EQ6 DEC Tuning Advice

[Atmos]

As there are a good number of people on here that have had a bit of experience in stripping and redoing EQ6's, I figured this would be the best place to come for a little bit of advice.

Over the new year I decided that I should strip my OLD EQ6 down and give it a regrease, it hadn't been done in a few years and I was beginning to get a lot of large spikes in RA. So I have since got RA and DEC running down as low as 0.4" RMS, sometimes managing to go nearly a minute without having to make a guiding hit, especially in DEC which sometimes gets as low as 0.25" RMS. The problem that I am having is with what I can only imagine is DEC backlash.

Over time pulses do need to be sent to DEC, if they had previously been going in the other direction it can take ~50 pulses to recover before it'll make a DEC movement. After this it may not need another serious move for a few minutes but it'll usually end up being in the opposite direction where it'll again have to clear the backlash before making an actual move.

I have got both axis so there is no wobble at all and there is no binding. I spent a bit of time playing with the DEC this yesterday afternoon, as when I last had it out a few nights ago PHD had to do something like 100 backlash compensation moves. Tonight it was down to 50 but buggered if I know how to remove them :/

Could it be that the belt isn't quite as taught as it needs to be?
Do I need to work on the worm gears (the two hex bolts)?
Maybe the washer thing that I need the needle nose for? (struggling to remember what that washer thing is called, under the black cap)

Before I stripped down the mount I would usually have about 2-4 backlash compensation moves in both axis so I know it is capable of doing much better than what it currently is.

Thanks for any help and advice :)

[glend]

Colin I have exactly the same issue with my NEQ6 Pro. It has improved after the Rowan Belt Mod but DEC drives me crazy. No apparent slop or binding and yet the oscillations in DEC continue. I think my belt is tight enough, and the worm, end cap bearing, and drive wheel are all aligned correctly. I will be following your thread with interest.:thanx:

[Atmos]

I know it only started doing it after I cleaned its innards so it is definitely user induced :P

[peter_4059]

Colin,

I believe these are the potential sources of DEC backlash:
1. Worm to ring gear interface
2. Worm longitudinal play
3. Belt gear to worm interface
4. Belt loose
5. Belt gear to stepper motor shaft interface

It sounds like you have checked 1 and 4. 2 is easy to check if you remove the cap that conceals the worm bearing and adjusting ring (you need needle nose pliers for this). If there is longitudinal movement of the worm between the worm bearings you can feel the worm move with your finger when you change DEC direction. If this is the case you need to tighten the ring that holds the worm in place. Have a look at "Stage 2 - Stripping the worm carrier" on Astrobaby's website: http://www.astro-baby.com/EQ6%20rebuild%20guide/EQ6-declination%20strip%20down.htm The third image shows the ring that adjusts tension. Don't over tighten this as the worm will not move freely.

3 and 5 are more work. Did you ensure you engaged the set screws for the belt gears on the flat section of the worm and motor shaft when you installed these and that the set screws were locked down tight. I've seen cases where the belt gear rotates slightly on the shaft because the set screw was not aligned with the flat on the shaft.

If all of these are ok then it is possible you always had this backlash but you were operating on a different section of the DEC ring gear. EQ6 ring gears often have a tight spot and a loose spot however since the DEC axis never completely rotates it is possible to adjust the mesh and the position of the ring gear so you normally operate in either the tight part or the loose part. You can change this by releasing the DEC clutch and rotate the DEC axis with the hand controller.

[glend]

I seem to recall that if your mount is well aligned you don't need DEC guiding running - just turn it off and guide on RA only. I have not tried it but since mine is on a pier I might give it a go.

[peter_4059]

Or intentionally misalign it slightly so it is always correcting dec in one direction.

[Tony_]

Hello Colin,

I have a CGEM mount with a similar problem (although it doesn't have belts). In my case the backlash is in the motor itself. My autoguider can switch off dec control in either direction. I always switch off one direction so that it only corrects for drift - this solved the backlash problem.

Regards,
Tony.

[Atmos]

I have been watching that and it appears to be okay, it is tight but not super tight. I both tried it tighter and looser, neither seemed to make any difference. On the plus side I have noted that I have physically see it happening when watching the worm shaft. When using the hand controller at a rate of 2, I can see the non-movement in the worm shaft for a second or two when switching directions. At the very least it means it is testable without having to keep toying outside.



It sounds like it may be this that is the problem. I didn't do the belt mod, it was done by the previous owner and I hadn't had any issues for the last 8 months. I could always give it a double check to make sure that they didn't somehow become lose in the strip down. I cannot imagine that they would have become lose but it seems like a logical explanation.



