开云体育

Henk,

AFAIK very few solids expand when cooled and certainly none of the materials used in the Losmandy mounts do this.?

The issue is the different rates of contraction between metals that is causing the binding issues in the drivetrain when backlash is set in a warm room. When chilled everything shrinks at different rates causing the backlash to shrink to zero and the lag to occur.?

Setting backlash on a cold mount will eliminate this issue. Of course the mount will exhibit excessive backlash when warm and last time I checked I mostly observe and image at night in the dark when it is a lot cooler.??

--

Chip Louie - Chief Daydreamer Imagination Hardware

Michael,

Just like Covid-19 the problem did not go away magically, something has changed in his system, we just don't know what. Chances are good it will return just as Covid-19 has,? unless the issue was just dirty contacts.?

--

Chip Louie - Chief Daydreamer Imagination Hardware

George,

The grinding was probably the cheap or damaged bearings they use in the Chinese mounts. I had a new Sirius HEQ5-PRO that had been assembled with the RA tapered bearings so tight that the cheap Chinese bearing seats were permanently damaged. I had to use a strap wrench to loosen the retainer. I replaced all of the larger tapered rollers on RA and DEC with SKFs and the mount was fantastically smooth. I had used Bellevue discs and precision ABEC7 bearings for the worms and Timken deep groove bearings elsewhere for the worm wheels. Huge improvements in movement and PE. The belt drive was a big help to, I loved doing the EQMOD stuff but the Atlas had no chance to carry the payloads I wanted for imaging so before I went down the road I did with my 2x Sirius mounts I dumped them all and bought a used G11.??

--

Chip Louie - Chief Daydreamer Imagination Hardware

Hi Thomas

I use shims between the outer gear block and the frame on my G11S.? The general procedure is to push in 2 directions, tangential towards the motor to remove the lateral backlash, and radial towards the crown wheel.? The latter is the tricky part because the crown wheel is never perfectly round and the motors can bind up.? What I do to sense if the radial pressure is too much is to disconnect the stepper and manually turn the worm.? I had to do it so often that I don't use the cover plates anymore. It is best to leave some tolerance so the motor does not bind up elsewhere.? The shims work but not consistently, I suspect that heat expansion may have something to do with it.? Also pay attention to how much you tighten the clutches because it interacts.? The spring loaded worms will probably fix this, otherwise use friction and/or weight imbalance to preload the worm gear.

Chip Louie,

I agree with what you said.....And, believe me, I actually did all of those things before I broke out into the other controller box!?

Remember, I have been at this fifteen or twenty years or something. The battery inside, for instance, was at 3.06. The connector (the arm that presses against the battery) was a little weak, but I re-bent it and it was tighter. The cables were swapped back and forth, reversed, and cleaned......Removed the motors from the worm gears to find that the motors themselves still were really stalled (it was not a mechanical thing in the worm gears). All that stuff did not cure it. When it did not, I tried the other controller. But, as you said, do all the other things first.?

Apparently posting here did cure it. I was just out in the desert for two nights, and it worked fine, with no stalling.?

Go figure.?

We will see how it goes next time. I hate intermittent failures.?

Alex
?

Alex,

I will say I hate to be an a--hole but we see these issues many times a month.?Stop swapping controllers and hand controllers.?

First verify battery voltage is 3V minimum to eliminate any issues here, if it is not you cannot know with authority that the battery is not part of the problem buy several to have on hand for when the Gemini sees to be strange but you can't figure it out, 99% of the time it is the battery and or a factory reset resolves these issues due to data corruption or incorrectly entered time/data/GPS coordinates, wrong format etc. once you have eliminated the other potential issues.

Stick with the Gemini that is "bad" and do all your testing with known parts, sounds stupid but mark them so you can ID them reliably. Next eliminate cables contacts and sockets, they are cheap and unreliable, get some contact cleaner and spray the Gemini contacts and push the DIN cables in and out several times to clean the contacts. Do the same on the servo side, clean, blow dry retest. If the SAME issue is present then while the error is visible gently jiggly the cables on the Gemini and see if the behavior stops or changes. If it does swap cables and test again, does the issue stay on the same axis or does it move with the cable? If it stays on the same axis even with different cables swap the RA and DEC servo motors. Reboot and retest, does the problem stay with the same servo motor when swapping cables? If yes you need a servo motor. If no then you have a bad Gemini 1 controller, send it to get repaired.

