Mini Report of My Ritchey Cretien First Light! ( M13 Image/hardly Post processed yet! )

[JeremyS]

👆🏻Olly’s response on the levelling-an-EQ-mount-tripod-fallacy should be a sticky somewhere given the number of times it comes up.

[ollypenrice]

All these mounts share identical polar alignment but only one rides on a pier pointing to the centre of the earth, meaning it has a level top.

The tracking at sidereal rate drives the RA housing round against a non-moving part of the mount at the correct speed. That non-moving part can be at any angle, it just needs not to move. The Go To function, likewise, is calibrated by the fixed encoders being informed of the position of the moving ones when pointed at the alignment stars. With no cone error and and perfect PA, a one star alignment contains all necessary information. (Or almost all, since atmospheric diffraction displaces the apparent position of stars slightly. This is only of interest to mounts with absolute encoders working without autoguiding.)

Olly

[wimvb]

In days past, eq3 and eq5 users would set up their tripods off level in order to avoid bending the altitude bolts of their mount. If you have the North leg of the tripod longer, you can polar align at a lower altitude than your local lattitude, and avoid unnecessary stress on the bolts.

[ollypenrice]

8 hours ago, wimvb said:

In days past, eq3 and eq5 users would set up their tripods off level in order to avoid bending the altitude bolts of their mount. If you have the North leg of the tripod longer, you can polar align at a lower altitude than your local lattitude, and avoid unnecessary stress on the bolts.

Yes, and a solution well known to northern users of Takahashi mounts which run out of altitude adjustment long before you get to Spitzbergen!  Just stick a brick under the northern leg of the tripod.

Olly

[Ibbo!]

13 minutes ago, ollypenrice said:

Yes, and a solution well known to northern users of Takahashi mounts which run out of altitude adjustment long before you get to Spitzbergen!  Just stick a brick under the northern leg of the tripod.

Olly

Been there (Kielder SP) & got the T shirt

 

[wimvb]

I experienced the arched bolt conundrum with my first mount, an eq3. So, after having learnt my abc’s of astrophotography, I chose as my second mount the AZ-EQ6, just because it had a different altitude adjustment arrangement. I never needed to resort to a brick.

[wesdon1]

On 03/07/2024 at 07:19, ollypenrice said:

Your sketch ignores the polar alignment which has been carried out on top of both tripods in the picture.

Imagine this: take your perfectly polar aligned mount and run a snug fitting steel shaft through its RA axis. Bolt one end of this shaft to the floor and the other to the wall. (Inconvenient, I know! ) Your polar alignment is now permanently fixed and perfect. Next, shorten your tripod legs so they don't reach the ground, loosen the level adjusters and swing the tripod around to an assortment of non-level positions. Does this alter your polar alignment? Nope, not at all.

Takahashi mounts have no levelling facility built into them but their polar alignment routine through the finderscope is by far the best in there is.

The Mesu in the picture is not level, the pier top is at the equatorial angle. The screw feet are there to perfect that angle, not to level anything.

Olly

@ollypenrice Hi Olly. I just did a bit of a thought experiment with your information, and I do now think you're correct. When the mount is pointing directly at NCP, the level isn't going to affect RA or DEC because the mount is effectively independent to the tripod, because the DEC axis is fixed relative to RA axis, regardless of the level of the tripod. So I think I now get it, finally! lol. Just as a side note though, I believe the tracking would still benefit from at least a roughly level tripod, because a very un-level mount would make your telescope and cameras/counter weights etc heavier in certain RA and DEC positions, messing up balance and thus messing up tracking accuracy? Am I correct? 

Thanks for educating me Olly, I'm always open and indeed grateful for more experienced peoples advice and guidance! 

Kind Regards, Wes.

[catburglar]

You’re balancing the OTA around the RA/Dec axes, so level won’t affect tracking

[ollypenrice]

29 minutes ago, wesdon1 said:

 I believe the tracking would still benefit from at least a roughly level tripod, because a very un-level mount would make your telescope and cameras/counter weights etc heavier in certain RA and DEC positions, messing up balance and thus messing up tracking accuracy? Am I correct? 

No, this isn't true either. Your RA axis is precisely the same as the polar axis and that is the axis (the only axis) around which your counterweights must balance the scope. Balance in Dec is, likewise, balance on either side of the Dec axis. Note that the Dec axis is not level because it is borne by the RA axis at the equatorial angle - but it is, none the less, perfectly possible to balance.

