Re: How to determine mechanical CWD position?
Thanks for the reminder.? I only saw the first video but not the second one.? I'm sure there's lots of good tips.
|
Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
Michael,
I replaced the gearbox on the RA axis when I first got the mount in 2013 with a McLennan gearbox, but if I remember correctly it has more backlash then the Losmandy one, so I still use the original gearbox on the Dec axis. The bulk of the backlash is due to the worm mesh anyways (no spring load on my mount), so the gearbox is not the main contributor on my mount.
Eric
|
Re: How to determine mechanical CWD position?
We recommend adjusting in one direction, I'm sure you know what to do but we also covered this and other best practices for polar align in this video
On Wed, Jul 14, 2021 at 08:18 AM, Brian Valente wrote:
I'm curious what results you are getting with the Sharpcap PA process??
The sharpcap results are fine, the problem is having to adjust the alt and az back and forth a lot every time at the beginning and end of polar alignment.
-- Brian?
Brian Valente portfolio
|
Re: At some point, I may begin commenting on the G11 DEC axis...
On Fri, Jul 16, 2021 at 04:29 AM, Henk Aling wrote:
Any idea what causes the positive feedback?? I've been wondering if the sampling delay could cause positive feedback if the sampling interval (exposure time) is too short especially for DEC.
Yes, but it's all speculative, so no comment.? There is no such thing as "too short exposure time" in a properly set up feedback algorithm.
|
Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
On Fri, Jul 16, 2021 at 02:17 PM, Michael Herman wrote:
That you show is typical of the stock gearbox play.? That's part #1 of hysteresis and you can't get rid of that unless you change gearbox model (to a McLennan type say). But I think all gearboxes have some hysteresis in the gearing.? (There is a spec in the McLennan types...see attached document.)
?
(...)?
?
If I left something out....I apologize.
?
For the DIY minded I would mention that OnStep requires no gearbox - the stepper is coupled directly to the worm axis.? Mode-switching stepper drivers slew the mount in slewing mode and track in micro-stepping mode.? I don't know if OnStep performs as well as Gemini but I have not seen or heard evidence of the contrary.
|
On Fri, Jul 16, 2021 at 08:06 AM, John Kmetz wrote:
On Thu, Jul 15, 2021 at 04:53 PM, alan137 wrote:
Even my new G11 DEC axis has this dark grey grease on the gears, and it is smudged on the sides, so 100% sure that's the original color.
The grease under the worm blocks is also this dark grey grease.
Anybody know what this is?
The JetLube MP-50, a molybdenum grease (used at the factory on the worms), is already is a blackish grey color when originally applied. SuperLube which many are using for their relubes, is clearish or slightly off water-white. Any grease will change color over time with use with oxidation and contamination from the metals it contacts. But color is not an indicator of lube degradation, if that is what you are thinking.
No, I am just using the grease color for identification, since I am a big proponent of not mixing greases.? Here's the grease colors that my mount came with: transfer gears and under the worm blocks : dark grey on the axis spindle (needle bearings): brownish clear worm: I dunno the tint, but NOT dark grey
|
Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
Eric ("cyclone")...
Very nice work!? I love that 2 image timed video photo of the Oldham coupler end!
That you show is typical of the stock gearbox play.? That's part #1 of hysteresis and you can't get rid of that unless you change gearbox model (to a McLennan type say). But I think all gearboxes have some hysteresis in the gearing.? (There is a spec in the McLennan types...see attached document.)
On the Tucked motor system you'd find additional time lag #2 (seen by the worm movement delay from the motor) from the necessary gaps between the 2 transfer gears.? You must not allow those transfer gears to jam so there has to be a little gap.? You can adjust the gearbox mounting to optimize that.? (The gear driving the Oldham coupler is on a fixed shaft...)
On the SLW you'd get yet another time lag #3 if the OPW that's pivoting at the left pivot point can move left to right in it's "nipple" pivot hole.? That also has to have a gap.? That affects the DEC autoguiding but not normally the RA that's always forcing the RA toward the west as the sky rotates.??
( So I think that bolting down the SLW so it does not move (defeating the SLW action) is a good approach to try eliminating that time delay.? )
Another time delay #4 affects the DEC mostly too is any worm left to right slip along it's axis if the worm mounting blocks get loose.? I think these blocks must get pushed loose over mount temperature changes anyway, so I recommend putting in one R4 Belleville spring under the far worm bearing, and reducing that bearing OD and lubing it to slide, to get rid of that time delay.