The idea is to isolate the problem and by rotating in other parts of an unknown condition you are making the debugging issues much bigger and more mysterious than necessary. Never swap parts until you can identity a potential bad part with the least number of parts, keep the data set as small as possible. This is basic trouble shooting, there is only one Gemini 1 controller, 1 hand controller, 3 cables and 2 servos. If this seems confusing write down the position of the parts tested and make jiggling a part of each testing step to try to see if there are multiple failure modes here. Only then once you have determined a repeatable failure do you swap in the other [arts to see if it resolves the issues and because we don't know if the other system is 100% good you may still have difficulty isolating the bad component(s).? ?

--

Chip Louie - Chief Daydreamer Imagination Hardware

Thomas B.,

If you are imaging be certain to do all of this in the COLD under similar conditions as you expect to use the mount for imaging. The mount has several different metals with very different material expansion values so if it seems perfect at normal warm room temperature and your counterweight shaft has minimal play it has a very good change of lagging in the cold after a few hours to come to ambient equalization. For visual use just be sure to not get it too close the shrinkage still happens but it is less important because your eyeballs won't care if there is a slight lag or sticking, it is just annoying. During the daytime the mount will seem to have a lot of backlash if adjusted in the cold, this is normal and desirable for a low PE in the cold with the old 2-piece worms. If you plan to keep this mount as a 2-piece worm you can look for the Losmandy G11 tuning guide by Michael Herman, it can get you to a very good place. Might be in the files, look around!??
??
--

Chip Louie - Chief Daydreamer Imagination Hardware

Keith N beat me to it.? That's basically what I've done in the past.? The trial run of your mount is the key.? A trial run will reveal any tight spots in your gear train which would cause a stall while you're using your mount.? Better to find that in your garage during daylight, than in the field under the stars.

Alex?

when you day “immediate”.....how immediate? ?Read on!

?...you can get stall messages on boot straight away as it is booting or straight at boot completion.

?
During boot, drive signals are applied to the Ra and DEC motors locking them. ?If they move at this stage you get motor warnings.

When boot is completed you see a HC MSG....”xxxx boot at xxx time....”, the HC then flashes sidereal rate (or the rate that’s been set) and starts moving at that rate. ?If encoder movement signals are not received or are slow it gives a stall or lag message.?

So try the other controller or swap over a motor ? Usually it’s a motor issue they are getting a little older seen them failing more and more after 15+ years ? Cannot complain though ?

--
Brendan

Thanks, Everybody......

I shall work through it as you suggested. I should let you know that i have been Losmandy Gemini user for fifteen or twenty years, and have been through motor stalls and all that many times, and worked through them. I always slow down the slew speed (usually to 600--have not tried 500), and make sure I have a full 13.8 power, swap cables and even motors......I have lots of cables and have swapped around. So, the basics I have been through.

I have a G11 in the little observatory. The GM-8 is the portable rig. I know to reconfigure on the cold start when moving the box from one to the other.

The one thing about this particular problem is the consistency of the problem and the solution. The original (problem) box stalls on startup (immediately) Every first of the evening. I swap in the other box, and it runs fine. I go back to CWD with the "other" box, install the original (problem) box? and start up and it runs fine. Has happened four times precisely like that.?

>>>>>>>>Q1: How long does it take for the stall to occur ? Right away at startup? or after a little while??

Immediate

>>>>>>>Q2: When you take the original GM8's Gemini-1 controller and put it on the "Other" mount (cold), does it have motor stalls also ? (even if it has different motors, etc)?

Have not tried it.

Alex

George,

You have it mostly right, it is fine for the worm to ride on the inner bearing surface. The purpose of the threaded cap is to take up the end play but this design cannot compensate for material expansion/contraction as temperature changes. So just as with the HEQ5-Pro/Sirius and EQ6/Atlas mounts you can use a Bellevue disc spring to allow for dimensional changes with air temperature changes. The worm is riding on the inner bearing so we will need a Bellevue disc with an outer diameter to match the outer bearing shell, just a bit smaller actually and the cone can be seated on the threaded plug which is tightened just enough to take up half the spring height no more. This will force the bearings on both sides of the worm to rotate at the same speed as the worm and take out any end play while also allowing some expansion/contraction for thermal changes. This works great but you must find the correct size Bellevue disc spring and ensure that the sprung bearing can move in the bore under the threaded cap/adjuster.?

--

Chip Louie - Chief Daydreamer Imagination Hardware

Alex?

First make sure you take note of what part belongs where (Mark them) prior to moving any cable or motor etc as moving may move the fault making it harder to locate or identify.?

On the mount that has the stall. ?Swap the Ra and DEC motors over prior to use. Does the stall fault problem follow the motor?

Yes, as said lower your max slew and goto rates down to 500x this helps but probably not “an on boot stall” ....unless it’s stalling in a slew or goto! ? Especially as you probably have the older motors ?I’m not saying it is a motor but by now I’ve seen them starting to fail, they are worked hard during their life, ?be prepared.....just sayin!

