New to PHD2 - Celestron AVX user

378 views
Skip to first unread message

Vitaly

unread,
Jul 7, 2025, 1:14:17 PM7/7/25
to Open PHD Guiding

Hi everyone,

Recently, I decided to stop using the ASIAIR and switch to a mini PC with NINA.

As a result, I had to spend some time getting to know the PHD2 guiding.

I’d like to ask the knowledgeable people in this group to take a look at my logs:

https://openphdguiding.org/logs/dl/PHD2_logs_A4Sv.zip

My calibrations were either good or acceptable, so I assume that part is fine.

What suggestions do you have for improving guiding on my mount?

My setup is perfectly balanced on both axes. Should I make it slightly east-heavy and camera-heavy instead? I’m not sure.

When I ran the Guiding Assistant, it didn’t give me a backlash compensation value. Instead, I consistently got the message:
"Mount never established consistent south moves – test failed."

Could the guiding speed for the DEC axis (0.5x sidereal) be too low for the AVX? I’m not sure.

I am not sure which algorithm should I use for RA: Hysteresis or Predictive PEC.
When the latter, should I then also PEC train my mount using Celestron's PECTool? I'm not sure.

Brian Valente

unread,
Jul 8, 2025, 10:47:02 AM7/8/25
to open-phd...@googlegroups.com
Hi Vitaly

Let's go one topic at a time because you have a lot going on here that needs some attention

First, you should upgrade to the latest dev release and use the calibration assistant to calibrate your mount. 

When you do guiding runs, it really helps to do them for longer. A handful of a few minutes doesn't really tell us the full picture. Doing runs that are minimum 10-15 minutes per run helps


Second, your pixel scale is quite coarse, 6.45"/pixel. Your guiding is hitting the limits of what's possible with this kind of setup, so you may consider a longer focal length guidescope or an oag. 

What is the physical connection of your guidescope? The typical rings/thumbscrews are a source of many problems for guidescope movement. 

A few pictures of your setup from different angles can often help to pinpoint issues with the setup

Regarding your calibration, the first one was good. it had a slight but acceptable orthogonality error. you can see the red (dec) and blue (ra) lines are fairly straight, the points are evenly spaced, and roughly at 90° from each other
image.png
Subsequent calibrations all show the dec is bunching up, which usually indicates some sort of stiction where your mount can't move the dec axis cleanly between the steps. All your subsequent calibrations show this problem, so the first good calibration seems to be outlier. This could also be poor Dec balance which you should confirm. I would test this by trying opposite side of the sky and doing a calibration, if it shows an improvement then you can re-inspect your dec balance.

image.png

The backlash test shows the dec axis taking quite a long time to reverse direction, and even then the reversal is barely moving. It took thirteen steps at 1800ms for the dec axis to even start the reversal (between the two green arrows,  that's 23 seconds), so the dec axis is basically unresponsive when it comes to reversal. This is going to need mechanical adjustment to improve this. 

image.png

The actual guiding runs, RA looks pretty rough. you have a lot of high frequency errors and the axis isn't very responsive. See below (blue is RA) the result is you have a lot of high frequency large excursions (line indicated by the red arrow) that are not caused by guiding, and when guiding is applied (the bars indicated by the green arrow)  it is largely unresponsive. It doesn't help that you experimented by dialing down the RA aggression. THis mount is going to need some really aggressive guiding and short exposure times (think 1 second) to even begin to address this
image.png
looking at the residual (uncorrectable) frequencies you have large repeating errors at 11,18,29, 99 (red arrows) and teh green arrow is probably your primary period, which again is quite high for what's uncorrectable. Unfortunately this indicates a really poorly performing RA axis that is likely suffering from one or more mechanical issues.

image.png
In Dec I can see you tried different approaches to Dec such as guiding only in one direction (not a bad idea), but unfortunately Dec is so unresponsive it doesn't really make a difference. In this 12 min run which Dec never really made it back to baseline despite numerous guide pulses. 
image.png
It's so much that even enabling dec backlash compensation (which at some point is probably a wise idea) won't help


Fiddling with the settings (aggression, exposure time, etc.) isn't getting you anywhere, this mount really needs some mechanical TLC imo. 

This host of issues isn't really uncommon for this type of mount. 

The silver lining is that your approach to using PHD is a good start. Get the software updated, use the calibration and guiding assistants to guide you, and produce longer runs that will help clarify.

Brian



--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/open-phd-guiding/19165071-6753-4a9e-8ab6-0d538f000618n%40googlegroups.com.


