Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
G11G First Night - Not So Good
|
Here are a couple of shots using PEMPro Viewer to show last nights initial run with a new G11G. The first one was good in DEC and not bad in RA. Then things got bad with the RA by the end of the night. Huge deviations at exactly 1/2 worm period. Anyone else seen this before. I tried learning and applying PEC a few times but that did not help. Each PEC learning curve looked completely different.
This morning I redid all of worm adjustments as per the video on YouTube. Discovered you must remove the motors so you can feel the drag when adjusting the last screw that pulls the worm back from the gear. Also if Losmandy is reading these please mill some slots down the sides of the the motor heat sinks to get better access to the mounting screws. Guiding is PHD2 with a ZWO ASI290mm on their 60mm guide scope. This is solidly mounted on top of a C8 and has usually produced? .9 to 1.5rms with a CGEM mount. Rick |
|
Hi Rick I'm a little confused about what you have here - are you measuring periodic error correction while guiding? if you are not doing PE correction, then the PEMPro log viewer is not a good way to view your results. use the PHD Log viewer instead.? i don't know how you are going about this, but there's a number of things you mention that are not correct - you do not need to remove the motors to adjust the worms, and adjusting the worms is probably not necessary for what you are trying to do right now anyways (the following only applies if you are doing PE correction) - PEC correction is only for RA, it has nothing to do with DEC - your initial PEMPro PE run (uncorrected) looks on the good side of pretty typical:?+/- 4" (the mount is spec'd at?+/-5") - the correction run (second run?) looks bad. What fitting algorithm did you use? Again, if I misunderstood what you are doing, my apologies, and use PHD log viewer instead On Sat, May 16, 2020 at 9:24 AM <rboudah@...> wrote: Here are a couple of shots using PEMPro Viewer to show last nights initial run with a new G11G. The first one was good in DEC and not bad in RA. Then tings got bad with the RA by the end of the night. Huge deviations at exactly 1/2 worm period. Anyone else seen this before. --
Brian? Brian Valente portfolio |
|
Here is the same data with the PHD2 Viewer.
I found there is a lot of drag on the spur gears from the motors and you can not feel the point at which the worm releases and yet still no backlash. The video shows what to look for and does not have motors attached. My initial results looks OK so I decided to give PEC a try but was getting no improvement. Then things started to get much worse as shown in PHD2 Final.jpg. As you can see the DEC is fine but RA got about 5 times worse with the peak at 1/2 worm frequency.? These were looking at 2 different areas of the sky the initial file was of M13 which was in the eastern sky the final file was the trifid nebula in the southern sky. Could there be a worm gear issue. All of the PHD2 settings were similar on these 2 attachments. I tried other control schemes but none worked any better. |
|
Hi Rick if you'd like some constructive feedback, i'd really need to see the actual guidelogs themselves it should show a good calibration, a guiding assistant run, and at least 15 mins of uninterrupted guiding regarding PEC, i'm not sure how you went about it? For me, it usually takes quite a while using PEMPro to gather the correct data with enough worm cycles, select the best fit algorithm, and rerun the results to see if the curve needs to be inverted.? If you're not doing something roughly along these lines with PEMPro, i'm not surprised PEC doesn't show any benefit, and would probably make it worse On Sat, May 16, 2020 at 11:22 AM <rboudah@...> wrote: Here is the same data with the PHD2 Viewer. --
Brian? Brian Valente portfolio |
|
>>>
For PEC I used the functions in the ASCOM driver. Got it - yeah i thought you might have Built-in PEC is really only for visual. I don't recommend it for imaging On Sat, May 16, 2020 at 12:43 PM <rboudah@...> wrote: Here is the log. --
Brian? Brian Valente portfolio |
|
Rick i've had a chance to go through this a bit and i have a few thoughts: - when you calibrate, make sure you bump the mount north first. You may be doing that now, but just want to confirm - you could improve your focus on your guidescope, the HFD is in the 4-6 range, I think it could be improved. (it appears the focus gets worse through the night) - aside from those points, calibration looks pretty good on the guiding itself,? Regarding DEC, it seems to be behaving reasonably well. I suggest seeding the backlash adjustment at about 800ms? with auto adjust enabled, and see how that adjusts itself. right now your?backlash is around 280ms, but it still appears to take a few more pulses to reverse. the auto adjust period will figure out where it needs to be (and adjusts this over the night as needed), but it could probably do it faster with that setting.? Regarding RA, as you pointed out the 1/2 periodic error (2x the fundamental, about 120s) is extremely high - about 3x what I typically see on this mount. you can see it dwarfs even the primary PE. As a result it's taking multiple guide pulses to correct, and then it appears to overcorrect. so you are seesawing back and forth on RA. smoothing that out is what needs to happen
You mentioned making adjustments to the worm itself, and also apparently you took the motor off when doing this? a couple things could be causing this: one is if you have a bad PEC loaded. the method you used would almost certainly cause this, so please make sure that PEC curve is completely cleared from your gemini the other possibility is that the adjustments you made may be exacerbating this periodic error. I'd suggest going back to square one on adjusting the spring loaded worm, and follow scott's video exactly as shown below including "from the blocks" first part. You want to do this with the motors connected but without your scope mounted: ?? ? On Sat, May 16, 2020 at 12:43 PM <rboudah@...> wrote: Here is the log. --
Brian? Brian Valente portfolio |
|
Rick FYI this is a graph of typical G11 pulled from the shelf and tested, you can see the 2x fundamental is hardly there? (the 1x PE is a bit high, so it went back on the bench for examination) On Sat, May 16, 2020 at 12:43 PM <rboudah@...> wrote: Here is the log. --
Brian? Brian Valente portfolio |
|
The results I sent were from the factory tuning. I just went in to double check this morning after last night results.
