some shopping and a bike ride, I can now reply.

No system is perfectly aligned, that's true, but we have to be able to live and work with certain tolerances.

The dithering would indeed be random, but not 100% in only one direction. I haven't activated the dithering function.

I've been performing position correction for years using a 3-star alignment with the SynScan controller. The last values were Mel = 0.4' and Maz = 0.1', which should be perfectly fine.

The plate solving is certainly a good thing. But it should also be possible to track correctly with PHD2. It works with my other systems, and there were no problems with the MGEN either. And if even a simple guiding controller in FireCapture can keep a planet centered for hours, then the complex PHD2 should be able to do the same with a guide star.

I've also considered off-axis guiding, for example with the ASI 2600 mc Duo.

So far, I've only used the plate solving to precisely align the mount with the celestial north pole in its home position.

CS and many greetings

Henning

Tobiasz · Apr 11, 2026, 05:01 PM
Differential flexure isn't about the alignment from the guide scope camera to the main imaging sensor, it’s about mechanical imperfections in the rigidness of your setup which introduces wobble in the system. It does not have to be the guide scope per se, it can be rigid enough and something else can be wobbly.

Many thanks Tobiasz!

RA drift also occurred when I used the C14 alone. I have little concern about the rigidity of the system. I suspect a control error between the PHD2 and the mount.

CS Henning

andrea tasselli · Apr 11, 2026, 05:11 PM
If you want a similar behavior as FireCapture do not use multi-stars tracking in PHD2.

A good suggestion that I will gladly test.

Thank you so much!

Henning Schmidt · Apr 11, 2026, 06:09 PM
Tobiasz · Apr 11, 2026, 05:01 PM
Differential flexure isn't about the alignment from the guide scope camera to the main imaging sensor, it’s about mechanical imperfections in the rigidness of your setup which introduces wobble in the system. It does not have to be the guide scope per se, it can be rigid enough and something else can be wobbly.
Many thanks Tobiasz!
RA drift also occurred when I used the C14 alone. I have little concern about the rigidity of the system. I suspect a control error between the PHD2 and the mount.
CS Henning

Without guiding @3910mm focal length? Whats the periodic error of your mount specifically?

Tobiasz · Apr 11, 2026, 06:17 PM
Without guiding @3910mm focal length?

I did not write that I work without guiding at a focal length of 3910 mm.

What is the purpose of your question?

CS Henning

Henning Schmidt · Apr 11, 2026, 06:25 PM
Tobiasz · Apr 11, 2026, 06:17 PM
Without guiding @3910mm focal length?
I did not write that I work without guiding at a focal length of 3910 mm.
What is the purpose of your question?
CS Henning

Specifying the information you provided with additional questions.

Henning Schmidt · Apr 11, 2026, 06:09 PM
RA drift also occurred when I used the C14 alone.

What does alone mean? C14 without dual saddle on the EQ8 but with guiding or without? If you used the the C14 and a guidescope only that does not rule out the differential flexure topic.

If you used the C14 “barebone” without guiding on the EQ8, then the trailing is most probably caused by periodic error, which is normal at 4m focal length.

Tobiasz · Apr 11, 2026, 06:32 PM
If you used the the C14 and a guidescope only that does not rule out the differential flexure topic.

Okay, I understand. "On its own" means without a double saddle, but also with a guide scope and guiding.

Indeed, differential flexure could also be a possible cause, as M51 was high in the sky near the zenith. I found the following post on this topic:

https://www.cloudynights.com/forums/topic/929009-differential-flexure-what-else-can-cause-this/

This would allow for mirror movement of the main telescope due to vibration, for example, from the camera fan. However, this doesn't explain the continuous drift, which progresses a little further each time. It would have to remain in one position, and the guide scope wouldn't have a mirror that could move back and forth.

Furthermore, the mount is located in an observatory with a separate concrete base.

I perhaps got/get a similar issue, and I'm putting it down to my unreliable 3ppa

A dirty fix (band aid) if you're using Nina legacy sequencer like I do, is rather than have one sequence of say 120 × 300 sec subs, create 10 sequences of 12 × 300 sec subs with center target and start guiding options on.

It's not fun creating it, but you can save the sequence.

Henning Schmidt · Apr 11, 2026, 09:06 PM
Tobiasz · Apr 11, 2026, 06:32 PM
If you used the the C14 and a guidescope only that does not rule out the differential flexure topic.
Okay, I understand. "On its own" means without a double saddle, but also with a guide scope and guiding.
Indeed, differential flexure could also be a possible cause, as M51 was high in the sky near the zenith. I found the following post on this topic:
https://www.cloudynights.com/forums/topic/929009-differential-flexure-what-else-can-cause-this/
This would allow for mirror movement of the main telescope due to vibration, for example, from the camera fan. However, this doesn't explain the continuous drift, which progresses a little further each time. It would have to remain in one position, and the guide scope wouldn't have a mirror that could move back and forth.
Furthermore, the mount is located in an observatory with a separate concrete base.

Yup, fan vibration would have a constant effect on the images.

As far as I know SCTs may experience mirror flop depending on the sky location which can be also a possibility here. Are you focusing via the primary mirror or with a dedicated focuser while the mirrors are locked?

Another test could be imaging with an OAG. In the best case an OAG should resolve most of the issues.

TiffsAndAstro · Apr 11, 2026, 09:26 PM
A dirty fix (band aid) if you're using Nina legacy sequencer like I do, is rather than have one sequence of say 120 × 300 sec subs, create 10 sequences of 12 × 300 sec subs with center target and start guiding options on.

Good idea, but the RA drift is already clearly visible in a single image.

Many thanks and CS henning

Tobiasz · Apr 11, 2026, 09:53 PM
Another test could be imaging with an OAG. In the best case an OAG should resolve most of the issues.

I will test that and only work with the electronic focuser.

However, if I use the OAG guide, then mirror wobble directly affects the guiding and could even potentially increase the error or not?

CS Henning

Henning Schmidt · Apr 11, 2026, 10:07 PM
Tobiasz · Apr 11, 2026, 09:53 PM
Another test could be imaging with an OAG. In the best case an OAG should resolve most of the issues.
I will test that and only work with the electronic focuser.
However, if I use the OAG guide, then mirror wobble directly affects the guiding and could even potentially increase the error or not?
CS Henning

If you engage the mirror locks, the mirror flop shouldn't happen at all. I don't have a SCT, but I would assume the mirror flop will only happen once (at the meridian) and should be fine for the rest of the night. So you should only have 1-2 bad frames at the meridian and that's it.

If you have constant mirror wobble during the night, the problem is somewhere else.

I don’t know if this is still true but in my experience with SCTs, any movement of the focuser can cause the image to shift.

Related discussions