--
Brian 



Brian Valente
Message has been deleted

Vitaly

unread,
Jul 10, 2025, 6:14:23 AM7/10/25
to Open PHD Guiding
Hi Brian.

Thank you for your comments, I really appreciate it!
Currently, my guider is mounted on the OTA (piggy-backed), and is rigidly fixed, IMO.
Unfortunately, using an OAG is not yet possible for me at the moment, so I’ll have to make do with what I have.

I decided to check the gears of the RA and Dec axes, and it looked like they needed to be brought closer together — which I did.
Mayby I overdid a bit and now I have issues from the binding instead...idk

Here are the logs:

Calibration now shows more issues as I understand from it's graph, there are some weird movements in both axes.
I have also increased the guiding speeds to 0.99 for RA and DEC and it seemed, probably, helped me to identify succesfully the Dec backlash compensation, which was, yes, very huge, to almost 8 seconds.
It didn't feel right to run guiding in this situation for a longer time. But if needed I could try doing that today when clear skies happen.
Switching to uni-directional this time seemed to help with Dec guiding, lowering the aggression to 50%.

вторник, 8 июля 2025 г. в 16:47:02 UTC+2, bval...@gmail.com:
Message has been deleted
Message has been deleted
Message has been deleted

Vitaly

unread,
Jul 13, 2025, 7:27:51 PM7/13/25
to open-phd...@googlegroups.com
Hi Brian,

Did you have a chance to check my second set of logs. Did I make the system worse?

Vitaly

On Fri, Jul 11, 2025, 05:29 Vitaly <vpust...@gmail.com> wrote:
Hi Brian,

My guide scope is piggy-backed on my OTA, I can't do OAG at the moment.
I have decided to have a look onto the gears for both axes. 
It seemed to me that the distance between the gears for the DEC could be reduced a bit, which I did. Did for RA as well, just on case.
Additionally I have increased the guiding speed for both axes to 0.99 of the siferal.
So those are the results after that manual manipulation: https://openphdguiding.org/logs/dl/PHD2_logs_JjZj.zip

Did I introduce binding now?

Calibration looks worse and the line for the backlash test is bumpy.

вторник, 8 июля 2025 г. в 16:47:02 UTC+2, bval...@gmail.com:
Hi Vitaly

Brian Valente

unread,
Jul 14, 2025, 5:30:55 PM7/14/25
to open-phd...@googlegroups.com
Hi Vitaly

From your new logs, it doesn't appear much has changed. You did a 15 minute unguided result which is great, but the guided results are still too short. 

your backlash looks about the same - far too much for good results (around 7 seconds backlash? in green below) the backlash chart looks about the same. 

image.png

I don't see much different in terms of RA. There may be a bit of stiction, but it's not a consistent problem. 

Regarding your comments about adjusting the gear mesh, I suggest you contact Celestron and ask for specific instructions. Just opening up the mount and looking by eye on how to adjust is not the best approach.

Also i see that you were adjusted the dec backlash aggression - do not do this: it did not help anything and instead makes your reversals sluggish. Leave it at 100%

Brian


On Thu, Jul 10, 2025 at 8:29 PM Vitaly <vpust...@gmail.com> wrote:
Hi Brian,

My guide scope is piggy-backed on my OTA, I can't do OAG at the moment.
I have decided to have a look onto the gears for both axes. 
It seemed to me that the distance between the gears for the DEC could be reduced a bit, which I did. Did for RA as well, just on case.
Additionally I have increased the guiding speed for both axes to 0.99 of the siferal.
So those are the results after that manual manipulation: https://openphdguiding.org/logs/dl/PHD2_logs_JjZj.zip

Did I introduce binding now?

Calibration looks worse and the line for the backlash test is bumpy.

вторник, 8 июля 2025 г. в 16:47:02 UTC+2, bval...@gmail.com:
Hi Vitaly

Vitaly

unread,
Jul 16, 2025, 4:21:02 PM7/16/25
to Open PHD Guiding
Hi Brian,

Thank you for your message.
I have sent a support email to Celestron.
While waiting for their response I found this article https://www.cloudynights.com/topic/701940-avx-backlash-wormgear-adjustment-20/
and attended doing something similar.

First of all I have discovered smth I haven't paid attention for before - the RA housing upper case was actually rubbing against the RA axis in one specific place. 
That might have affected the RA guiding.

photo_2025-07-16_20-47-41.jpg
I fixed it by unscrewing a bit the interfaces shaft (Motor Board cover) away. Temporarily put a tape to try hold it away from the axis body, I might need to open the housing again.

