Topics

Q. for Ray Gralak about PEMPro


CurtisC
 

Ray: As you know, I've had considerable angst about the uncorrected PE of my 2010-vintage Mach1GTO.  We had a lengthy discussion about this a few days ago in one of the other threads.  You looked at my logs.  I can't definitively explain why one run on Sep 2 measured only 6-something p-to-p (which is the best number I've ever achieved), and then the next run on the same evening measured 11-something.  I consider these to be unrepeatable anomalies.  In fact, all of my other recent runs (and I've done a lot of them) are in the area of 17 to 19.  My latest one yields 0.87 with correction, even though the raw curve is 17.49.  One could argue that I should be happy with the corrected curve and drop the whole matter.  

In any case, I have a question.

I found this comment from you in 2017: "The only fundamental you should try to measure is the 1.0x fundamental, so uncheck all the other fundamentals because they are random false fundamentals." Really? I found a comment from George W. back in 2014 saying we should use fundamentals 1, 2, and 4 for the Mach1. I think 2 and 4 relate to the periods of the spur gears. Should we use only fundamental 1 with the Mach1GTO, or should we use 1, 2, and 4, or should we use something else? When I run PEMPro, it automatically checks 1, 2, and 3. I found that unchecking 3 and checking 4 makes no visible difference in the curve, but then I read a comment from you somewhere that PEMPro doesn't recalculate the displayed curve if we fiddle with the checked fundamentals.


Ray Gralak
 

Hi Curtis,

I can't definitively explain why one run on Sep 2 measured only
6-something p-to-p (which is the best number I've ever
achieved), and then the next run on the same evening measured
11-something.
You are right that theer isn't a definitive explanation, but I think I mentioned some possible causes. For example:
* Non-optimal gear meshing.
* Debris in the grease
* Damaged worm wheel teeth.

I found this comment from you in 2017: "The only fundamental
you should try to measure is the 1.0x fundamental,
so uncheck all the other fundamentals because they are random
false fundamentals." Really?
Yes, but it depends on what the data tells you. Usually, you will want just the 1.0x fundamental unless the graph and frequency spectrum clearly show other fundamentals repeated between multiple worm cycles. This, is the case in your latest data.

However, at least one of those fundamentals doesn't have a stable amplitude, so you may not be able to completely eliminate it. Because it is not stable, you can't tell if it might have higher or lower amplitudes on other parts of the worm wheel, so it might be better NOT to try to fix that fundamental.

somewhere that PEMPro doesn't recalculate the displayed curve if
we fiddle with the checked fundamentals.
When I run PEMPro, it automatically checks 1, 2, and 3. I found that
unchecking 3 and checking 4 makes no visible difference in the curve,
but then I read a comment from you somewhere that PEMPro doesn't
recalculate the displayed curve if we fiddle with the checked fundamentals.
I don't know where you saw that, but I think you might be taking that out of context (or I had my coffee yet! :-). When you are in the Create PEC Curve window in PEMPro, depending on the amplitude of the frequency you might not see much change, but PEMPro recalculates the curve every time a checkbox is enabled or disabled. It has always worked that way.

-Ray Gralak
Author of PEMPro
Author of APCC (Astro-Physics Command Center): https://www.astro-physics.com/apcc-pro
Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of CurtisC via groups.io
Sent: Saturday, September 12, 2020 1:12 AM
To: main@ap-gto.groups.io
Subject: [ap-gto] Q. for Ray Gralak about PEMPro

Ray: As you know, I've had considerable angst about the uncorrected PE of my 2010-vintage Mach1GTO. We had
a lengthy discussion about this a few days ago in one of the other threads. You looked at my logs. I can't
definitively explain why one run on Sep 2 measured only 6-something p-to-p (which is the best number I've ever
achieved), and then the next run on the same evening measured 11-something. I consider these to be
unrepeatable anomalies. In fact, all of my other recent runs (and I've done a lot of them) are in the area of 17 to
19. My latest one yields 0.87 with correction, even though the raw curve is 17.49. One could argue that I should
be happy with the corrected curve and drop the whole matter.

In any case, I have a question.

I found this comment from you in 2017: "The only fundamental you should try to measure is the 1.0x fundamental,
so uncheck all the other fundamentals because they are random false fundamentals." Really? I found a comment
from George W. back in 2014 saying we should use fundamentals 1, 2, and 4 for the Mach1. I think 2 and 4 relate
to the periods of the spur gears. Should we use only fundamental 1 with the Mach1GTO, or should we use 1, 2,
and 4, or should we use something else? When I run PEMPro, it automatically checks 1, 2, and 3. I found that
unchecking 3 and checking 4 makes no visible difference in the curve, but then I read a comment from you
somewhere that PEMPro doesn't recalculate the displayed curve if we fiddle with the checked fundamentals.


CurtisC
 

Thank you for your reply.  Swapping fundamental 3 for 4 didn't make any visible difference in the graph.  When I tried going only with fundamental 1 once or twice, I did get a distinctly better result.  Remeshing the worm hasn't had much effect, provided I use the same fundamentals.  It's certainly possible that I've been doing the remesh incorrectly since 2010.  I wonder what fundamentals A-P used when they generated the original PEMPro curves at the factory.

