Re: G8 DEC axis backlash, fixing, questions
I normally get around 1000 for dec backlash. Phd2 manages it very well. As long as I have pa below 5 arc min I usually get dec rms well below 1. If I try and remove the backlash by tighting the worm the backlash does improve, but at the expense of rms and good images.?
|
Re: G8 DEC axis backlash, fixing, questions
>>>
I always assumed DEC backlash was somewhat consistent.? This make it harder to correct.
PHD's auto backlash compensation does that for you, which is why you want to use that approach
On Wed, Apr 7, 2021 at 11:02 AM Jim Waters < jimwaters@...> wrote: I always assumed DEC backlash was somewhat consistent.? This make it harder to correct.
------------------------
Jim W
Phoenix, AZ. USA
-- Brian?
Brian Valente portfolio
|
Re: G8 DEC axis backlash, fixing, questions
I always assumed DEC backlash was somewhat consistent.? This make it harder to correct.
------------------------
Jim W
Phoenix, AZ. USA
|
Re: G8 DEC axis backlash, fixing, questions
You can - sort of
you can measure it by summing the pulse guides until it reverses direction
But it won't be consistent for comparison or understanding over time. Backlash changes through the night depending on altitude, side of pier, shifting gear, etc. so it's an approximation
You could?measure it (and repeat measure it) guiding from a fixed point in the sky like intersection of meridian/celestial equator?
Can you determine backlash from the logs or do you need to run Guiding Assistant?
------------------------
Jim W
Phoenix, AZ. USA
-- Brian?
Brian Valente portfolio
|
Re: G8 DEC axis backlash, fixing, questions
Can you determine backlash from the logs or do you need to run Guiding Assistant?
------------------------
Jim W
Phoenix, AZ. USA
|
Re: Pixinsight Tutorials?
PS funny you asked this, i was literally processing an image with WBPP just now
This is a couple hours on eta carinae nebula, just WBPP and some curves. so even the basic processing works pretty well.
On Wed, Apr 7, 2021 at 8:51 AM Scott Popjes via <imagemover= [email protected]> wrote: I am taking the plunge into learning Pixinsight.? ?Anyone have any recommendations for good tutorials on YouTube?
-- Brian?
Brian Valente portfolio
|
Re: Pixinsight Tutorials?
Hi Scott
My suggestion is starting with WBPP (weighting batch preprocessing script) which can get you from data to an RGB image in the fastest way possible while still learning an approach that will work for you as you get better at PI
There are four main options:
FREE - light vortex astronomy has some good pages. they are quite outdated but relatively easy to follow?? - youtube - tons of stuff on youtube. not a huge fan because it's a bit piecemeal but there's a lot of content there
PAID - Rogelio has a book on learning PI. it's popular. more of a reference than a tutorial - warren keller has ip4ap subscription. tons of content, imo more of a reference? - adam block has fundamentals? which are great and very thorough but quite detailed and long
Personally, i really like starting with Adam Block's FastTrack. it's like $40 and uses WBPP to go from zero to finished pic inside a day. it's not everything, but a great place to start
On Wed, Apr 7, 2021 at 8:51 AM Scott Popjes via <imagemover= [email protected]> wrote: I am taking the plunge into learning Pixinsight.? ?Anyone have any recommendations for good tutorials on YouTube?
-- Brian?
Brian Valente portfolio
|
I am taking the plunge into learning Pixinsight.? ?Anyone have any recommendations for good tutorials on YouTube?
|
Re: Nice discussions leading toward perfection...
On Wed, Apr 7, 2021 at 09:44 AM, Tom & Barbara Coverdale wrote:
Thanks John! That was generally what I thought but I was wondering if there was a focal length above which guiding became a necessity. I enjoy the educational value of this thread. I plan to keep trying to get my guiding working. Thanks again. Tom
Tom, I also have a 102mm scope with a 0.7X focal reducer (500mm FL) and I always use PHD2 to optimize guiding. The FL of your main scope times 0.2-0.25 is about where the focal length of your guide scope should be. But it sounds like you may already have one. What are your PHD2 issues? Connection or operation? John
|
Re: Nice discussions leading toward perfection...