If you balance a wheel you can tip its axle to any angle and it will remain balanced. If this were not true, bicycle and motorcycle wheels would go out of balance as the machine banked in a corner. Fortunately for those of us who like riding bikes, this is not so!

Olly

Edit, crossed with Catburglar who said the same with greater brevity! 

By the way, the problem of variable balance in different parts of the sky does exist and is caused by lack of dynamic balance. The simplest balancing scenarios say slide the CW up and down the bar till the scope will sit in any RA position and slide the scope back and forth in the saddle plate till it will sit still in any Dec position.

This assumes that, as you look up along the tube from the back, the left hand side of the the scope weighs the same as the right hand side - but does it? Finders? Guidescopes? Robotic focusers? Here you have to set the CW bar horizontal and the scope vertical and then balance the left and right hand sides of the OTA.

Edited July 4 by ollypenrice

[wesdon1]

8 minutes ago, ollypenrice said:

No, this isn't true either. Your RA axis is precisely the same as the polar axis and that is the axis (the only axis) around which your counterweights must balance the scope. Balance in Dec is, likewise, balance on either side of the Dec axis. Note that the Dec axis is not level because it is borne by the RA axis at the equatorial angle - but it is, none the less, perfectly possible to balance.

If you balance a wheel you can tip its axle to any angle and it will remain balanced. If this were not true, bicycle and motorcycle wheels would go out of balance as the machine banked in a corner. Fortunately for those of us who like riding bikes, this is not so!

Olly

Edit, crossed with Catburglar who said the same with greater brevity! 

@ollypenrice Oh ok I see what you mean Olly. Thanks again Olly! I really do appreciate the advice and guidance!

Kind Regards, Wes.

[Louis D]

I would think that if you greatly extended the north leg until the other two legs are nearly vertical, you would end up with an unstable rig prone to tipping over southward.

[wimvb]

2 hours ago, Louis D said:

I would think that if you greatly extended the north leg until the other two legs are nearly vertical, you would end up with an unstable rig prone to tipping over southward.

Yes. This must also be a problem with a harmonic drive mount without counterweights. It must be very unbalanced when pointing at zenith.

[Louis D]

20 minutes ago, wimvb said:

Yes. This must also be a problem with a harmonic drive mount without counterweights. It must be very unbalanced when pointing at zenith.

I would probably use the harmonic drive in alt-az at first, so quite similar to an offset alt-az manual mount as far as keeping the legs spread wide to avoid tipping.

[wesdon1]

On 03/07/2024 at 09:30, JeremyS said:

👆🏻Olly’s response on the levelling-an-EQ-mount-tripod-fallacy should be a sticky somewhere given the number of times it comes up.

@JeremyS Indeed! Jeremy I see this topic being passionately debated across the board online! I think someone needs to do a definitive real world test, imaging with a really un-level mount/tripod, and then imaging same night with perfectly level mount, and post the video on YouTube? That might once and for all settle the debate? ( probably not!? LOL ) In fact, although I now accept @ollypenrice arguments/advice about the issue, I might just make said video anyway? It would be very interesting!? 

[Ibbo!]

6 hours ago, Louis D said:

I would think that if you greatly extended the north leg until the other two legs are nearly vertical, you would end up with an unstable rig prone to tipping over southward.

That is a centre of gravity issue not a level issue.

 

My NJP is not level on its pier but nicely polar aligned, the concrete former moved a bit while pouring.

The EQ6 gets polar aligned but not leveled on its tripod - it is about level (well a blind man on a passing horse would not notice) solar I can leave for hours and apart from PE it tracks very well also with the C11 and 2.5 powermate using  ROI for planetary.

When I use it for deep sky longer subs it is left a bit off PA as it has bad backlash in dec so I only guide in one direction.

 

Every time I swap scopes it moves a bit so hence a quick PA is required.

 

[ollypenrice]

6 hours ago, wesdon1 said:

@JeremyS Indeed! Jeremy I see this topic being passionately debated across the board online! I think someone needs to do a definitive real world test, imaging with a really un-level mount/tripod, and then imaging same night with perfectly level mount, and post the video on YouTube? That might once and for all settle the debate? ( probably not!? LOL ) In fact, although I now accept @ollypenrice arguments/advice about the issue, I might just make said video anyway? It would be very interesting!? 

