Possible asymmetry in NEQ6 axes

[codemonkey (Lee)]

Does anyone have any suggestions on what might cause asymmetry in the axes of an EQ6 mount?

I had problems with my DEC axis prior to doing a belt mod, which are persisting after doing it. I've now also noticed this in the RA axis, but I'm not sure if it was always like that and I've just never noticed. The RA isn't really a problem actually, it's just the DEC giving me grief.

So what I mean by asymmetry is that axis will move more freely at one point of rotation than at another. This makes gear meshing more difficult because to get it loose enough to rotate a full 360 degrees without binding, I think it makes it too loose in other parts.

I've got what is appearing to be very bad DEC backlash (takes PHD2 ~25 steps to clear when calibrating), but I've spent hours trying to get the gear meshing as close as I can without binding and I don't think I'm that far off, so my only thought at this point is that it's asymmetrical.

Everything is well-greased and all the bearings seem to be in good condition and move smoothly by hand.

Any thoughts?

[Shiraz (Ray)]

Hi Lee. I take it that you are adjusting the gear mesh as tight as possible without binding and are finding that you get different mesh backlash at different DEC angles? I think that all mounts have this issue to a greater or lesser extent - unless the gears are spring loaded or unless you happen to have fluked a combination of bearings, shafts and gears where the various unavoidable run-outs cancel out.

Fix it by slightly un-balancing the DEC axis so that there is always some load to keep the gear faces in contact on only one side when in mesh (I use a small lead weight that I clip onto an appropriate position on the OTA). Probably also a good idea to make sure that the bearings are turning freely at all DEC angles as well - stiction can reduce the effectiveness of this method. You would probably have to unbalance any mount without spring loaded worms, so it doesn't really make any difference to the outcome if the backlash varies a little or a lot in your particular mount. There is no real point in trying to fix it unless you are concerned that your mount must be perfect, even if it won't make any difference to your imaging - then you could try some high ABEC main bearings as a first step.

[peter_4059 (Peter)]

Lee,

I think there are a number of potential sources of backlash:
1. Mesh between worm and ring gear
2. End play in the worm (ie longitudinal movement of the worm between the worm bearings).
3. Belt too loose
4. Belt gears not locked on worm/motor shaft

If you've eliminated 2,3 and 4 and are having problems getting the mesh tight enough without the axis binding, I think you are experiencing run-out in the ring gear. This is common on these mounts and is a manufacturing tolerance issue. I think the remedy is hobbing the ring gear however I believe this is not a simple exercise.

The dec axis does not make a full rotation in normal operation and if the polar alignment is good it barely moves at all. On mine, I set the worm mesh at the tightest point so it never binds. There are dec backlash settings in EQMOD and I think newer versions of PHD2 are also trying to measure and compensate for this.

Peter

[LewisM]

Unfortunately, of the 3 NEQ6's I owned, I found this asymmetry in 2 of the 3. the third one had been "hypertuned" and had relieved this discrepancy (and it did track and guide flawlessly - thanks Grady! )

Of the 3 HEQ5's I have owned, ALL did it.

My advice is strip them down, rebuild them, using good bearings and delrin washers. They NEED it.

[codemonkey (Lee)]

Originally Posted by Shiraz

Hi Lee. I take it that you are adjusting the gear mesh as tight as possible without binding and are finding that you get different mesh backlash at different DEC angles? I think that all mounts have this issue to a greater or lesser extent - unless the gears are spring loaded or unless you happen to have fluked a combination of bearings, shafts and gears where the various unavoidable run-outs cancel out.

Fix it by slightly un-balancing the DEC axis so that there is always some load to keep the gear faces in contact on only one side when in mesh (I use a small lead weight that I clip onto an appropriate position on the OTA). Probably also a good idea to make sure that the bearings are turning freely at all DEC angles as well - stiction can reduce the effectiveness of this method. You would probably have to unbalance any mount without spring loaded worms, so it doesn't really make any difference to the outcome if the backlash varies a little or a lot in your particular mount. There is no real point in trying to fix it unless you are concerned that your mount must be perfect, even if it won't make any difference to your imaging - then you could try some high ABEC main bearings as a first step.

Thanks Ray :-)

I'm theorising about the problem really, I haven't confirmed different backlash at different orientations... it's possible I'm wrong, but I really don't think I'm going to be that far out, not far enough to cause PHD to take ~25 steps to clear it. I've spent a fair amount of time on it, making tiny adjustments to back it off (< 1/8th of a turn) after first getting it to the point of binding.

I can understand how creating an imbalance on the RA axis can work, because it's always moving in the same direction and to guide in the opposite direction it just slows down, so as long as there's an imbalance, it will solve any guiding issues there.

I don't understand how an imbalance in DEC will have the same effect, unless you only ever guide in one direction, since DEC isn't moving, so to guide in the opposite direction it has to physically move it, thus losing the slack, if that makes sense. Is there something here that I'm not getting?

Originally Posted by peter_4059

Lee,

I think there are a number of potential sources of backlash:
1. Mesh between worm and ring gear
2. End play in the worm (ie longitudinal movement of the worm between the worm bearings).
3. Belt too loose
4. Belt gears not locked on worm/motor shaft

If you've eliminated 2,3 and 4 and are having problems getting the mesh tight enough without the axis binding, I think you are experiencing run-out in the ring gear. This is common on these mounts and is a manufacturing tolerance issue. I think the remedy is hobbing the ring gear however I believe this is not a simple exercise.