Thanks John! That was generally what I thought but I was wondering if there was a focal length above which guiding became a necessity. I enjoy the educational value of this thread. I plan to keep trying to get my guiding working. Thanks again. Tom
--- jjkmetz54@... wrote: From: "John Kmetz" <jjkmetz54@...> To: [email protected]Subject: Re: [Losmandy_users_io] Nice discussions leading toward perfection... Date: Tue, 06 Apr 2021 23:01:51 -0700
On Tue, Apr 6, 2021 at 10:16 PM, Tom & Barbara Coverdale wrote:
Mike et al,
This is a very interesting discussion and I am following?every comment.?But as a noob I often have more questions than answers! The one?question thats nags me thru these guiding and technical discussions is "when does one need to be this technically?sophisticated
with their guiding?" For instance I have a GM-8 and a Stellarvue 102mm and generally do not guide since I?fail in so many ways to get guiding to work for me.I?typical shoot 30sec to 3 min exposures depending on?targets and "seeing" from Atmospheric. I like my images and am learning both the?nuances of imaging but the workflow and challenges of Pixinsight. So at what focal length, exposure time etc does guiding become?truly?necessary??Thanks. Tom
Tom, Most people use PHD2 or another app for correcting their guiding unless they are very happy with the tracking of their mount. Super steady tracking allows for sharper and rounder stars and can be used for much longer exposure times for your subframes. More advanced astroimagers are willing to pay a premium for the best possible tracking and are willing to buy mounts which can approach $10K or even much more. The tech discussions you see here are mostly centered around squeezing the maximum performance out of the moderately priced Losmandy mounts.? So you will see much about fine tuning the mechanical drive train, or using various software settings to gain the most advantage. The longer the focal length of your scope and the smaller the arcsecond diameter of the astronomical target of interest, the more accurate the guiding needs to be.? If you are happy with your images, then you are good. The tech talk from all the scientific and engineering folks is more an expression of interest in the topic, and their willingness to apply their talents to maximizing operation. It can be overwhelming and you don't really need to absorb it all to enjoy the basic use of your mount. John
?
|
Re: G8 DEC axis backlash, fixing, questions
On 4/6/2021 11:51 PM, Brian Valente wrote: you can try TVC 10. if it's working well, you should see backlash amount decrease (but not be eliminated)
if it's too much your DEC will oscillate. it will be pretty obvious. just drop it by a point or two.
10 is pretty low. you could also increase it if you want to see more improvement. I probably wouldn't go past 25 for now FWIW..... I find that my GM811 DEC backlash as reported by PHD2 varies between something less than 200mS and about 1800mS depending on where I am pointing and possibly other unknown variables. Typical is around 1200mS. I can set TVC to a non-zero value and reduce the PHD reported backlash nicely. In my case, TVC=10 seems to be a good number. But that runs the risk of reducing the overall backlash to (or below) zero depending on pointing location. If backlash approaches zero, the system can oscillate. This is especially noticeable at the start of a new session or after a slew because PHD2 may be attempting to compensate for more backlash than is currently present. I have decided to use TVC=0 as long as the reported DEC backlash is less than 2000mS. PHD2 can easily compensate for that amount of backlash and the only observable effect is that on DEC reversals PHD2 appears to wait for the duration of the applied backlash compensation before beginning a new exposure. Paul -- Paul Goelz Rochester Hills, MI USA pgoelz@... www.pgoelz.com
|
Re: G8 DEC axis backlash, fixing, questions
Well in my case the more you tighten the clutch, the tighter the worm mesh gets.? So either the ring gear gets pulled down a little with a tighter clutch, or at low clutch pressure, the upper bearing is not sufficiently preloaded, and the pressure from the worm pushes the ring gear / axis away.? The way to check is to replace the clutch pad with some imcompressible material and then measure the gap between the setting circle and main body with feeler gauges while tightening the axis.? But I didn't want to bother going thru all the trouble of having the clean and then regrease the shaft to take out the clutch pad.
