VintageBigBlue.org

 

Re: Please help us review Ovision results and understand them.


Apr 27, 2009

 


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

#42619 Apr 27, 2009

Hi All,



My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.



tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/



FS is mine, PO is Peter's and AB is Andrew's.



All three are rates as +/-4, all thre stat Simulated Seeing.

Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.



Few things we noted, all three seem to have very poor first period results as compared to 2nd period.



In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.



Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.



We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?



Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.



Regards

Fahim



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

#42624 Apr 27, 2009

Hi Fahim,



I took a look at your spreadsheet, but I hope others will double-check what I found.



The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.



The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.



What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).



Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:



cycle 1 is about 8 arcsecnds peak-to-peak

cycle 2 is about 5 arcseconds p2p

cycle 3 is about 6 arcseconds p2p



Regards,



-Paul



--- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@...> wrote:

>

> Hi All,

>

> My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

>

> tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

>

> FS is mine, PO is Peter's and AB is Andrew's.

>

> All three are rates as +/-4, all thre stat Simulated Seeing.

> Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

>

> Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

>

> In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

>

> Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

>

> We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

>

> Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

>

> Regards

> Fahim

>



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

#42625 Apr 27, 2009

Paul,



Thanks for your input, yes there does seem to be a change in p-p from cycle to cycle and no consistent pattern. This is what worried us all as all three seem to be similarly inconsistent.



Can you verify which file you reviewed. As I put 3 up there end in FS, AB and PO. I think perhaps you looked at AB, FS and PO by my review have a P-p of almost 10 in the first cycle.



Regards

Fahim

--- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@...> wrote:

>

> Hi Fahim,

>

> I took a look at your spreadsheet, but I hope others will double-check what I found.

>

> The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.

>

> The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.

>

> What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).

>

> Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:

>

> cycle 1 is about 8 arcsecnds peak-to-peak

> cycle 2 is about 5 arcseconds p2p

> cycle 3 is about 6 arcseconds p2p

>

> Regards,

>

> -Paul

>

>

> --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> >

> > Hi All,

> >

> > My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> > Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

> >

> > tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

> >

> > FS is mine, PO is Peter's and AB is Andrew's.

> >

> > All three are rates as +/-4, all thre stat Simulated Seeing.

> > Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

> >

> > Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

> >

> > In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

> >

> > Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

> >

> > We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

> >

> > Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

> >

> > Regards

> > Fahim

> >

>







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

#42628 Apr 27, 2009

Fahim,



The only file I looked at was the FS one (yours, I believe).



Here are the 3 cycles of data overlayed on one worm period:



pk.darkhorizons.org/tmp/fahim.jpg



The blue line (the bottom one) was cycle 1, the yellow line in the middle -- cycle 3, and the purple line, cycle 2.



Then, I removed a 30 second drift from the data to produce this:



pk.darkhorizons.org/tmp/fahim-drift.jpg



Don't know where the drift is coming from, if it is real... The vertical scale is arcseconds in both plots.



Regards,



-Paul



--- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@...> wrote:

>

> Paul,

>

> Thanks for your input, yes there does seem to be a change in p-p from cycle to cycle and no consistent pattern. This is what worried us all as all three seem to be similarly inconsistent.

>

> Can you verify which file you reviewed. As I put 3 up there end in FS, AB and PO. I think perhaps you looked at AB, FS and PO by my review have a P-p of almost 10 in the first cycle.

>

> Regards

> Fahim

>

> --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> >

> > Hi Fahim,

> >

> > I took a look at your spreadsheet, but I hope others will double-check what I found.

> >

> > The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.

> >

> > The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.

> >

> > What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).

> >

> > Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:

> >

> > cycle 1 is about 8 arcsecnds peak-to-peak

> > cycle 2 is about 5 arcseconds p2p

> > cycle 3 is about 6 arcseconds p2p

> >

> > Regards,

> >

> > -Paul

> >

> >

> > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > >

> > > Hi All,

> > >

> > > My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> > > Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

> > >

> > > tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

> > >

> > > FS is mine, PO is Peter's and AB is Andrew's.