As for the DEC, I might have fixed a huge stiffness it had this whole time...3+ years
Moving the axis by hand in this default assembly I could feel there is a lot of resistance in the movement.

I have dismantled the DEC motor and took out the worm gear which moves the axis.
But to do that I had to use my drill-screwdriver otherwise I could not unscrew it myself, all of the bolts were crazy tightly screwed on.

After readjusting the tension I believe the DEC axis behaves smoother.

Now I am waiting for clear skies in order to test everything again.

понедельник, 14 июля 2025 г. в 23:30:55 UTC+2, bval...@gmail.com:

Vitaly

unread,
Aug 15, 2025, 7:15:02 PM8/15/25
to Open PHD Guiding
Hi,

So..I am back....I did manage to srew up my DEC axis and had to order a replacement.
Unfortunatelly Celestron Support till today has not provided me with the definitive guidance on the backlash tweaking, I asked them a month ago.

On the positive side - my new DEC axis behaves a bit better.
Could you check the uploaded logs https://openphdguiding.org/logs/dl/PHD2_logs_T7Ht.zip

There is some improvement in DEC backlash compensation, though still not enough.
As I am not going to touch the axis again I suppose the only way to deal with the backlash is to make correction only in one direction, right?

Kind regards,
Vitaly



среда, 16 июля 2025 г. в 22:21:02 UTC+2, Vitaly:

Vitaly

unread,
Aug 20, 2025, 4:34:30 PM8/20/25
to Open PHD Guiding
Could you check on the logs for me?

Kind regards,
Vitaly

суббота, 16 августа 2025 г. в 01:15:02 UTC+2, Vitaly:

Brian Valente

unread,
Aug 21, 2025, 12:32:14 AM8/21/25
to open-phd...@googlegroups.com
Hi Vitaly

Thanks for checking back in, sometimes we miss followup emails

Not much has changed since your last session. You still have high Dec backlash and several periodic errors in RA.


your calibration looks good

The backlash was still mostly the same as before:
image.png
This is confirmed by your 90 second run where Dec needed to reverse, and it took quite a long time to do this (the largest guidepulse at the top is 682ms, you have 27 guide pulses, so very roughly i'd estimate the backlash at 10 seconds?)
image.png

In the guided run, because dec did not have to reverse, it was relatively well behaved at a total 0.41" rms, which is pretty good. RA still showed the same behavior: A few high frequency spikes particularly around ~19 seconds, and the primary PE still being very high to the point where a mount-based PE should be applied (or a mechanical fix)

image.png


At this point, sending more guidelogs after doing minor adjustments in hopes of improving things is not going to be productive. I really think you need to resolve the mechanical aspects of your mounts' performance before you are going to see any substantial improvement. 

Brian

Vitaly

unread,
Aug 21, 2025, 5:30:46 AM8/21/25
to Open PHD Guiding
Hi Brian,

thank you for checking on once again.
As I am not sure I want to tweak anything else mechanically, in order not to break the axis again, I see only the following solutions for possible left improvements:

1. Train PEC and activate it in the mount. Use Hysterisis algorithm in PHD2.
2. Set the polar alignment error to around 5-10 arcminutes and dec guide only in one direction.

No other options, I suppose?

Vitaly


четверг, 21 августа 2025 г. в 06:32:14 UTC+2, bval...@gmail.com:

Brian Valente

unread,
Aug 21, 2025, 9:23:22 AM8/21/25
to open-phd...@googlegroups.com
Hi Vitaly

Your problems are mechanical in nature, so it seems to me you would want to look at mechanical solutions. If your axis is broken and you are unable to repair or replace it, the two options you mentioned are probably your best alternatives, but there's only so much software can do with a damaged mount

Best of luck to you

Brian

Vitaly

unread,
Aug 21, 2025, 9:30:27 AM8/21/25
to Open PHD Guiding
Hi,

ahm, at the moment my mount, the one with replaced DEC axis, is not broken, and I do hope it will not. I was referring to my previous message on 6th of August.

I don't have instructions on how to mechanically reduce the DEC backlash, and Celestron Support Team seems to ignoring my request from July.
So for now I will stick with those 2 options. Thank you.

четверг, 21 августа 2025 г. в 15:23:22 UTC+2, bval...@gmail.com:

Brian Valente

unread,
Aug 21, 2025, 9:38:38 AM8/21/25
to open-phd...@googlegroups.com
You may want to find a user group specific to your mount and ask there. CloudyNights etc. 



Reply all
Reply to author
Forward
0 new messages