|
Re: G8 DEC axis backlash, fixing, questions
On Tue, Apr 6, 2021 at 04:09 PM, Henk Aling wrote:
Does this mean that the back-off screw is meant to keep the worm from riding up the ring gear because if it does, the assembly wobbles?? That would be disappointing.
I have never seen the Losmandy spring loaded worms in detail other than the assembly/adjustment video.? If it has a wobble problem, would it help to pull the block down at the pivot by applying a (Belleville) spring underneath it?? An old design by Chip on CN had springs at the bottom IIRC.? In the worst case add a brace around the worm assembly that provides a pivot at the top to keep it from wobbling.? Pivots on both sides should greatly reduce the wobble.
It's the preload from the regular screw (with spring) that keeps the worm from riding up. I tried putting an o-ring underneath the allen head for the block bolts, but it didn't make a difference.
|
Re: Nice discussions leading toward perfection...
On Tue, Apr 6, 2021 at 10:16 PM, Tom & Barbara Coverdale wrote:
Mike et al,
This is a very interesting discussion and I am following?every comment.?But as a noob I often have more questions than answers! The one?question thats nags me thru these guiding and technical discussions is "when does one need to be this technically?sophisticated
with their guiding?" For instance I have a GM-8 and a Stellarvue 102mm and generally do not guide since I?fail in so many ways to get guiding to work for me.I?typical shoot 30sec to 3 min exposures depending on?targets and "seeing" from Atmospheric. I like my images and am learning both the?nuances of imaging but the workflow and challenges of Pixinsight. So at what focal length, exposure time etc does guiding become?truly?necessary??Thanks. Tom
Tom, Most people use PHD2 or another app for correcting their guiding unless they are very happy with the tracking of their mount. Super steady tracking allows for sharper and rounder stars and can be used for much longer exposure times for your subframes. More advanced astroimagers are willing to pay a premium for the best possible tracking and are willing to buy mounts which can approach $10K or even much more. The tech discussions you see here are mostly centered around squeezing the maximum performance out of the moderately priced Losmandy mounts.? So you will see much about fine tuning the mechanical drive train, or using various software settings to gain the most advantage. The longer the focal length of your scope and the smaller the arcsecond diameter of the astronomical target of interest, the more accurate the guiding needs to be.? If you are happy with your images, then you are good. The tech talk from all the scientific and engineering folks is more an expression of interest in the topic, and their willingness to apply their talents to maximizing operation. It can be overwhelming and you don't really need to absorb it all to enjoy the basic use of your mount. John
?
|
Re: G8 DEC axis backlash, fixing, questions
>>>Won't the Losmandy / PHD2 Anti-TVC Gods come down and zap me for using it?????
haha no. not anymore! it's a slightly different approach than before. Before I think people were trying to eliminate backlash, and it was not a good approach
you can try TVC 10. if it's working well, you should see backlash amount decrease (but not be eliminated)
if it's too much your DEC will oscillate. it will be pretty obvious. just drop it by a point or two.?
10 is pretty low. you could also increase it if you want to see more improvement. I probably wouldn't go past 25 for now
Won't the Losmandy / PHD2 Anti-TVC Gods come down and zap me for using it???? Sure; I am willing to try.? I need to order some but I want to try using?NyoGel 774 Damping Grease on some of the DEC gear to gear surfaces to see what happens (Not on the worm).? I DO NOT recommend anybody do this.? Let me try it first. I will supply before and after PHD2 logs.?
I will try setting TVC to 10 first and supplying the before and after PHD2 logs also.? How would I know that PHD2 is having an issue with this - TVC set to 10?