Takahashi mount owners do this test nearly every time they set up their mounts but, yes, why not? On the other hand it is perfectly possible to see from pure thought, or even by looking at an English Equatorial mount.

The two square columns which carry the polar aligned beam are, in this case, vertical and built on a level floor. It is quite obvious, though, that they could emerge from any angle (out of a wall, for example) and still hold the polar aligned beam exactly where it is now. 

Olly

[wimvb]

On 03/07/2024 at 13:57, ollypenrice said:

All these mounts share identical polar alignment but only one rides on a pier pointing to the centre of the earth, meaning it has a level top.

The tracking at sidereal rate drives the RA housing round against a non-moving part of the mount at the correct speed. That non-moving part can be at any angle, it just needs not to move. The Go To function, likewise, is calibrated by the fixed encoders being informed of the position of the moving ones when pointed at the alignment stars. With no cone error and and perfect PA, a one star alignment contains all necessary information. (Or almost all, since atmospheric diffraction displaces the apparent position of stars slightly. This is only of interest to mounts with absolute encoders working without autoguiding.)

Olly

This is indeed a very clear explanation. One can only add that when the tripod or pier top isn’t level, the altitude an aximuth controls of the mount won’t allow you to adjust altitude and azimuth independent from each other. Polar alignment may become confusing, but it’s still possible.

[ollypenrice]

3 hours ago, wimvb said:

This is indeed a very clear explanation. One can only add that when the tripod or pier top isn’t level, the altitude an aximuth controls of the mount won’t allow you to adjust altitude and azimuth independent from each other. Polar alignment may become confusing, but it’s still possible.

Yes, that's true, but in reality it takes more time to level a pier precisely than it does to align using, for example, Takahashi's polar scope method which is still the fastest there is, faster than software based methods. A perfect polar alignment is not even, necessarily, a good thing. It puts both axes in the backlash zone if you are using a gear-driven mount.

Olly

[wesdon1]

On 05/07/2024 at 18:29, ollypenrice said:

Yes, that's true, but in reality it takes more time to level a pier precisely than it does to align using, for example, Takahashi's polar scope method which is still the fastest there is, faster than software based methods. A perfect polar alignment is not even, necessarily, a good thing. It puts both axes in the backlash zone if you are using a gear-driven mount.

Olly

@ollypenrice Wow I never even considered backlash from perfect PA?? I have little issues with backlash on my HEQ5, at least nothing that I can notice? What are the biggest symptoms of backlash on a mount, and is there any solutions? Thanks!

[wimvb]

1 hour ago, wesdon1 said:

@ollypenrice Wow I never even considered backlash from perfect PA?? I have little issues with backlash on my HEQ5, at least nothing that I can notice? What are the biggest symptoms of backlash on a mount, and is there any solutions? Thanks!

Backlash only shows if gears are moved in both directions, back and forth. If your mount shows severe backlash, you can feel it if you wiggle the DEC or RA axis. Moderate backlash shows in the guide graph. PHD will send guide pulses, but they don't seem to do anything, then suddenly one more pulse moves the axis a lot. This usually looks like a saw tooth pattern in the guide graph. You measure backlash with the guide assistant in PHD. If PHD needs to compensate for more than a few 100 ms, it's a good thing to tighten the worm gear. But too much will cause binding of the gear at another position, that's why it's best to not try to get backlash down to zero. PHD can compensate for moderate backlash by sending an extra pulse whenever it senses a change in guiding direction.

If a gear has backlash, there is a small gap between the teeth of both wheels. When a mount isn't balanced, the axis that has backlash tends to "fall" back to one side, untill the teeth touch each other. If the axis moves in the direction such that it always pushes the gear, you won't notice anything. But if the axis moves in the opposite direction, it will always try to 'fall back' a little after each move, and that axis seeems sluggish to react to small movements. By changing the imbalance, you can see to it that the gears always push. The RA axis for example always tracks (moves) in one direction. If you make the RA axis slightly east heavy, you will have the gear always pushing. For this reason, the guide speed should also be always less than sidereal speed (1 x in PHD settings). If you set a higher speed than this, PHD can cause a reversal in RA direction, and RA backlash can start to show.

