Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
On Fri, Jul 23, 2021 at 01:27 PM, Brian Valente wrote:
Can you post your guidelogs for that night? that would be?great, thanks
Brian, I'm not quite ready to release actual guide logs at this point. I'm still studying problems and making changes etc. I did post the FFT analysis for the attached session from two nights back, a few post back. You may wish to take a look. It revealed a large 2600s error in Ra which I attributed to the thrust and or needle bearings or some combination. The take away is I've reduced the native 240s error considerably to the point it guides out almost entirely (mostly through changing the coupler and my own spring loaded design). Also, the 80s error from the bearings is substantially reduced. You can see from this image (attached) that the Dec axis is still playing up a bit. I will be going back to the 25:1 ratio gearbox tonight and using 1024 optical encoder. I fear I damaged my 80:1 gearbox due to a misalignment problem I found and written about with regard to the worm and ring gear and the force needed to rotate the worm. This alignment problem was also picked up by Michael Herman who looked? at one of his mounts and found a similar problem. Its stops a true precise mesh from being obtained. I set the initial backlash prior to guiding at 200ms, and in this case it adaptively brought it down to 43ms. Last night is was 40ms, so a good confirmation. But from the Dec graph something old is going on. Once this is sorted I'll release log files. Peter
|
Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
Peter,
I lost track of the Dec axis configuration that resulted in the reported 40 msec backlash. Are these the most relevant configuration items:
-no spring load
-homemade one piece worm block
-added a 3rd needle bearing on Dec axis
-replaced the upper thrust bearing with thicker one
Eric
|
Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
Hi Peter
Can you post your guidelogs for that night? that would be?great, thanks
New bits arrived today. Will change out the Ra thrust bearings a see if reduces the 2600s error shown in the previous posting. These thrust bearings are German sourced from McMaster!
Also, received the 1024 optical encoders from US Digital. I hate to tear down what is already a superbly performing system, but this is all in the interest of science and you guys. Will put back the 25: 1 ratio gearbox in the Dec axis and install the 1024 step encoder. Lets see if we can achieve same high level of guiding precision while preserving slew rates.
Oh my backlash was 40ms last night and Predictive PEC absolutely nailed the 240s error (see attachment) by totally tuning into it (please note earlier post how I reduced the 80S error), and still no PEC.? The attached image is a 400s Ha single frame of the Cave Nebula, 2.4 um/px. Seeing was horrible and transparency was dire due to lingering smoke in the atmosphere - full moon.
-- Brian?
Brian Valente portfolio
|
Re: Needle bearings and axial play, is this a significant performance problem and how difficult is it to fix?
New bits arrived today. Will change out the Ra thrust bearings a see if reduces the 2600s error shown in the previous posting. These thrust bearings are German sourced from McMaster!
Also, received the 1024 optical encoders from US Digital. I hate to tear down what is already a superbly performing system, but this is all in the interest of science and you guys. Will put back the 25: 1 ratio gearbox in the Dec axis and install the 1024 step encoder. Lets see if we can achieve same high level of guiding precision while preserving slew rates.
Oh my backlash was 40ms last night and Predictive PEC absolutely nailed the 240s error (see attachment) by totally tuning into it (please note earlier post how I reduced the 80S error), and still no PEC.? The attached image is a 400s Ha single frame of the Cave Nebula, 2.4 um/px. Seeing was horrible and transparency was dire due to lingering smoke in the atmosphere - full moon.
|
Re: Need advice selling my Losmandy G11 492 Digital Drive
1. If you choose send money to a friend there are no fees.? You have to really not trust the person you are buying from to force them to pay a fee, which they then charge you.?
2. If you ship in a Pelican case. Zip tie it and slap a label on it. I'd look carefully at insurance, it can add a lot to shipping and may not be worth it, read the small print...? If shipping to a neighboring state or two, the post office will be the least expensive option, else UPS or Fedex.? Buy the label online not at a store. Don't over package, like thousands of feet of taped bubble wrap, with every nut and screw equally wrapped:((??
