Persistent RA Oscillations on HEQ5 Pro (Belt Mod + New Bearings)

40 views
Skip to first unread message

NoahHD

unread,
Mar 20, 2026, 12:15:45 PM (yesterday) Mar 20
to Open PHD Guiding

Dear PHD2 Community,

I am reaching out to get some expert feedback on my guiding performance with an HEQ5 Pro mount. Despite several mechanical upgrades, I am still struggling with a mediocre total RMS and a specific periodic issue in the RA axis.

My Setup:

  • Mount: Skywatcher HEQ5 Pro (Rowan Belt Mod installed)

  • Telescope: Quattro 150p

  • Software: NINA, GSServer, PHD2 (Multi-Star Guiding enabled)

I recently replaced all factory bearings with high-quality SKF and FAG bearings. The Rowan Belt Kit is installed. I noticed a periodic error peak at approximately 122 seconds in my frequency analysis. Suspecting an eccentric motor pulley, I have recently attempted to re-center the motor pinion and ensure the grub screws are not tilting the pulley on the D-shaft. I have adjusted the worm gear to have a tiny amount of felt play to prevent stiction/binding.

My guiding logs show a persistent "sawtooth" pattern in the RA axis. While DEC is relatively quiet, RA seems to struggle with a rhythmic deviation that matches the motor's rotation period (~122s). This leads to a total RMS that is higher than expected for this level of tuning.

Due to bad weather, I haven't had the chance to test the latest pinion re-centering. I am planning to try the Predictive PEC (PPEC) algorithm next.

Looking at my setup, are there any other mechanical "usual suspects" I should check for this 122s oscillation?

Could a slightly over-tightened belt cause this kind of frequency peak, or is it strictly an eccentricity issue?

I have attached my latest Guide Log from a previous session (before the latest pinion adjustment) for reference. Any insights on how to push this mount into the "sub-0.7" RMS range" would be greatly appreciated!

Best regards

Daniel

PHD2_GuideLog_2026-03-17_165809.txt

Brian Valente

unread,
Mar 20, 2026, 12:27:32 PM (yesterday) Mar 20
to open-phd...@googlegroups.com
Hi Daniel

Here's what we can tell from your guidelogs (and this may be recap for you because it sounds like you've done some review of the logs):

you have three high PEs, presumably it's 1x, 2x, and 3x the primary. 1X is highest error and then is reduced with each harmonic, so that's pretty typical:
image.png

In the residual (uncorrected) error results, the larger period errors (1x, 2x) are pretty well handled by PHD, and again that's expected since longer periods are easier to guide out
the 122.7 second period is reduced from 1.5" to 0.35", but it is the highest remaining error so it's the constraint to your RA guiding:

image.png

As to the nature of this error, that answer is not something that comes from the guidelogs. Skywatcher or other SW users could comment on what is the source of that. You can see some unevenness in the RA calibration (the blue dots are not evenly spaced):
image.png


You can certainly use PPEC to at least test/address that error. you would set the period to 122.7 seconds, disable auto adjust, and I would start with predictive and reactive set to 70

You may also want to drop your exposure time to 1.5-2 seconds, since you are aiming for a higher frequency period. 

it takes a few cycles for the predictive part to kick in


Also in the future, please run a guiding assistant for a few worm cycles which produces unguided output and better shows the raw mount performance.

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/7050ad7b-a4f0-4931-acb2-b84740e84a88n%40googlegroups.com.


--
Brian 



Brian Valente

NoahHD

unread,
Mar 20, 2026, 12:57:44 PM (yesterday) Mar 20
to Open PHD Guiding

Hi Brian, thank you very much for this detailed analysis! 

I will run the Guiding Assistant for at least 3-4 full worm cycles (approx. 30-45 minutes) without guiding, as you requested. I will test switching to the PPEC algorithm for RA with the settings you suggested and will also reduce my exposure time to 1.5s - 2.0s.