{ The G11T has 3 transfer gears at the RA Titan worm drive.? I found that a 2 pulley belt drive eliminates that time delay in those gears (that thanks to Allan Ruckle who showed me his nice implementation).? }
Then there is the worm to ring gear gap.? That was always the obvious place where a gap must be optimized....going back decades.? For that, we used to (and I still do) use spark plug feeler gauges to adjust that.? That's hysteresis element #5.? Many owners get rid of that gap by using a weighted cord wrapped around the axis. [On the RA axis, hang the weight to the East side to minimize chatter of the ring gear when the worm rubs it perpendicularly].?
Recent emails suggest on some units there could be a gap between the 1.25 inch shaft and the needle bearing ID.? I don't find that only units but it's another thing to test for.? I guess that is potential hysteresis element #6.?
Lastly, check that your dovetail mounting at the top of the DEC axis bolts are tight and that dovetail is not able to wobble.?
If I left something out....I apologize.
Have fun all,
Michael
How do I do that?
As briefly described in my previous post, I measured the backlash directly by taking pictures of the coupling. To be more specific:
-Remove the motor-gearbox assembly from the mount. Keeping half the coupling on the gearbox output shaft helps with the backlash measurement.?You should be able to rotate the output shaft with your fingers and feel the backlash.
-Take a pictures at each end of the backlash. I installed my camera on a tripod so I can easily align the images. The camera was facing the end of the shaft (or coupling half).
-In Photoshop, or other imaging software, measure the angle difference between the 2 images; this is the gearbox backlash. Convert the measured backlash to other units, for instance:
Backlash angle * 239.34 /? 360 deg = backlash in seconds at sidereal rate
-For the worm backlash, repeat the above taking pictures of the other coupling half on the worm which is still installed on the mount.
-If you believe there is backlash in the coupling itself, you could assemble the coupling, fix one end, and measure the backlash at the other end using the same approach.
Here is an example of the two pictures for the gearbox backlash, combined in a gif file:
Eric
|
Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
How do I do that?
As briefly described in my previous post, I measured the backlash directly by taking pictures of the coupling. To be more specific: -Remove the motor-gearbox assembly from the mount. Keeping half the coupling on the gearbox output shaft helps with the backlash measurement.?You should be able to rotate the output shaft with your fingers and feel the backlash. -Take a pictures at each end of the backlash. I installed my camera on a tripod so I can easily align the images. The camera was facing the end of the shaft (or coupling half). -In Photoshop, or other imaging software, measure the angle difference between the 2 images; this is the gearbox backlash. Convert the measured backlash to other units, for instance: Backlash angle * 239.34 /? 360 deg = backlash in seconds at sidereal rate -For the worm backlash, repeat the above taking pictures of the other coupling half on the worm which is still installed on the mount. -If you believe there is backlash in the coupling itself, you could assemble the coupling, fix one end, and measure the backlash at the other end using the same approach. Here is an example of the two pictures for the gearbox backlash, combined in a gif file: Eric
|
Re: New G11G Looking for final tweaking advice
Since
I identify with everything you write below, I would suggest you just try to
move the mount by hand to see how much play when you A) wiggle the end of the
counterweight shaft and B) at the camera end of the optics.. If it doesn¡¯t
move, leave it alone and carry on until you find a reason to adjust it..
?
If
it moves too much, then adjust the worms.. BUT! Your numbers already say leave
it alone..
?
Derek
?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Nick Ambrose
Sent: Friday, July 09, 2021 8:07
AM
To: [email protected]
Subject: Re: [Losmandy_users_io]
New G11G Looking for final tweaking advice
?
Thanks you both, Brian
& John
In my opinion, my images look amazing (for my level of experience at least - I
am far from an expert here)
My FWHM & Eccentricity are a bit higher than I'd like but I dont think this
is anything to do with the mount. I used the "Carey" mask after
auto-focus and unless I am mistaken, it was perfectly in focus, so any increase
in FWHM above what I want is due to my gear and conditions
Similar on the eccentricity, it's in the low 0.5s which is good, but not 0.4s.
I think this is the "duo-band" filter mostly as well as a little bit
of sensor tilt I think I saw using the CCD tools.
My guiding RMS is fantastic, even for a less forgiving imaging train
These are the results from 4.5 hours of guiding the other day
Total RMS? ? ? ? DEC? ? ? ?
? ? ? ? ? ? ? ? ? ? ?
? ? ? RA
0.62"? ? ? ? ? ? ? ?
?0.35"? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? 0.51"
0.59"? ? ? ? ? ? ? ?
?0.30"? ? ? ? ? ? ? ? ?
? ? ? ? ? ? ? ? 0.50"
0.53"? ? ? ? ? ? ? ?
?0.29" (this is 0.04pixel!)? ? ?0.45"
0.59"? ?
? ? ? ? ? ? ?0.32"? ? ?
? ? ? ? ? ? ? ? ? ? ?