It is possible I could have messed things up with the PEC that I did about halfway though.? How are you testing the unit in your example.? How are you getting that data. Do you have these test results for all units leaving your shop. I will try your DEC suggestions. Do you have a document describing some PHD2 setting that should work best with a new G11G. Also how to implement a PEC correction. Thanks, Rick |
|
Hi Rick >>>How are you testing the unit in your example.? How are you getting that data. >>>Do you have these test results for all units leaving your shop. no, we spot test, but they are largely very consistent.? >>>I will try your DEC suggestions. great! >>>Do you have a document describing some PHD2 setting that should work best with a new G11G. Also how to implement a PEC correction. the baseline guiding is a document i put together for exactly this reason - it's about as far as anyone can get before you get into tuning things on an individual?basis. We use that over at PHD forums all the time, FYI regarding PEC, Ray Graylak (author of PEMPro) has good instructions, but i'm happy to help people through the process. it's one of the things that we've been discussing as a tutorial or zoom meeting On Sat, May 16, 2020 at 8:21 PM <rboudah@...> wrote: The results I sent were from the factory tuning. I just went in to double check this morning after last night results. --
Brian? Brian Valente portfolio |
|
Regarding the mounting screws you can file the side of the motors to make a flat face and have some clearance for the hex bolt heads.
There's a couple of things you can look at for your PE, frequency spikes and tracking noise. I'll assume that your mount is new enough that the RA disc is square with the RA shaft and when your clutch is engaged the pressure is evenly spread over the surface. The ideal vertical worm meshing will depend on how high/low is your spur gear. So with your clutch locked (not over tightened that you "smash" your thrust bearings) push the worm block in the gear to mate it and see if there's a thin gap under it. If there is shim it with some alu shim (coke can or other if thinner). Once you have that sorted out engaged it completely in the spur gear then release the pressure as you encounter "hard spots". If you have a smooth motion during your PE run then you're good. If it's noisy randomly you might have to thick a grease in your thrust bearings or the worm/gear grease is sticky too. So check that as well. If your PE is smooth but still too big check your couplings and the centering of the motor pinion in your gearbox. With the single worm block it is very hard to misalign the worm bearings anymore but try to keep the couplers inline as much as possible. That should turn by hand really easy with no pressure at all. If your G11 sounds like a coffee grinder then recenter the motor pinion in the gearbox until it stops then secure the motor. When everything sounds good and quieter run your PE again. |
|
Re:
"With the single worm block it is very hard to misalign the worm bearings anymore but try to keep the couplers inline as much as possible." I agree it's hard to grossly misalign with this design, but the newer OPW in not a one-piece-worm and is in effect the old 2-block design with the cover bolted on.? With the OPW removed, if I loosen the bearing block 'front' screws to just barely snug, the worm rotates very smoothly by hand.? As soon I snug those up to a reasonable/usable level, the worm definitely binds more as I rotate it, implying that something is now misaligned (e.g., due to machining accuracy of bearing slots, bearing blocks outer surface, bearing block mounting channel) .? I'm not saying that it necessarily degrades PE (haven't measured directly).? The newer OPW does make meshing easier and more precise, but aligning the Oldham coupler is just guess-work compared to the 2-block design where you can see it in situ. Keith |
Greg
to navigate to use esc to dismiss