I emailed Jeff this morning and he is not interested in splitting the mount sale between two people, neither is he even reading this thread he started. Seriously..........................................
Peter
|
Re: Meridian Flips and NINA
On Fri, Jul 23, 2021 at 08:57 AM, Edward Plumer wrote:
Fascinating!? From this video, I am also hard pressed to see how this is not an issue in the Gemini firmware itself since the discontinuity clearly occurs in the HC. When I get back home to my scope, I will try to reproduce as well.
(1)? I lost track of your configurations ... did you see this on both the released and beta version of Gemini firmware??
(2)? To echo an earlier question, can you confirm this only occurs during a slew operation, not during normal sidereal tracking?
(3)? Are you only seeing this around the Meridian, which John relayed is likely to have trickier math?
(4)? If so, do you have any sense of how close you have to be to the meridian during a slew to see the anomaly?
(5)? I gather from your last comment that this does not happen everywhere along the meridian?
As you continue to document this, it might be good to note your lat/lon, time-of-day, rough RA/DEC, etc before the slews to see if there is any pattern that might be useful to someone who knows the code. Also allow others to reproduce the issue by resetting their clock and location in the HC to match your documented occurances.
In the meantime, as a work-around in your automation it might be good to set BOTH "Pause Before Meridian" and "Minutes after Meridian" in NINA to a minute or two, based on what you report for question #4 above. Leave "Max Minutes after Meridian" to the larger value based on WSL. Use the spreadsheet to get consistent values for the other parameters. This should ensure that NINA is not pushing the slew buttons or grabbing bogus coordinates during the window around the meridian.?
I would like to hear whether those settings sufficiently "hide" the occurrence of the anomaly.
By the way, I uploaded a newer version of the spreadsheet with some added round-off error checks.
Edward,? This issue I do think does lurks at the route of other issues. I experienced it for my self about 6 months ago during galaxy season, but attributed to a configuration problem on my end. I learned to work around the problem buy imaging up to an hour past the meridian.? ?I reported numerous times that I could not get the meridian flip to work correctly. I now realize that the flip did take place, although thank you for your spread sheet. The scope ended up staying on the west side pointing at the ground. I now know the reason for this which I will explain. I've set a lot of info over to Rene, but this is what I think is happening:
?
1. Gemini responds correctly to the coordinates provided by the planetary software to the area of the sky where the bullseye vanishes. This is close to the meridian, but the invisible zone is cone shaped with the tip at the poll star.?
?
2. When the bullseye disappears, the scope still responds correctly to co-ordinates given in that general area, even though the bullseye has now appeared in the southern hemisphere. That appearance in the southern hemisphere of the bullseyes and its associated coordinates is why the scope ends up pointing at the ground after a flip attempt for me.?
?
I think there must be a translation problem from the internal co-ordinate system to the co-ordinates sent back to the planetary software.? ?I'm sure Rene will get to the bottom of this.?
Peter
?
|
Re: Meridian Flips and NINA
It is in the "Files" section, the direct link is?
/g/Losmandy_users/files/Edward-Gemini-NINA.xlsx
--
Edward
|
Re: Meridian Flips and NINA
"?I uploaded a newer version of the spreadsheet with some added round-off error checks."
Thanks Edward.
------------------------
Jim W
Phoenix, AZ. USA
Losmandy GM811G, NINA 1.11, ASI2600MC Pro, Sky-Watcher Scopes
|
Re: Meridian Flips and NINA
>>
By the way, I uploaded a newer version of the spreadsheet with some added round-off error checks.
thanks edward. can you post the link again?
On Fri, Jul 23, 2021 at 8:57 AM Edward Plumer < eplumer@...> wrote: Fascinating!? From this video, I am also hard pressed to see how this is not an issue in the Gemini firmware itself since the discontinuity clearly occurs in the HC. When I get back home to my scope, I will try to reproduce as well.