Thanks again, Ray.


Ray Gralak
 

Hi Curtis,

Swapping fundamental 3 for 4 didn't make any visible difference in the graph.
If the 3rd and 4th fundamentals are small, then you may not see much difference. If you want to verify the graph is changing, then try unchecking the 1x fundamental, or even all the fundamentals. Please report back if you think that the graph is still not changing for you.

-Ray Gralak
Author of PEMPro
Author of APCC (Astro-Physics Command Center): https://www.astro-physics.com/apcc-pro
Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver


-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of CurtisC via groups.io
Sent: Saturday, September 12, 2020 10:42 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] Q. for Ray Gralak about PEMPro

Thank you for your reply. Swapping fundamental 3 for 4 didn't make any visible difference in the graph. When I
tried going only with fundamental 1 once or twice, I did get a distinctly better result. Remeshing the worm hasn't
had much effect, provided I use the same fundamentals. It's certainly possible that I've been doing the remesh
incorrectly since 2010. I wonder what fundamentals A-P used when they generated the original PEMPro curves at
the factory.

Thanks again, Ray.


CurtisC
 
Edited

I've just spent some time in the observatory playing with last night's "best" run, the one that I loaded into the mount.  Unchecking 3 and checking 4 does make a difference, not so much on the graph (although there is a small change), but in the reported p-to-p.  With 3 on and 4 off the p-to-p is 18.26.  With 3 off and 4 on it drops to 17.47.  That's the one I sent to the mount last night.  I also tried using only fundamental 1.  I forget the resulting p-to-p, but it didn't change 17-18 to 3 or 4.  The smoky sky reduces me to worrying about periodic error instead of imaging.  I can image stars well enough to do PEMPro runs.  It's a shame, because seeing is very good.


CurtisC
 

On Sat, Sep 12, 2020 at 11:34 AM, Ray Gralak wrote:
* Damaged worm wheel teeth.
The curve looks smooth to me.  I'd think it would show some irregularities if the worm were damaged.  But you looked at my logs.  What do you think?


Ray Gralak
 

Hi Curtis,

 

>          * Damaged worm wheel teeth.

>

> The curve looks smooth to me.  I'd think it would show some

> irregularities if the worm were damaged.  But you

> looked at my logs.  What do you think?

 

I think you are confusing the worm gear with the worm wheel. The worm wheel is the large gear with 225 teeth that drives the mount. It rotates once per sidereal day. The worm turns one revolution for each of the 225 teeth, thus 225 times in a sidereal day (which is about 4 minutes less than 24 hours). Here's an example screenshot of an online image of a worm and worm wheel:

 

 

 

When you capture periodic error over, say 6 worm cycles, you are measuring the periodic error over 6 of the 225 teeth of the worm wheel. The shape of each worm wheel tooth can contribute to tracking error recorded by PEMPro. That's why in some PE graphs the shape is not precisely the same between worm cycles.

 

If some of the worm wheel teeth are malformed or have large particles, that can cause the corresponding worm cycles to look different. If some worm wheel teeth are damaged or not spaced evenly, this could cause the periodic error to be larger than that contributed by the worm alone. Also, improper gear meshing can cause varying worm peak/peak values because the worm gear can ride too tightly (or loosely) on the worm wheel.

 

-Ray Gralak

Author of PEMPro

Author of APCC (Astro-Physics Command Center): https://www.astro-physics.com/apcc-pro

Author of Astro-Physics V2 ASCOM Driver: https://www.siriusimaging.com/apdriver

 

 

> -----Original Message-----

> From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of CurtisC via groups.io

> Sent: Saturday, September 12, 2020 8:55 PM

> To: main@ap-gto.groups.io

> Subject: Re: [ap-gto] Q. for Ray Gralak about PEMPro

>

> On Sat, Sep 12, 2020 at 11:34 AM, Ray Gralak wrote:

>

>

>          * Damaged worm wheel teeth.

>

> The curve looks smooth to me.  I'd think it would show some irregularities if the worm were damaged.  But you

> looked at my logs.  What do you think?


CurtisC
 

Good point!  I read "worm," I think of the worm, not the "worm wheel."


CurtisC
 

But I do know the difference.  It was my error for not reading your post correctly. 

For the record, I have no reason to think that the worm gear has suffered any damage in the 10+ years I've owned the mount.  No, the worm was never scraped across the gear.  It never happened.


Christopher Erickson
 

The worm is also known as the worm gear.

The worm wheel is always known as the worm wheel.

Calling the worm wheel the worm gear is incorrect.


On Sun, Sep 13, 2020 at 10:30 AM CurtisC via groups.io <calypte=verizon.net@groups.io> wrote:
But I do know the difference.  It was my error for not reading your post correctly. 

For the record, I have no reason to think that the worm gear has suffered any damage in the 10+ years I've owned the mount.  No, the worm was never scraped across the gear.  It never happened.


CurtisC
 

Thank you for the clarification.