> > >

> > > All three are rates as +/-4, all thre stat Simulated Seeing.

> > > Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

> > >

> > > Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

> > >

> > > In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

> > >

> > > Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

> > >

> > > We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

> > >

> > > Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

> > >

> > > Regards

> > > Fahim

> > >

> >

>



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

#42631 Apr 27, 2009

Paul,



Which sheet in the workbook and which column did you plot in particular. The PE is given in "measures" in column D. And I dont see a -8 in there. But your first graph shows a -8.



Ovision has used a graph of the raw data from the encoder to generate a trendline y=mx+b and then he has removed this drift line from the final values as calculated in column D of "measures". I also am not sure why there is a drift. But if you look at the other excel files in the files section for NS worms you will see some more pronounced drifts.



Regards

Fahim

--- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@...> wrote:

>

> Fahim,

>

> The only file I looked at was the FS one (yours, I believe).

>

> Here are the 3 cycles of data overlayed on one worm period:

>

> pk.darkhorizons.org/tmp/fahim.jpg

>

> The blue line (the bottom one) was cycle 1, the yellow line in the middle -- cycle 3, and the purple line, cycle 2.

>

> Then, I removed a 30 second drift from the data to produce this:

>

> pk.darkhorizons.org/tmp/fahim-drift.jpg

>

> Don't know where the drift is coming from, if it is real... The vertical scale is arcseconds in both plots.

>

> Regards,

>

> -Paul

>

>

> --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> >

> > Paul,

> >

> > Thanks for your input, yes there does seem to be a change in p-p from cycle to cycle and no consistent pattern. This is what worried us all as all three seem to be similarly inconsistent.

> >

> > Can you verify which file you reviewed. As I put 3 up there end in FS, AB and PO. I think perhaps you looked at AB, FS and PO by my review have a P-p of almost 10 in the first cycle.

> >

> > Regards

> > Fahim

> >

> > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > >

> > > Hi Fahim,

> > >

> > > I took a look at your spreadsheet, but I hope others will double-check what I found.

> > >

> > > The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.

> > >

> > > The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.

> > >

> > > What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).

> > >

> > > Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:

> > >

> > > cycle 1 is about 8 arcsecnds peak-to-peak

> > > cycle 2 is about 5 arcseconds p2p

> > > cycle 3 is about 6 arcseconds p2p

> > >

> > > Regards,

> > >

> > > -Paul

> > >

> > >

> > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > >

> > > > Hi All,

> > > >

> > > > My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> > > > Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

> > > >

> > > > tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

> > > >

> > > > FS is mine, PO is Peter's and AB is Andrew's.

> > > >

> > > > All three are rates as +/-4, all thre stat Simulated Seeing.

> > > > Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

> > > >

> > > > Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

> > > >

> > > > In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

> > > >

> > > > Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

> > > >

> > > > We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

> > > >

> > > > Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

> > > >

> > > > Regards

> > > > Fahim

> > > >

> > >

> >

>







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

#42634 Apr 27, 2009

Fahim, I used the DATA CODEUR sheet, columns "temps sec" and "reso 0.125", which are the same as "mesures" sheet columns A and B.



I first interpolated the data to 1 second intervals, and then split at 240 seconds into three cycles. I then normalized the starting points for the three cycles to produce the first plot.



The second plot was actually using a 60 seconds line fit to the data to remove drift (not 30 sec as I said in the post).



Regards,



-Paul

--- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@...> wrote:

>

> Paul,

>

> Which sheet in the workbook and which column did you plot in particular. The PE is given in "measures" in column D. And I dont see a -8 in there. But your first graph shows a -8.