? ? ? 0.49"
And this would be perfectly good even at 1 arcs/pxl
I watched the RMS (since I have to be with the mount when I am imaging) and I
never saw the spikes that David is seeing
Also, as John says, my 76.2 second peak looks pretty low (0.3" ?) so I
don't see that it would have a huge impact anywhere
I will definitely do the PEM Pro correction but this will have to wait until I
get an astro-cam or a better guider. Programming PEC with a 6.45 arcs/pxl guide
cam I don't think makes much sense
I do see fairly often, RA going to +/- 1.5-2" deviations (usually it is
well inside +/-1" on my graph
It seems like if my image scale was 1" / pxl, a deviation of 2" no
matter how short-lived would impact the image, since the sensor is continually
imaging, no ?
For me, that's not having an effect right now but I have no idea how (or even
if) I need or should try to correct that
I'm looking to image, not to mess with stuff (especially as I am not at all
mechanically minded).
I just don't want to take a lot of images and then realize there was an issue
which I should have corrected
Definitely not looking to stir things up here, trying to learn.
I'll be posting some final images as soon as possible for people to look at,
but the current one I am working on is far away from having enough integration
time yet ...
Thanks so much for the help & advice here
|
Re: New G11G Looking for final tweaking advice
David,
Looks like a common curve, nice overlapping data, Peak to Peak = 8.2, RMS = 2.8. Did you analyze, create a PEC curve and then upload to the mount when the pop-up prompted? This should be done while PEMPro and the mount are still synced. When you start your guiding session, then make sure the PEC box in the lower right of the Gemini app is checked for the curve to be applied.
|
>>>
The JetLube MP-50, a molybdenum grease (used at the factory on the worms), is already is a blackish grey color when applied at the factory.
Yep
the MP-50 is what we use on the worms, and the arctic?lube elsewhere
On Fri, Jul 16, 2021 at 8:07 AM John Kmetz < jjkmetz54@...> wrote: [Edited Message Follows]
On Thu, Jul 15, 2021 at 04:53 PM, alan137 wrote:
Even my new G11 DEC axis has this dark grey grease on the gears, and it is smudged on the sides, so 100% sure that's the original color.
The grease under the worm blocks is also this dark grey grease.
Anybody know what this is?
The JetLube MP-50, a molybdenum grease (used at the factory on the worms), is already is a blackish grey color when originally applied. SuperLube which many are using for their relubes, is clearish or slightly off water-white. Any grease will change color over time with use with oxidation and contamination from the metals it contacts. But color is not an indicator of lube degradation, if that is what you are thinking.
-- Brian?
Brian Valente portfolio
|
Re: Adding grease to existing greased worm and wheel
I
have ignored what PHD2 says I have for backlash completely, as I believe it is
more likely stiction and my guiding is rather good.. Also note that since the
mount ¡°works¡± perfectly for what I want to do, I don¡¯t fret over perfection as
I have much bigger problems with ¡°normal¡± things that no one else would have
problems with as my eyes are cratering from ¡°extreme Glaucoma¡±.. Becoming hard
to read the email too..
?
I
am not far from selling the entire setup and being quite depressed for the rest
of my days..
?
As
for the expense of this lubrication, I was quite lucky that the person who
recommended it, sold me enough from what he already had.. Also, I have no
backlash as measured by hand moving the counterweight shaft or eyepiece
diagonal.. My mount has never completed the PHD2 measurement of backlash (I
tried about a dozen times over 6 months) so I started ignoring it and things
have been great since then..
?
Derek
?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Jim Waters
Sent: Wednesday, July 07, 2021
3:18 PM
To: [email protected]
Subject: Re: [Losmandy_users_io]
Adding grease to existing greased worm and wheel
?
On Wed, Jul 7, 2021 at 02:12 PM, Derek C Breit wrote:
And a mixture of 3:1 Lubriplate 105 and Nye Fluorocarbon Gel 868H for
the worms
Does this help with backlash?
------------------------
Jim W
Phoenix, AZ. USA
Losmandy GM811G, NINA 1.11, ASI2600MC Pro, Sky-Watcher Scopes
|
On Thu, Jul 15, 2021 at 04:53 PM, alan137 wrote:
Even my new G11 DEC axis has this dark grey grease on the gears, and it is smudged on the sides, so 100% sure that's the original color.
The grease under the worm blocks is also this dark grey grease.
Anybody know what this is?
The JetLube MP-50, a molybdenum grease (used at the factory on the worms), is already is a blackish grey color when originally applied. SuperLube which many are using for their relubes, is clearish or slightly off water-white. Any grease will change color over time with use with oxidation and contamination from the metals it contacts. But color is not an indicator of lube degradation, if that is what you are thinking.
|
Another Approach to Worm Axle End Play
I've got a couple dead Meade LT6/8 mounts and have been toying of reviving them with OnStep. So I picked up a drive and worm block assembly yesterday and broke it down. Thought I'd share the end play adjuster as this subject comes up here.