For me and I suspect many other DEC backlash it the weakest point in the Losmandy design.? I am not sold on the SLW being the fix for this issue.??
------------------------
Jim W
Phoenix, AZ. USA
-- Brian?
Brian Valente portfolio
|
Re: G8 DEC axis backlash, fixing, questions
Won't the Losmandy / PHD2 Anti-TVC Gods come down and zap me for using it???? Sure; I am willing to try.? I need to order some but I want to try using?NyoGel 774 Damping Grease on some of the DEC gear to gear surfaces to see what happens (Not on the worm).? I DO NOT recommend anybody do this.? Let me try it first. I will supply before and after PHD2 logs.?
I will try setting TVC to 10 first and supplying the before and after PHD2 logs also.? How would I know that PHD2 is having an issue with this - TVC set to 10?
For me and I suspect many other DEC backlash it the weakest point in the Losmandy design.? I am not sold on the SLW being the fix for this issue.??
------------------------
Jim W
Phoenix, AZ. USA
|
Re: G8 DEC axis backlash, fixing, questions
Henk, et al,
Just a quick overview...
The thread here has focussed attention on the newest version GM811 units with tucked motors and spring loaded worms.??
The major problem reported by many is a high hysteresis? meaning a long time lag between telescope movement and a worm direction reversal. This is making autoguiding difficult or ...PHD2 fails.?
Several possible causes must be ruled out one at a time.? Here is a checklist of tests and procedures:
1. Make the worm to ring gear mesh tight and look for a loose dovetail or slipping clutch disk.? ?How: On the DEC axis drive, back out the "back off" small bolt near the spring, until the worm will not turn.? Do not run the motor.? Grab the dovetail and try to wiggle it.? It should be completely immobile.? If it wiggles, suspect the dovetail bolts are either loose or bottomed out.? If bottomed out you must file the end down to shorten the bolt. Get the dovetail to be tightly held.
2. The GM8 clutch disks are small 3.0 inch OD and 1.25 inch ID.? Does the DEC axis clutch slip?? If so the disk may be oily.? Take the DEC axis apart and clean the plastic and mating metal surfaces. See attached PDF about alternative clutch pads.?
3. Now the worm still will not turn because we jammed it.? But the worm might still shift left to right.? Let's look at the chance the worm assembly can shift around.? How: when you grab the dovetail and try to rotate it, the worm is attached.? See if you notice the worm OPW assembly shift while you are wobbling the dovetail around in the DEC direction.? If the OPW can shift, then extra time will be needed for the autoguide system to respond.??
See if you can determine the cause of the shift.? You can first tighten the left pivot point block bolt.? Now wobble the DEC axis.? Do you see the worm move?? If the worm moves then the end block is not tight enough toward the left bearing block.? ?Perhaps the bolt holding the far block has loosened or shifted.? Fix this (or add a Belleville spring...some work here to remove and undersize the bearing, do the spring will work).??
4. If the worm won't move by wobbling the DEC dovetail, great.? Next step is to see how much the transfer gear on the gearbox needs to move before the DEC axis responds.? How: remove the DEC motor.? This allows the gearbox output shaft to freely turn.? Try to rotate the gearbox output gear back and forth.? How far can you rotate that before it feels the jammed worm?? If it rotates a lot, sustect a loose Oldham coupler metal part.? To get to that, bolt down both left and right bearing blocks.? Then remove the thick aluminum OPW cover.? Now you can see the Oldham coupler completely. Ensure both metal ends are tight.? (I often take out the tiny grub setscrews, and drill and tap to install 4-40 setscrews.? Those enable me to use a larger hex socket wrench.? If you need some if those set screws contact me.)
5.? Also check to see that the 2 transfer gears setscrews are firm.? (Sometimes these have tiny setscrews and can't be loosened or tightened!)..See note above about? replacing them with better 4-40 setscrews that can more easily be removed or tightened.