Once the weather allows, i will post the new logs and the Guiding Assistant results. Thanks again for pointing out the unevenness in the calibration dots :)

Best regards,

Daniel

alex reinhardt

unread,
10:49 AM (7 hours ago) 10:49 AM
to Open PHD Guiding
Hi Daniel,

I have exactly the same setup as you do so I'm interested to see the comparison.  FYI, on the HEQ5 with the Rowan belt mod, the mechanical drive periods are:
1. Belt tooth engagement on main gear = 13.6 sec
2. Drive motor = 122 sec
3. Worm = 638 sec.  

Hmm, the 320ish sec period should not be there but just thinking, that works out to about half a revolution of the main 47 tooth gear.  Is it possible that gear is either off center/crocked or maybe the belt is running up and down on the gear and against the idler??  Or, another thought, it is possible it's rubbing the cover?  I noticed on mine today that the Ra belt has a wee bit of fraying happening on the edges after running for 6 months over the winter (certainly not full time...I wish we had no clouds!).  I did "loosen" up the idler a bit today allowing some up/down float to see if that helps.  Unfortunately, we are at the mercy of manufacturing tolerances on the motor to mounting plate fit so if the motor shaft is not perfectly parallel/aligned with the main gear axis, the belt will migrate up and down.   I personally struggle with the 13.6 sec period that PHD2 can't do anything about.  I'm playing around with the belt tension now that it's not -20C outside to see if that will helps, the chap from Rowan suggested that was the first place to start.  

In terms of guide settings, here's what I have settled on after much playing around, I never see the 122 sec period using the PPEC algorithm (after a few cycles till it gets it figured out as Brian said).  My exposure setting is normally 1.5 sec but I do sometimes go down to 1.0 or up to 2.0 (max) on crappy nights to see if either are better; sometimes 1.0 is but not very much. 1.5 is a good compromise.  I do run the guiding assistant on occasion and it suggests Ra Min move ranging from about .2 to .3 depending on the night.  I've never touched the Dec settings, they work fine.   On the GSS side, I have Ra guide rate at 80% and Dec rate at 70%. I saw somewhere on the net way back when that having a faster guide rate is better for a faster "response" mount, not sure whether that is valid or not but it seems to work OK.  Note you have to recalibrate if you change these numbers.  My rig is pier mounted and my polar alignment is better than 5' depending on what numbers you look at (PHD2 drift or NINA 3pt alignment).  

I typically see .6-.7" total guide error with Ra being bigger than Dec in my bortle 4-5 skies.  I saw .3" total for about 10 min the other night but then I also got 1.2" a while before that so it's really seeing dependent.   It's really hard to know whether a change in the algorithm values is responsible for a change or if it's seeing, at least once you are in the right ballpark and in the tweaking mode.  I've also attached a typical freq analysis for info, here for the full 1hr 3min segment.

Brain/Bruce, just a question re the period length it determines as I've got auto adjust on, here 962 sec.  I've looked at my other guide logs, or even on the same night at different segments, and this number varies from about 650 to 1350 sec.  Just curious on why that would be the case, I mean it's guiding fine with no issues so...?  

Hope this helps!

Alex

Screenshot 2026-03-20 172248.jpg

Screenshot 2026-03-20 180523.jpg

Brian Valente

unread,
11:06 AM (7 hours ago) 11:06 AM
to open-phd...@googlegroups.com
Alex
thanks for providing some feedback

regarding your question:

>>>Brain/Bruce, just a question re the period length it determines as I've got auto adjust on, here 962 sec.  I've looked at my other guide logs, or even on the same night at different segments, and this number varies from about 650 to 1350 sec.  Just curious on why that would be the case,

imo Auto adjust is primarily for when you don't know your mount's PE or don't have a specific targeted period. It will try to find a a periodic error but It can guess wrong. Anything above your primary worm period is probably not doing much but also not doing any harm. I suggest you review your logs, determine what is your highest error period, and use that as your setting with auto adjust disabled. 

areinhardt784

unread,
2:03 PM (4 hours ago) 2:03 PM
to open-phd...@googlegroups.com
Thanks Brian, I'll try that.  



Sent from my Galaxy
Reply all
Reply to author
Forward
0 new messages