By introducing a small polar misalignment, you can also have the DEC axis constantly move in one direction. A small imbalance here will work the same as for RA.

[Louis D]

Why is it that harmonic drives also have gears, but don't seem to suffer from backlash?  I've never fully comprehended this.

[ollypenrice]

10 hours ago, Louis D said:

Why is it that harmonic drives also have gears, but don't seem to suffer from backlash?  I've never fully comprehended this.

In my partial understanding the red flexispline is deformed by the green wave generator until its teeth are in contact with the circular blue gear. I suspect (rightly or wrongly) that this contact arrests the deformation of the flexispline at the point of no backlash.

Any engineers out there?

Olly

[wimvb]

Backlash only shows when either the rotation is reversed, or when there is an imbalance, and the secondary part (wheel) of the gear "falls" back. For backlash to occur at all, there must also be a gap between teeth of each part (see image), so that only one side of each tooth of the driving wheel pushes against one side of each tooth of the secondary wheel. In the harmonic drive animation, there is a constant motion pushing the secondary section, and at the point of contact, the entire tooth makes contact. There is no room for the secondary to fall back. In a two wheel spur gear system the gear would be binding if it were this tight.

I'm not a mechanical engineer either, but this is how I understand it.

[wesdon1]

On 13/07/2024 at 14:29, wimvb said:

Backlash only shows if gears are moved in both directions, back and forth. If your mount shows severe backlash, you can feel it if you wiggle the DEC or RA axis. Moderate backlash shows in the guide graph. PHD will send guide pulses, but they don't seem to do anything, then suddenly one more pulse moves the axis a lot. This usually looks like a saw tooth pattern in the guide graph. You measure backlash with the guide assistant in PHD. If PHD needs to compensate for more than a few 100 ms, it's a good thing to tighten the worm gear. But too much will cause binding of the gear at another position, that's why it's best to not try to get backlash down to zero. PHD can compensate for moderate backlash by sending an extra pulse whenever it senses a change in guiding direction.

If a gear has backlash, there is a small gap between the teeth of both wheels. When a mount isn't balanced, the axis that has backlash tends to "fall" back to one side, untill the teeth touch each other. If the axis moves in the direction such that it always pushes the gear, you won't notice anything. But if the axis moves in the opposite direction, it will always try to 'fall back' a little after each move, and that axis seeems sluggish to react to small movements. By changing the imbalance, you can see to it that the gears always push. The RA axis for example always tracks (moves) in one direction. If you make the RA axis slightly east heavy, you will have the gear always pushing. For this reason, the guide speed should also be always less than sidereal speed (1 x in PHD settings). If you set a higher speed than this, PHD can cause a reversal in RA direction, and RA backlash can start to show.

By introducing a small polar misalignment, you can also have the DEC axis constantly move in one direction. A small imbalance here will work the same as for RA.

@wimvb Hi and thank you so much for such a comprehensive explanation, it makes so much sense to me! I actually have a background in building, modifying Radio Controlled RC Cars, Planes, helicopters etc etc so if I do ever need to open HEQ5 up to work on it, I should be very comfortable. Thanks again, I'd be lost without the more experienced members like yourself!

Off subject somewhat, I was/am considering either "belt modding" my HEQ5 or possibly buying a EQ6-R Pro. I'll probably go for the latter later this year when funds allow. I am guessing the EQ6 will be even better made and have more robust gearing than my well made HEQ5?

Again Thank You so much for all your advice and help, It really is appreciated!

Kindest Regards, Wes.

 

[wesdon1]

1 hour ago, wimvb said:

Backlash only shows when either the rotation is reversed, or when there is an imbalance, and the secondary part (wheel) of the gear "falls" back. For backlash to occur at all, there must also be a gap between teeth of each part (see image), so that only one side of each tooth of the driving wheel pushes against one side of each tooth of the secondary wheel. In the harmonic drive animation, there is a constant motion pushing the secondary section, and at the point of contact, the entire tooth makes contact. There is no room for the secondary to fall back. In a two wheel spur gear system the gear would be binding if it were this tight.

I'm not a mechanical engineer either, but this is how I understand it.

I've never seen inside a Harmonic Drive but from the animations it appears to me as though you're 100% correct. Those Harmonic Drives are such a game changer for the hobby. I will eventually own one but not for a couple of years, I don't think! I'm getting an EQ6-R Pro first! lol.