6.? Now that every item has been checked for hysteresis with the worm locked into the ring gear, there will be too much friction. You can't leave it like this in the evaluation condition with the worm jammed.? ?The next step is to drive the "back off" bolt in to relieve the heavy spring pressure forcing the worm into the ring gear.? ?Use just enough bolt inward movement so the worm will turn by hand using the gearbox transfer gear.? (The DEC motor is still off!).?
7. Now put the motor back on and run the Gemini DEC axis in both directions.? There should be no Lags or Stalled messages.? If you get a stalled message try reducing the digital setting for max Slew rate motor speed.? If that won't solve it you must back the worm off more with the back off bolt, but be careful...the more you back that off the more time there will be when it switches direction.?
--------
I'm done....I hope this step by step investigation will turn up good suspects and close the mystery case for you.
Best, Michael?
On Tue, Apr 6, 2021, 4:13 PM Henk Aling < haling@...> wrote: On Sun, Apr 4, 2021 at 10:21 PM, alan137 wrote:
I also tried a configuration with a very soft spring and no back-off bolt for a truly spring-loaded worm.? With this set up it is very important to have a hard stop that prevents the worm from lifting up and skipping over the ring gear.? I found that twisting the axis one direction lifts the worm up, and the other direction pulls it down into the gears, so there must be some kind of asymmetric shape to the ring gear tooth profile.? Anyway, I could sometimes also get a very low backlash, but this system was too unreliable in terms of always getting a low backlash.? I think the low spring force makes the worm block "wobbly" and it doesn't seem to sit on the ring gear consistently.? Too bad.? And if there is too much axis friction, a turning worm gear will try to "ride up" on the ring gear, whereas in a system where the worm is locked in place, it would be forced to push the ring gear along.?
Does this mean that the back-off screw is meant to keep the worm from riding up the ring gear because if it does, the assembly wobbles?? That would be disappointing.
From my own experiments (DIY v3 see the other thread), I have no trouble pushing the worm into the ring gear without the motor binding up.? However since both bottom screws are loose (by slightly loosened screws), the worm assembly can tilt and it does - mainly because I have steppers attached with a flex coupler.? Because the motor and worm are slightly misaligned the left block wobbles.? I will attempt to fix that by applying a Belleville washer at the bottom of the bottom screw of the near block so it pulls the near block down, hopefully tight enough to level it (and thereby the whole worm assembly) against the bottom. If that works I still have to worry about the block staying in place horizontally (the spring pushes it to the far side of the hole for the near bottom block) so in the worst case I have to build a stop so the block can't travel within its hole.
I have never seen the Losmandy spring loaded worms in detail other than the assembly/adjustment video.? If it has a wobble problem, would it help to pull the block down at the pivot by applying a (Belleville) spring underneath it?? An old design by Chip on CN had springs at the bottom IIRC.? In the worst case add a brace around the worm assembly that provides a pivot at the top to keep it from wobbling.? Pivots on both sides should greatly reduce the wobble.
|
Re: G8 DEC axis backlash, fixing, questions
HI Jim
nice to see you here :)
can you upload a guidelog?to see how it's going?
one additional thing you might try is increasing the TVC. you might try a value of 10 and see how that goes. we used to steer people away from TVC because it could potentially fight with PHD, but now we are more thinking it can help take the edge off the backlash in certain circumstances
I need to read through this entire post but for me a lot of my GM811G DEC backlash was caused by one of the screw holes on my DEC gear box being stripped.? The motor would rock back and forth when changing DEC directions.??I purchased another gear box, disassembled the DEC worm assembly, adjusted stuff and put it back together. Worm mesh is critical and you need to play around with it.
Its manageable now but I have to keep on top of things.? I still have more DEC backlash than I would like.? The PHD2 dither 'settle time' is better.? Before I would get dither settle time errors and my error value was set to 16 seconds - if I recall correctly.
------------------------
Jim W
Phoenix, AZ. USA
-- Brian?