Pretty sure this isn't the case. As I haven't done any PEC training I have in the past just found it easier to sometimes point the telescope at a star and then tell TheSkyX "You're here" so the worm cycle should have had a chance to move around every now and then. Sometimes don't bother Parking either :P

[Atmos]

Would that be an Altitude or Azimuth misalignment?



I should probably double check that it isn't the motor causing the issue. It is an issue that only started AFTER I decided to strip the mount down and clean it so I know it is something that I have caused, now I want to make it go away :P

[peter_4059]

I think you would probably need to do both because the DEC drift due to altitude is more pronounced when you are pointing near the horizon and due to azimuth when you are near the meridian - hence the way drift aligning works.

You would need to ensure you make the adjustment to each axis consistently so it doesn't swap from a north drift to a south drift as you track from horizon to meridian as this would cause the guider to change the direction of the DEC corrections.

[codemonkey]

I had the exact same issue with my EQ6. It ended up being a combination of the worm end float and the worm meshing (i.e. 1 & 2 in Peter's post).

Hours of tweaking these finally got it to the point where it takes <= 5 pulses in DEC to clear the backlash. Also keep in mind that having the float too tight can appear as backlash as well, as was the case in mine.

I've also had the set screw come loose on the RA axis after my belt mod. I was doing PE recording at the time and things didn't seem right; when I reviewed the PE curve it was really smooth, so that's one way to identify that. In my experience (sample size of 1 here) the amplitude of the PE will be about the same as it otherwise would, but it will appear much less noisy than usual.

[codemonkey]

And that's why I hated using the "guide in one direction" advice. I'd go back up to the scope and find that it had started drifting in the other direction and I'd lost a bunch of subs.

Best advice is, where possible, fix the mechanical issue so that you can guide in DEC.

[peter_4059]

I agree best to make the mount as good as you can but failing that guiding in one direction is better than continually fighting backlash. The misalignment is not that hard to do:

Scope pointing to east horizon, star drifting north => raise altitude
Scope pointing to meridian (CW in east), star drifting north => rotate AZ clockwise


Easiest way is to use PHD2 drift align tool and ensure the trend line is going the same way (either up or down) for both the AZ and the ALT adjustment steps. Also if the slope of the line is the same in both cases the the drift rate should be similar.

[RobF]

Great thread and summary guys. Will have to run through your checklist next time my EQ6 gets opened up Peter.

[Atmos]

Okay I THINK I may have fixed the problem. I had the good intentions of stripping the mount down again to open her up and the thought occurred to me... If it was an issue within the worm housing (any of the parts in there), I would be able to feel the backlash by manually moving the worm drive with a pair of pliers. So I put tension on the belt with my finger and made small and careful movements in both directions... Didn't feel any backlash.

Connected the motor on its own and run it back and forth to see whether there was any backlash in the motor (maybe I needed to tighten one of the hex bolts that holds the gear), didn't feel anything.

Finally I decided to go all anal (I apologise for the potentially terrible image there) about getting a very good tension and mesh with the motor and belt. This I feel is best done with a second pair of small hands, my 26 year old hands don't fit in the EQ6 housing very well! This is one of those moments where I wish I had a 10 year old daughter :P
Anyway, between using a large screw driver as a leaver to ensure good tension on the belt, I manually moved the worm drive, and therefore the belt, so as to make sure that everything was in place. Then I tightened the three hex bolts holding the hear in place and the backlash seems to have gone now.

Although the RA seemed fine I figured I'd do the same procedure as well just to make sure that it can work at its best. All in all it probably took a good 20-30 minutes to do the testing and be thorough (sounds better than anal) about getting good meshing between the belt and motor gearing.

I guess this is the big benefit of belts over gears, I can now better understand why the belt helps remove backlash :)

[codemonkey]

How're you measuring the backlash, Colin? That was the difficult part for me. Easy to make adjustments, hard to know what impact they had until you were under the stars.

Sticking an eyepiece in the scope and moving it back and forth and watching for delays was useful when the backlash was really bad, but after getting it down a bit I found it difficult to detect.

[codemonkey]

Aha, thanks for that Peter. I always have difficulty visualising things like this.

[Atmos]

Totally true! I am hoping that tonight is at least a tiny bit clear. I don't think it'll be acceptable for any kind of imaging but hopefully I'll at least be able to test the backlash.