The spring's tension is actually transmitted to the bearings in the motor. The bearings are handled by tight sleeve spacers.
|
Re: First image with GllG
Incredible result with fairly "humble" equipment. Not referring to the? G11G, of course?
|
Re: At some point, I may begin commenting on the G11 DEC axis...
On Fri, Jul 16, 2021 at 05:05 AM, alan137 wrote:
From the factory, the mesh was too tight and the axis had severe "forwards-lash" problems, where the DEC guiding would start to drift a little south, but north pulses made it go even further south in a case of positive feedback.? After a few seconds, the DEC value would (over)shoot back up and this cycle repeats back and forth a few times.? In terms of the PHD backlash graph, it had the slight overshoot characteristics that I mentioned in my long post about the G8 axis.? After a few rounds of slightly tightening the "back off bolt", I was able to get very good backlash values of less than a second.? (according to the plot, not the number that PHD calls out, which can be somewhat random)
Any idea what causes the positive feedback?? I've been wondering if the sampling delay could cause positive feedback if the sampling interval (exposure time) is too short especially for DEC.
BTW, does anybody know how to make PHD do less than 500ms pulses on the backlash test?
Ekos with OnStep allows a pulse time interval of any positive value FWIW.? Alternatively, can't you just lower the guide rate if the goal is to achieve a minimum step change?
|
Re: How to determine mechanical CWD position?
This is what I ended up doing:
Set the RA axis position via bubble levels.
To set the dec position:
1) Perform a sharpcap polar alignment
2) set the mount back to CWD position
2.5) restart the sharpcap polar alignment, but don't go past the first screen
3) manually move the DEC axis until the pole is centered left-right
4) (This also made me realize that I needed to shim up one end of the telescope dovetail to make the pole centered up-down)
5) now mark the DEC axis
The whole point of doing this procedure is that next time I do a polar alignment, I can center the pole on the first screen of the procedure, and after rotating the RA axis 90 degrees, there should only be minor adjustments required.
|
Re: How to determine mechanical CWD position?
On Wed, Jul 14, 2021 at 08:18 AM, Brian Valente wrote:
I'm curious what results you are getting with the Sharpcap PA process??
The sharpcap results are fine, the problem is having to adjust the alt and az back and forth a lot every time at the beginning and end of polar alignment.
|
Runaway slew while using PHD2
Hello folks,?
Despite using the latest 2018 Beta firmware and the ASCOM SP1 drivers, I still had the problem where issuing a goto under Stellarium causes the mount to run off in some random direction, and there was no STOP button available on the hand controller, so I had to quickly shut off the mount.? The problem seems to be due to a PHD2 setting about "stop guiding when mount begins slewing".? I accidentally left that unchecked when setting up a new profile, so it was probably the source of the problem.? Usually this happens when I only need to move over a little bit, like centering a nearby bright star to use a Bahtinov focus mask.? And I can't be expected to remember to turn off guiding every time.
Still, I don't really know why it happened.? Maybe PHD tries to apply some guiding pulses, but that shouldn't cause the mount to go flying off.? Second, why is it that the mount can go at full slew speed but doesn't give me a STOP button in this case?? It's reasons like this that I don't trust Gemini2 for fully remote imaging.
|
At some point, I may begin commenting on the G11 DEC axis...
I got a new one recently and have been testing it with a 12" F4 solid tube newtonian, approx 50 lbs.
The weather has been spotty clouds, so no real testing, but initial impressions are positive.
From the factory, the mesh was too tight and the axis had severe "forwards-lash" problems, where the DEC guiding would start to drift a little south, but north pulses made it go even further south in a case of positive feedback.? After a few seconds, the DEC value would (over)shoot back up and this cycle repeats back and forth a few times.? In terms of the PHD backlash graph, it had the slight overshoot characteristics that I mentioned in my long post about the G8 axis.? After a few rounds of slightly tightening the "back off bolt", I was able to get very good backlash values of less than a second.? (according to the plot, not the number that PHD calls out, which can be somewhat random)
So far, it's been a lot easier to test and adjust the backlash than on the G8, where sometimes I couldn't tell whether the worm was too tight or loose, or simply could not get the backlash down to less than a few seconds. If adjustments are this straightforwards, and there aren't any other weird problems, then I would be fine with just spending a few extra minutes every time making an adjustment, rather than trying to come up with some elaborate "spring-loaded worm" mechanism.
For now, it seems the G11 DEC axis is more than twice as good as the G8.
BTW, does anybody know how to make PHD do less than 500ms pulses on the backlash test?
|
Michael Herman