Brian Valente portfolio
|
Re: G8 DEC axis backlash, fixing, questions
I need to read through this entire post but for me a lot of my GM811G DEC backlash was caused by one of the screw holes on my DEC gear box being stripped.? The motor would rock back and forth when changing DEC directions.??I purchased another gear box, disassembled the DEC worm assembly, adjusted stuff and put it back together. Worm mesh is critical and you need to play around with it.
Its manageable now but I have to keep on top of things.? I still have more DEC backlash than I would like.? The PHD2 dither 'settle time' is better.? Before I would get dither settle time errors and my error value was set to 16 seconds - if I recall correctly.
------------------------
Jim W
Phoenix, AZ. USA
|
Re: Nice discussions leading toward perfection...
Mike et al, This is a very interesting discussion and I am following?every comment.?But as a noob I often have more questions than answers! The one?question thats nags me thru these guiding and technical discussions is "when does one need to be this technically?sophisticated with their guiding?" For instance I have a GM-8 and a Stellarvue 102mm and generally do not guide since I?fail in so many ways to get guiding to work for me.I?typical shoot 30sec to 3 min exposures depending on?targets and "seeing" from Atmospheric. I like my images and am learning both the?nuances of imaging but the workflow and challenges of Pixinsight. So at what focal length, exposure time etc does guiding become?truly?necessary??Thanks. Tom
--- mherman346@... wrote: From: "Michael Herman" <mherman346@...> To: [email protected]Subject: Re: [Losmandy_users_io] Nice discussions leading toward perfection... Date: Mon, 5 Apr 2021 02:59:52 -0700 This is a nice forum (especially for us insomniacs) to discuss improvement and diagnostic ideas.? I'm sure all the other owners are busy using their mounts to get great images.? That's the "silent majority"!??
I had great results from my G11T the other night when shooting M81 and M51.? Of course those are less of a challenge for tracking since they are close to the NCP.? Still, I was shooting at FL of ~2850mm.? I felt that I really didn't need auto guiding, though I was using an OAG in case my polar alignment was not perfect.? My 300 sec exposures were perfect...stars were all round dots.??
That G11T has been souped up.? I put on a 1:2 ratio belt drive, and McLennan 25:1 gearbox, so it is operating as a? 50:1 gearbox providing double the torque.? And any variation in the gearbox or belt is cut in half too.???I shimmed the RA worm case by very thin brass shims, to try and get the center of the worm exactly centered on the ring gear center.? I also just installed a?pair of back to back Belleville washers (30mm OD for the large worm bearings of the Titan) at the screwcap end of the RA worm, do the worm is not going to shift during temperature changes.? I reduced the weight hanging to 1 pound on the East side RA cord.? And I changed from Superlube to CRC caliper grease with MolyD, graphite, and teflon on the RA worm and ring gear.??
The system is running very smoothly.? The load is a 14 inch C14EdgeHD (~62 lbs) the 0.7 FL reducer (~5 lbs) QHY168C camera (~4 lbs), nautilus filter wheel, Celestron big OAG, and a heavy 80 mm refractor on rails on the top.? And 82 lbs of counterweights.? The total motor current running sidereal was ~ 0.25 amps at 17 volts into a high torque motor.? It's driving a total weight on the RA of about 165 lbs...well balanced but for the 1 pound hanging weight.?
Incredibly, I could still see M81 center in the camera image through opaque sky clouds.? It was enough to test the mount and get the camera imaging working.? Here is a picture of the rig and sky conditions before I shut it down.? Anyone sensible would have packed up earlier.??
As soon as you get your system working perfectly, it acts like a magnet for clouds.? Better than iodide cloud seeding!??
All the best, Michael
My opinion is that you need just enough clearance to allow the bearing to tilt inside the block by a tiny bit.? Maybe about 0.001 - 0.002" clearance
|
Jamie Amendolagine
Michael Herman