(1)? I lost track of your configurations ... did you see this on both the released and beta version of Gemini firmware??
(2)? To echo an earlier question, can you confirm this only occurs during a slew operation, not during normal sidereal tracking?
(3)? Are you only seeing this around the Meridian, which John relayed is likely to have trickier math?
(4)? If so, do you have any sense of how close you have to be to the meridian during a slew to see the anomaly?
(5)? I gather from your last comment that this does not happen everywhere along the meridian?
As you continue to document this, it might be good to note your lat/lon, time-of-day, rough RA/DEC, etc before the slews to see if there is any pattern that might be useful to someone who knows the code. Also allow others to reproduce the issue by resetting their clock and location in the HC to match your documented occurances.
In the meantime, as a work-around in your automation it might be good to set BOTH "Pause Before Meridian" and "Minutes after Meridian" in NINA to a minute or two, based on what you report for question #4 above. Leave "Max Minutes after Meridian" to the larger value based on WSL. Use the spreadsheet to get consistent values for the other parameters. This should ensure that NINA is not pushing the slew buttons or grabbing bogus coordinates during the window around the meridian.?
I would like to hear whether those settings sufficiently "hide" the occurrence of the anomaly.
By the way, I uploaded a newer version of the spreadsheet with some added round-off error checks.
--
Edward
-- Brian?
Brian Valente portfolio
|
Re: Meridian Flips and NINA
By the way, the spreadsheet password is "nina". Not trying to be coy, just ensuring users don't accidently changes cells that are not meant to be entered.
--
Edward
|
Re: Meridian Flips and NINA
Fascinating!? From this video, I am also hard pressed to see how this is not an issue in the Gemini firmware itself since the discontinuity clearly occurs in the HC. When I get back home to my scope, I will try to reproduce as well.
(1)? I lost track of your configurations ... did you see this on both the released and beta version of Gemini firmware??
(2)? To echo an earlier question, can you confirm this only occurs during a slew operation, not during normal sidereal tracking?
(3)? Are you only seeing this around the Meridian, which John relayed is likely to have trickier math?
(4)? If so, do you have any sense of how close you have to be to the meridian during a slew to see the anomaly?
(5)? I gather from your last comment that this does not happen everywhere along the meridian?
As you continue to document this, it might be good to note your lat/lon, time-of-day, rough RA/DEC, etc before the slews to see if there is any pattern that might be useful to someone who knows the code. Also allow others to reproduce the issue by resetting their clock and location in the HC to match your documented occurances.
In the meantime, as a work-around in your automation it might be good to set BOTH "Pause Before Meridian" and "Minutes after Meridian" in NINA to a minute or two, based on what you report for question #4 above. Leave "Max Minutes after Meridian" to the larger value based on WSL. Use the spreadsheet to get consistent values for the other parameters. This should ensure that NINA is not pushing the slew buttons or grabbing bogus coordinates during the window around the meridian.?
I would like to hear whether those settings sufficiently "hide" the occurrence of the anomaly.
By the way, I uploaded a newer version of the spreadsheet with some added round-off error checks.
--
Edward
|
Re: Need advice selling my Losmandy G11 492 Digital Drive
I'd rephrase a bit. "The seller must demand that the buyer cover the fees." Generally, on CN, that only happens when the item is selling for several hundred bucks.?
Essentially all transactions today involve a 3rd party getting a piece to aid in the money changing?hands. Go to the farmers market and most vendors?accept credit or debit cards. Otherwise? sales will go elsewhere or not happen at all. I've sold about 550 OnStep controller kits worldwide in the last couple years, all through?Paypal. With sales that average $50 the fees are more in the 5-6% range and often there are additional?costs for currency conversions when you go outside the EU.?
It's just?a cost of business. Without it I wouldn't be selling to Abu Dhabi. It just gets built into the price. Don't get me started about shipping costs and reliability.?