>

> Ovision has used a graph of the raw data from the encoder to generate a trendline y=mx+b and then he has removed this drift line from the final values as calculated in column D of "measures". I also am not sure why there is a drift. But if you look at the other excel files in the files section for NS worms you will see some more pronounced drifts.

>

> Regards

> Fahim

>

> --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> >

> > Fahim,

> >

> > The only file I looked at was the FS one (yours, I believe).

> >

> > Here are the 3 cycles of data overlayed on one worm period:

> >

> > pk.darkhorizons.org/tmp/fahim.jpg

> >

> > The blue line (the bottom one) was cycle 1, the yellow line in the middle -- cycle 3, and the purple line, cycle 2.

> >

> > Then, I removed a 30 second drift from the data to produce this:

> >

> > pk.darkhorizons.org/tmp/fahim-drift.jpg

> >

> > Don't know where the drift is coming from, if it is real... The vertical scale is arcseconds in both plots.

> >

> > Regards,

> >

> > -Paul

> >

> >

> > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > >

> > > Paul,

> > >

> > > Thanks for your input, yes there does seem to be a change in p-p from cycle to cycle and no consistent pattern. This is what worried us all as all three seem to be similarly inconsistent.

> > >

> > > Can you verify which file you reviewed. As I put 3 up there end in FS, AB and PO. I think perhaps you looked at AB, FS and PO by my review have a P-p of almost 10 in the first cycle.

> > >

> > > Regards

> > > Fahim

> > >

> > > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > > >

> > > > Hi Fahim,

> > > >

> > > > I took a look at your spreadsheet, but I hope others will double-check what I found.

> > > >

> > > > The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.

> > > >

> > > > The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.

> > > >

> > > > What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).

> > > >

> > > > Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:

> > > >

> > > > cycle 1 is about 8 arcsecnds peak-to-peak

> > > > cycle 2 is about 5 arcseconds p2p

> > > > cycle 3 is about 6 arcseconds p2p

> > > >

> > > > Regards,

> > > >

> > > > -Paul

> > > >

> > > >

> > > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > > >

> > > > > Hi All,

> > > > >

> > > > > My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> > > > > Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

> > > > >

> > > > > tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

> > > > >

> > > > > FS is mine, PO is Peter's and AB is Andrew's.

> > > > >

> > > > > All three are rates as +/-4, all thre stat Simulated Seeing.

> > > > > Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

> > > > >

> > > > > Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

> > > > >

> > > > > In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

> > > > >

> > > > > Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

> > > > >

> > > > > We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

> > > > >

> > > > > Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

> > > > >

> > > > > Regards

> > > > > Fahim

> > > > >

> > > >

> > >

> >

>







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

#42636 Apr 28, 2009

Thanks for taking a look and offering your help Paul. I'm one of the recipients of the worms as well, and I produced a graph similar to yours, although I didn't subtract the trendline, I just subtracted the average value to have all 3 cycles average zero. My first cycle is strange how it moves upwards at the end which is not present on the other two cycles



andrew.bundschuh.googlepages.com/andrews-ovision-worm.jpg



Regards,

Andrew





--- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@...> wrote:

>

> Fahim, I used the DATA CODEUR sheet, columns "temps sec" and "reso 0.125", which are the same as "mesures" sheet columns A and B.

>

> I first interpolated the data to 1 second intervals, and then split at 240 seconds into three cycles. I then normalized the starting points for the three cycles to produce the first plot.

>

> The second plot was actually using a 60 seconds line fit to the data to remove drift (not 30 sec as I said in the post).

>

> Regards,

>

> -Paul

>

> --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> >

> > Paul,

> >

> > Which sheet in the workbook and which column did you plot in particular. The PE is given in "measures" in column D. And I dont see a -8 in there. But your first graph shows a -8.

> >

> > Ovision has used a graph of the raw data from the encoder to generate a trendline y=mx+b and then he has removed this drift line from the final values as calculated in column D of "measures". I also am not sure why there is a drift. But if you look at the other excel files in the files section for NS worms you will see some more pronounced drifts.