[codemonkey]

Yeah, that's the frustrating part. Hmm. If you had a fake star that you could bring to focus, then fired up your mount and PHD with tracking disabled on the mount, you should be able to fairly easily quantify the backlash.

[ZeroID]

Following this with interest also. My EQ6 (no belts) is thrashing around in DEC and RA, seems to be continually correcting and over correcting. I reduced the aggressiveness in both axis which seems to have helped a wee bit but I think it needs to come apart for a clean, regrease and retune.
Probably take the opportunity to rebuild my pier head as well, it's a bit mickey mouse at present.
It's the loss of imaging opportunities that makes me hesitate, at F7 the software can fix the issue mostly but at F14 it becomes a real problem.

[Atmos]

Skies tonight weren't even worth setting up for imaging but I did test out the mount, down to 5 backlash steps so I am pretty happy with that :)

[codemonkey]

Gotta be happy with that! Congrats.

[RobF]

Reading some of the great tips here motivated me to operate on my (previously belt modded) NEQ6 last weekend. Tackled a few things that had been bugging me:

- large PE (probably average, but around +/- 15 arcsecs)
- RA and DEC backlash (quite a bit of give in both if moved by hand)

Very happy with the outcome. In no particularly order, things changed:

- ABEC 11 bearings in place of ceramic bearings previously used during belt mod (Tandum suggestion as per http://www.iceinspace.com.au/forum/showthread.php?t=88499&highlight=abec
- Replaced a few key 6008 bearings that didn't feel super smooth
- Removed dust caps off all 6008 and worm bearings (another Tandum suggestion - apparently Tak does that in their mounts)
- Swapped RA and DEC worm and gear
- Put more tension on belts when re-assembled than previously
- Spent a lot more time working on mesh and tuning of each axis. The order suggested by "Astrobaby" seemed to work really well. (http://www.astro-baby.com/EQ6%20rebuild%20guide/EQ6%20worm%20alignment.htm)

Very little higher frequency noise on PEC plot and PE on the RA worm is now about +/- 7 arcsec, albeit with a blip each worm cycle or a few arcsecs that doesn't matter too much at my focal lengths (2.5 arcssec/pixel typically). No discernible looseness in DEC or RA by hand - feels much tighter with no binding. Backlash seems better, but no hard data to prove - certainly less overcorrection in Dec now.


In hindsight, things that probably helped most:

- 2nd time ever pulling it apart - better understanding this time (first time was just hoping it would slew when re-assembled! :) ). This time took the time to properly tune meshing.
- Lucky improvement swapping gear/worms
- getting rid of the ceramic bearings - these appeared and felt to have more "give" than traditional bearings - may not have been a good idea in the first place
- Better mesh (as above) - really thinking about the issues in post#4 from Peter as each variable was checked and tweaked.

Satisfied the NEQ6 is now performing at least as good as the HEQ5Pro has always managed, probably significantly better after a few backyard sessions this week.

Would be interested if anyone else has tried Tandum's suggestion re taking out bearing dust caps. Presumably this greatly increases the chances of future bearing contamination, but I'd be prepared to have to service more often for less vibration during tracking.

[RobF]

Also noted there is replacement kit for EQ6s now that looks great. Screw in counterweight arm with wider bore. Peter_4059 did something similar himself and seemed very happy - almost no slop.

http://www.telescope.com/Mounts-Tripods/Mount-Tripod-Accessories/Orion-Counterweight-Shaft-Upgrade-for-Atlas-EQ-G-Mount/c/2/sc/38/p/102787.uts

Saving my dollars for now however. Price seems reasonable but counterweights would need to be bored out. Bet that chrome would look snazzy though ;)

[codemonkey]

Good work Rob! I'll be stripping mine back down again soon to try and resolve the issue I was having in RA. Betting money the pulley has come loose again. This time I'll sort it out once and for all with some thread locker, as suggested Dave from Rowan Engineering.

I've got that aftermarket counterweight shaft, but I'm not sure how much impact it had since I think I replaced it at the same time I did the belt mod.

The counterweights from the AZ-EQ6 fit that Orion one perfectly if you happen to have any of those laying around. I bought some new ones rather than boring out existing ones.

[RobF]

I think you end up with a better idea what guiding settings to tweak too after stripping your mount down a couple of times and looking at a few cycles of PEC. Gives you a realistic expectation of backlash, PE drift, minimum move settings etc.

Getting decent alignment and moving up min move parameters up around the RMS values seems to help too. At least until I win the lotto and by my Astrophysics dream mount :)