Just shipped 4 kits to a fellow in Italy. Talk about triumph of hope over experience. His first kit landed in Warsaw a couple days out of JFK. It was then flown to Milan where it sat for 4 months. Considering it lost, I shipped a second. The two packages arrived a day apart.?
This happens a lot in the EU, like maybe 5% of the time. Every member state has their own postal service and customs service. When I order components from China they clear U.S. Customs before they are exported. Individual EU states can't offer that kind of coverage. Anyone who knows the history of the U.S. knows how well things worked under the Articles of Federation. The country only works because it became clear to the states that they had to give up some of the prerogatives of statehood. Unless and until the EU? members accept that fact they will not have a functioning union.
|
Brian and Keith,
Thanks for clarifying the situation with these upgrade options.? I will likely wait on the SLW idea.? But I might do the RA extension.? My setup is not permanent and it might be handy to split out the DEC portion.
Steve
|
Re: Need advice selling my Losmandy G11 492 Digital Drive
as @pcborland is interested in the mount head?
I am interested in the tripod base as I just bought a G11 locally but am not happy with the lighter weight tripod that it was sold with.
I think perhaps we can figure this out??
There are ways to ship this and if you keep the weight down below 50# per box it is less expensive.?
Flat rate boxes are cheap for very heavy items like weights. Perhaps @pcborland would mull over also buying the weights because they are 32mm bore?
I can be reached on Cloudy Nights where you can message me directly or email at @mikeby.mikeby and the last half is at? gmail dot com
?
|
Re: Meridian Flips and NINA
On Thu, Jul 22, 2021 at 06:50 PM, <pcboreland@...> wrote:
I would thing there has got to be a problem in the math.
Peter, As I recall with my exchanges with Rene in 2018 when he wrote the Beta ware, he said there was some tricky calculations which occur near the meridian. Several of the issues some of us have seen over the past few years have that common theme, like the mount not responding to an app slew command to an object east or west unless you were already on that side of the meridian. Or the scope pointing into the ground. Or the wild slews. The Beta ware fixed many of these quirks for me, but perhaps the size of the fix needs to be enlarged, whatever that means in software speak. The Gemini logs I forwarded at that time helped with the solution. If you know how to read the code lines, you might see further into what happens at the noted points in recorded time when you see a problem.?
|
Re: Need advice selling my Losmandy G11 492 Digital Drive
Just did a PayPal sale recently. The buyer must add 3% to their transaction amount for the seller to get full price requested.?
|
Re: Need advice selling my Losmandy G11 492 Digital Drive
Bob H. here, if you consider "friends" in Pay Pal you lose a large part of the protections that PayPal gives to buyer and seller, especially the seller if something goes wrong.? Additionally there are 2 types of shipping costs, 1 covers just the transportation and the 2nd option gives a greater degree of protection of the shipped goods, so clarifying the difference to a buyer will protect both parties if problems arise. Lastly Cloudy nights is a viable option and there are methods to checking background of buyer mostly this will give a greater degree of safety to a seller. well that's my thoughts and data base on selling your mount. Respectfully submitted.
|
Re: Meridian Flips and NINA
Hay everyone,
I've uploaded another video that shows the problem some 7 hours later in the day, for those here that might have insight as to what might be going on here. Since I first showed the problem this morning, I have since reset Gemini II and re-entered all my data.?
Its a fascinating problem as it moves along the meridian so hard to nail. I would thing there has got to be a problem in the math.
Peter
?
|
Hi Steve.? I was/am in the same boat, and for me it doesn't make much financial sense to upgrade vs. selling what I have and purchasing new.? You might be interested in this thread:? /g/Losmandy_users/message/67900. Keith
|
I'm looking to switch from Celestron to Losmandy!
Any idea when the G11G will become available??
I signed up for the waiting list.
Thanks!
|
Edward Plumer