VintageBigBlue.org

 

Re: [Losmandy_users] RE: PE Shots


Mar 14, 2001

 


----------------------------

#3209 Mar 14, 2001

Well I finally got a partially cloudy where I could just set up the

Gemini/G-11 system and play around with it. I have uploaded to the

files section two jpegs that show what my PE looked like before and

after installing the Gemini system. No bearings were replaced or

added just the motors replaced. I hope this doesn't start a firestorm

about motors or anything but I thought you guys would like to see

what the results were. Of course I had set up the mount a couple of

times between so could not reproduce the exact offset from the pole

on both shots. Both shots were done with the same CCD camera and OTA

and at the same temps and durations.



Stef



----------------------------

#3210 Mar 14, 2001

Stef,



it's 0:45 a.m. here but you've already made my day (mmh ... if you would

have named the file PE_WITH_Gemini.jpg ;-))

> Well I finally got a partially cloudy where I could just set up the

> Gemini/G-11 system and play around with it. I have uploaded to the

> files section two jpegs that show what my PE looked like before and

> after installing the Gemini system. No bearings were replaced or

> added just the motors replaced. I hope this doesn't start a firestorm

> about motors or anything but I thought you guys would like to see

> what the results were. Of course I had set up the mount a couple of

> times between so could not reproduce the exact offset from the pole

> on both shots. Both shots were done with the same CCD camera and OTA

> and at the same temps and durations.



Now what there is really PE. With the continous tracking of the servos

the inertia of the OTA helps to smooth out the curve. With steppers,

the inertia is a problem at the beginning of a step and at the end of a

step and if it get to near to the Eigenfrequency (own frequency) it may

get berserk but even a little difference f.i. in friction results in

visible peaks.



Thank you,

Rene'



----------------------------

#3211 Mar 14, 2001

Hi Rene',



Congratulations. That is an impressive result.



You may be interested in some tests I have been doing. There was a marked

improvement in tracking when I started to drive my G-11 with Mel's PC based

scope control program. The primary difference, I felt, was due to

microstepped vs halfstepped motors.



I wanted to analyze the actual shaft angle feeding the worm plotted against

time. The test rig I made up is a 2048 CPR encoder coupled to the output

shaft of the gearhead. This is read in quadrature by a little program I

wrote that samples the encoder via the PC parallel port. The pathlengths

through the code have been made as equal as possible without coding in

assembly. On a P133, I can sample at around 50K per second. Here are a

some links to a screen capture of a graph of the Hurst ABS gearhead output

and the output of a Vexta 1.8 degree motor adapted to a Hurst 20:1 gearbox

driven by Mel's code. Samples run in the X axis and encoder count runs in

the Y. The data represents about .6 of a second. There is also a pic of

the modified gearhead itself.



www.ameritech.net/users/mniemi000/Hrdcpy.jpg

www.ameritech.net/users/mniemi000/vexta2.jpg



-Marty



Martin Niemi



www.ameritech.net/users/mniemi000/index.html



Mailto:mniemi000@...



----- Original Message -----

From: goerlich@...>

To: Losmandy_users@yahoogroups.com>

Sent: Wednesday, March 14, 2001 6:45 PM

Subject: Re: [Losmandy_users] RE: PE Shots





> Stef,

>

> it's 0:45 a.m. here but you've already made my day (mmh ... if you would

> have named the file PE_WITH_Gemini.jpg ;-))

>

> > Well I finally got a partially cloudy where I could just set up the

> > Gemini/G-11 system and play around with it. I have uploaded to the

> > files section two jpegs that show what my PE looked like before and

> > after installing the Gemini system. No bearings were replaced or

> > added just the motors replaced. I hope this doesn't start a firestorm

> > about motors or anything but I thought you guys would like to see

> > what the results were. Of course I had set up the mount a couple of

> > times between so could not reproduce the exact offset from the pole

> > on both shots. Both shots were done with the same CCD camera and OTA

> > and at the same temps and durations.

>

> Now what there is really PE. With the continous tracking of the servos

> the inertia of the OTA helps to smooth out the curve. With steppers,

> the inertia is a problem at the beginning of a step and at the end of a

> step and if it get to near to the Eigenfrequency (own frequency) it may

> get berserk but even a little difference f.i. in friction results in

> visible peaks.

>

> Thank you,

> Rene'

>

>

> To unsubscribe from this group, send an email to:

> Losmandy_users-unsubscribe@egroups.com

>

>

>

> Your use of Yahoo! Groups is subject to docs.yahoo.com/info/terms/

>

>

>







----------------------------

#3215 Mar 14, 2001

Hi Marty,

> Congratulations. That is an impressive result.



Thank you. My part is only the precise pulsing of the servo part

that was developed by Scott and the people of Aveox.

> You may be interested in some tests I have been doing. There was a marked

> improvement in tracking when I started to drive my G-11 with Mel's PC based

> scope control program. The primary difference, I felt, was due to

> microstepped vs halfstepped motors.



That's exactly where I came from. The first incarnations of the Gemini

were microstepping (and they are still alive). It was a big difference, really,

much smoother, less prone to vibrations, less current consumption. Even better

with other motors (I am the guy that disassembled the Hurst motors and adjusted

the stepper cages, www.tfh-berlin.de/~goerlich/cg11tune.html 8-))

But it was always a compromise between speed, step rate, dynamics, accuracy (not

to loose steps), power consumption, electronic control.