The dec axis does not make a full rotation in normal operation and if the polar alignment is good it barely moves at all. On mine, I set the worm mesh at the tightest point so it never binds. There are dec backlash settings in EQMOD and I think newer versions of PHD2 are also trying to measure and compensate for this.

Peter

Thanks Peter :-)

I'm reasonably confident that I've gotten #1 as good as I can get it. I've spent a long time on it so unless new information comes to hand, or I can get something better by pure luck, I don't think there's much more I can do there.

#2 is a definite possibility, I'll play around with that. When I had it all apart I played with that and noticed that it seemed as soon as I did the worm float up beyond it "not being loose" the worm moved poorly, so I figured I'd just keep that "not loose".

I don't think #3/#4 are likely, I've just taken the mount apart again on Fri/Sat and made sure everything was tight there. Possible I guess, but less likely.

I've seen some backlash before, usually in the area of 3-4 steps, but ~25 is nuts. There's no way I can guide in DEC with that. My only option is to either:

#1 Get polar alignment as close to perfect as possible and never guide in DEC
#2 Intentionally misalign the mount and only guide in one direction - sometimes this over-corrects
#3 Imbalance DEC?

Hobbing the ring gear sounds slightly outside of the realms of the possible for me. Buying a new one might be possible, if I knew it to be an issue.

Interestingly enough I tried messing with the EQMOD backlash setting last night and it seemed to make no difference between 0 and 2000msec.

Originally Posted by LewisM

Unfortunately, of the 3 NEQ6's I owned, I found this asymmetry in 2 of the 3. the third one had been "hypertuned" and had relieved this discrepancy (and it did track and guide flawlessly - thanks Grady! )

Of the 3 HEQ5's I have owned, ALL did it.

My advice is strip them down, rebuild them, using good bearings and delrin washers. They NEED it.

Thanks Lewis :-)

I've stripped it all down a couple of times now and checked the grease and the bearings. Everything seems fine, at least to a person that doesn't really have a clue what he's doing.

The only abnormality that I noticed was one of the washers in the RA axis wasn't dead flat, not sure if that'll cause issues once the weight of the bearings is on it.

[Shiraz (Ray)]

the backlash is a free region where the gears are not in physical contact and you can move the axis from side to side across the backlash gap by pulling and pushing on the OTA. But now imagine that you only pull on the OTA and hold it - the gears will be pulled into contact and if you keep pulling, the gears will stay in contact, regardless of which direction you drive. If you imbalance the load, you get the same result and take the backlash right out of the equation.

Don't be worried too much about DEC backlash - you can easily manage it and there is probably no need to try to repair your mount if you have ruled out all of the possibilities listed by Peter.

edit: thinking about it though, 25 steps is a lot - have you checked the settings in phd2 to make sure that the cal step size is big enough?

[peter_4059 (Peter)]

Lee are you saying you used to have 3-4 steps of backlash and now you have 25? That sounds like an adjustment issue rather than an inherent flaw in the mount

[codemonkey (Lee)]

Originally Posted by Shiraz

the backlash is a free region where the gears are not in physical contact and you can move the axis from side to side across the backlash gap by pulling and pushing on the OTA. But now imagine that you only pull on the OTA and hold it - the gears will be pulled into contact and if you keep pulling, the gears will stay in contact, regardless of which direction you drive. If you imbalance the load, you get the same result and take the backlash right out of the equation.

Don't be worried too much about DEC backlash - you can easily manage it and there is probably no need to try to repair your mount if you have ruled out all of the possibilities listed by Peter.

edit: thinking about it though, 25 steps is a lot - have you checked the settings in phd2 to make sure that the cal step size is big enough?

Cheers mate.

I guess I was thinking for DEC to guide in the opposite direction to the imbalance, it would need to first take up the slack. Bit if it falls rather than gets pushed due to the imbalance that would probably have the same effect, which is essentially what you're saying. Makes sense, thanks!

The step size is fine, RA and DEC calibrate in about 12 steps once the backlash has cleared.

[codemonkey (Lee)]

Originally Posted by peter_4059

Lee are you saying you used to have 3-4 steps of backlash and now you have 25? That sounds like an adjustment issue rather than an inherent flaw in the mount

I've seen 3-4 on other mounts I've owned in the past. This one wasn't bad in the beginning, but I don't recall exactly what it was. It has gotten worse, but it got worse before I adjusted anything. The only reason I ever adjusted anything was to try and fix it.

I managed to make it worse at one point, taking about ~42 steps to calibrate (PHD counted to 30, then decided it was moving, and took another ~12 steps before it actually did move), but I've never managed to make it better.

[codemonkey (Lee)]

Actually, thinking about it, I'm not sure if it actually ever did get worse.

Previously I was only guiding in one direction in DEC, and was intentionally slightly out of alignment. I've also started imaging at different parts of the sky, so it's possible when it didn't seem so bad before that it was a matter of how DEC approached the subjects... it could be that it was taking up the backlash getting to the subjects, then calibrating fine, and because I was only guiding in one direction I'd never have seen it.