Also, as said, by moving the cables you are possibly hiding the fault. Try swapping over the cables from the “other Gemini” prior to boot. Does this fix the stall? ?Not at the same time moving the motors!

Be methodical is swapping parts over, keep track of swaps, mark the parts, try swap only one at a time. ?

See how this goes see if you can narrow it down. ? It’s either the 1st Gemini, it’s motor or its cable!

cheers
--
Brendan

Hi Dietmar,
No. I don't have the greatest luck with Meridian flips.
Even though all indications say I'm at the same spot, I have historically had a difficult time finding my object again.
I'm on a new avenue with my imaging now. As soon as I realized the value for me in Scott's RAEXT device to me for my transporting my mount parts out and back in, I further modularized my mount. Now I can easily transport the 4 main sections.
But recently, I've been doing imaging without Meridian flips at all, per Scott's advice with the RAEXT.
Instead of fooling with a meridian flip, I simply find my object, and begin imaging it. Then let the mount, PHD2, and my camera work. I usually will stop the imaging at certain points and save the image, then continue on.
I'm using an Atik Infinity OSC (one shot color) camera, and it stacks the image in it's program. I save my images in a TIFF format, which is why SGP was not for me. Being limited to FITS files ONLY was a major drawback for me. My imaging is 100% electronically used. Web posted, or email transported. FITS files are just a major bog to me.
I would have bought into SGP if not for the FITS only file format and SGP's creator being an ass about not opening SGP to at least one more modern file format. His loss.
I've somewhat lost interest in N.I.N.A. even. Because I can choose a victim, and suck the light out of it and build beautiful images by just using my GM811G with the RAEXT to image over my entire view of the night sky in most instances.
My East to West, Southern facing view is my most object free. So finding pre-meridian subjects I can now amass files with huge stack rates and build images in one pass.
Also, I like to be more hands on. Even when running my mount and telescope remotely from my home office.

So if I do get back to NINA, it will be more as saving and plate solving roll. And not as a all in one data acquisition program.
But so far, I'm enjoying my GM811G with the No-Flip RAEXT, and doing very in-depth file building. (1300 10 second subs, 13,000 seconds in an image = hell of a lot of detail, without processing.)
--
SonnyE


(I suggest viewed in full screen)

I would first check the Slew Speed setting in the Gemini 1.? It is usually set to 800.? Lowering to 600 may resolve the issue.? I had a similar issue with a Gemini 2.

Hi Alex,

If I understand,?
1: You have 2 mounts with Gemini-1 Controllers, 1 is "GM8", another is ??. ("Other")
2: Only the GM8 mount, with it's original Gemini-1 control box, has the motor stall symptom.
3: When you temporarily install the "Other" control box on the "GM8", the "GM8" works with no stalls.
4: After returning the Gemini-1 controllers to their original configuration, the "GM8" STILL works.
5: It's only upon the next night of operation or testing (the 'GM8's (mount), controllers, and it's motors, etc. have had time to cool back to ambient temp.) that the problem resurfaces.

Q1: How long does it take for the stall to occur ? Right away at startup? or after a little while??

Q2: When you take the original GM8's Gemini-1 controller and put it on the "Other" mount (cold), does it have motor stalls also ? (even if it has different motors, etc)?
(is it possible to perform this test, or are there other limiting factors?)

Presumably the cabling is the same during both tests, and the only thing that changes is the controller board, what else could be different?

Assuming the cabling, supply power, and mount motor load requirements are similar, isn't the only physical thing that is effectively changing, is the motors have been on (maybe for awhile),
compared to them being cold after a day of non-use.

Curious,
Astronut Tim

On startup, either cold or warm, my GM8 Gemini 1 has a motor stall.?

I swap in a different control box (from my other Gemini 1 mount) and there is no motor stall.?

Then, I swap in the first control box, and the mount operates fine.?

Next night I come back and that same control box has a stall on startup.?

IN other words, something on startup stalls the motor with one box, but not the other. And, once I use the second box, the first box will work properly.

Any ideas?

Alex

Here's an extreme example of the problem. This is a worm block commonly found in up market Celestron Nexstar mounts made in the first decade of the 2000s, like the Nexstar GPS Alt/Az and CGE mounts.?

While playing with on of these I notice there was a lot of play in the worm. Tightening the adjustment screw until the play disappeared produced a horrid notchiness like a out of round tire.?



Digging into the design of the block, this what I discovered. The block is one piece. In order to assemble it one end was bored to let the worm and the adjustment en bearing pass through it. The outer diameter of the worm is larger than the bearing fitted to it. Thus the bearing is left to float around in its seat.?