At the time when the basic functionality was implemented Scott and I met and

our developments complemented each other. If there is the possibility

to use servos and to control them completely in a closed loop, there is no

question what to use.

> I wanted to analyze the actual shaft angle feeding the worm plotted against

> time. The test rig I made up is a 2048 CPR encoder coupled to the output

> shaft of the gearhead. This is read in quadrature by a little program I

> wrote that samples the encoder via the PC parallel port. The pathlengths

> through the code have been made as equal as possible without coding in

> assembly. On a P133, I can sample at around 50K per second. Here are a

> some links to a screen capture of a graph of the Hurst ABS gearhead output

> and the output of a Vexta 1.8 degree motor adapted to a Hurst 20:1 gearbox

> driven by Mel's code. Samples run in the X axis and encoder count runs in

> the Y. The data represents about .6 of a second. There is also a pic of

> the modified gearhead itself.

>

> www.ameritech.net/users/mniemi000/Hrdcpy.jpg

> www.ameritech.net/users/mniemi000/vexta2.jpg



Hey, you cannibalized poor old 5.25" floppy drives, didn't you?? I still can't

throw away this drives without keeping the stepper 8-)

Their design is much better, and they are quite expensive if you buy them new.

I've made some experiments with them too, also 10:1 microstepping, and I think

the result were similiar, but I did not measure them accurately.



Rene'



----------------------------

#3218 Mar 14, 2001

--- In Losmandy_users@y..., goerlich@t... wrote:

Rene'



Your wish is my command. It is now titled PE_With_Gemini.jpg :)

BTW the pointing was awesome last night. It was regularly putting the

Goto objects into a 0.40degree FOV eyepiece everytime after I made

the first 4 to 5 alignments. It was also pegging Saturn and Jupiter

even after meridian flops in goto seeks from one side of the sky to

the other. The only time I had to realign was when I had bumped the

scope and the encoders kicked in. They were close but still not as

dead on as the motor's encoders. Hopefully on the weekend of the 24th

we will get clear weather so I can run my G-11/Gemini and new GT-1

side by side and see how they work for dropping stuff on CCD chips

and handle the tracking. Based on the results so far, both of them

are gonna be real workhorses and used a good bit.



Stef



> Stef,

>

> it's 0:45 a.m. here but you've already made my day (mmh ... if you

would

> have named the file PE_WITH_Gemini.jpg ;-))

>

>

> Thank you,

> Rene'



----------------------------

#3233 Mar 15, 2001

Stef,

> Your wish is my command. It is now titled PE_With_Gemini.jpg :)



Thank you 8-)

> BTW the pointing was awesome last night. It was regularly putting the

> Goto objects into a 0.40degree FOV eyepiece everytime after I made

> the first 4 to 5 alignments. It was also pegging Saturn and Jupiter

> even after meridian flops in goto seeks from one side of the sky to

> the other. The only time I had to realign was when I had bumped the

> scope and the encoders kicked in. They were close but still not as

> dead on as the motor's encoders. Hopefully on the weekend of the 24th

> we will get clear weather so I can run my G-11/Gemini and new GT-1

> side by side and see how they work for dropping stuff on CCD chips

> and handle the tracking. Based on the results so far, both of them

> are gonna be real workhorses and used a good bit.



I was astonished how good the modelling can work after some alignments.

You could have the Gemini switched off shortly and select the Warm Restart

after power up and use the Initial (Re)Alignment to synchronize without

changing the modelling after the bump.

The good news is, that software version 1.3 will include a "Synchronize"

that will do exactly the same without switching off/on.



Rene'



----------------------------

#3249 Mar 15, 2001

Interesting comments. Inertia...how about a flywheel on the worm to provide

more inertia to smooth out the herky jerky stepper .....???



goerlich@... wrote:

> Stef,

>

> it's 0:45 a.m. here but you've already made my day (mmh ... if you would

> have named the file PE_WITH_Gemini.jpg ;-))

>

> > Well I finally got a partially cloudy where I could just set up the

> > Gemini/G-11 system and play around with it. I have uploaded to the

> > files section two jpegs that show what my PE looked like before and

> > after installing the Gemini system. No bearings were replaced or

> > added just the motors replaced. I hope this doesn't start a firestorm

> > about motors or anything but I thought you guys would like to see

> > what the results were. Of course I had set up the mount a couple of

> > times between so could not reproduce the exact offset from the pole

> > on both shots. Both shots were done with the same CCD camera and OTA

> > and at the same temps and durations.

>

> Now what there is really PE. With the continous tracking of the servos

> the inertia of the OTA helps to smooth out the curve. With steppers,

> the inertia is a problem at the beginning of a step and at the end of a

> step and if it get to near to the Eigenfrequency (own frequency) it may

> get berserk but even a little difference f.i. in friction results in

> visible peaks.

>

> Thank you,

> Rene'

>

>

> To unsubscribe from this group, send an email to:

> Losmandy_users-unsubscribe@egroups.com

>

>

>

> Your use of Yahoo! Groups is subject to docs.yahoo.com/info/terms/



--

Rob Watson

members.home.net/robonthemove/







Contact Us
This Site's Privacy Policy
Google's privacy policies

S
e
n
i
o
r
T
u
b
e
.
o
r
g