> >

> > Regards

> > Fahim

> >

> > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > >

> > > Fahim,

> > >

> > > The only file I looked at was the FS one (yours, I believe).

> > >

> > > Here are the 3 cycles of data overlayed on one worm period:

> > >

> > > pk.darkhorizons.org/tmp/fahim.jpg

> > >

> > > The blue line (the bottom one) was cycle 1, the yellow line in the middle -- cycle 3, and the purple line, cycle 2.

> > >

> > > Then, I removed a 30 second drift from the data to produce this:

> > >

> > > pk.darkhorizons.org/tmp/fahim-drift.jpg

> > >

> > > Don't know where the drift is coming from, if it is real... The vertical scale is arcseconds in both plots.

> > >

> > > Regards,

> > >

> > > -Paul

> > >

> > >

> > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > >

> > > > Paul,

> > > >

> > > > Thanks for your input, yes there does seem to be a change in p-p from cycle to cycle and no consistent pattern. This is what worried us all as all three seem to be similarly inconsistent.

> > > >

> > > > Can you verify which file you reviewed. As I put 3 up there end in FS, AB and PO. I think perhaps you looked at AB, FS and PO by my review have a P-p of almost 10 in the first cycle.

> > > >

> > > > Regards

> > > > Fahim

> > > >

> > > > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > > > >

> > > > > Hi Fahim,

> > > > >

> > > > > I took a look at your spreadsheet, but I hope others will double-check what I found.

> > > > >

> > > > > The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.

> > > > >

> > > > > The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.

> > > > >

> > > > > What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).

> > > > >

> > > > > Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:

> > > > >

> > > > > cycle 1 is about 8 arcsecnds peak-to-peak

> > > > > cycle 2 is about 5 arcseconds p2p

> > > > > cycle 3 is about 6 arcseconds p2p

> > > > >

> > > > > Regards,

> > > > >

> > > > > -Paul

> > > > >

> > > > >

> > > > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > > > >

> > > > > > Hi All,

> > > > > >

> > > > > > My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> > > > > > Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

> > > > > >

> > > > > > tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

> > > > > >

> > > > > > FS is mine, PO is Peter's and AB is Andrew's.

> > > > > >

> > > > > > All three are rates as +/-4, all thre stat Simulated Seeing.

> > > > > > Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

> > > > > >

> > > > > > Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

> > > > > >

> > > > > > In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

> > > > > >

> > > > > > Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

> > > > > >

> > > > > > We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

> > > > > >

> > > > > > Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

> > > > > >

> > > > > > Regards

> > > > > > Fahim

> > > > > >

> > > > >

> > > >

> > >

> >

>







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

#42637 Apr 28, 2009

Some insight into how Ovision get there Simulated data.

www.mda-telescoop.com/index.php?option=com_content&task=view&id=50&Itemid=80



The formating of the data is exactly the same.



Does anyone know how to get this data into Peas or PECPrep.



--- In Losmandy_users@yahoogroups.com, "andrew.bundschuh" andrew.bundschuh@...> wrote:

>

> Thanks for taking a look and offering your help Paul. I'm one of the recipients of the worms as well, and I produced a graph similar to yours, although I didn't subtract the trendline, I just subtracted the average value to have all 3 cycles average zero. My first cycle is strange how it moves upwards at the end which is not present on the other two cycles

>

> andrew.bundschuh.googlepages.com/andrews-ovision-worm.jpg

>

> Regards,

> Andrew

>

>

>

> --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> >

> > Fahim, I used the DATA CODEUR sheet, columns "temps sec" and "reso 0.125", which are the same as "mesures" sheet columns A and B.

> >

> > I first interpolated the data to 1 second intervals, and then split at 240 seconds into three cycles. I then normalized the starting points for the three cycles to produce the first plot.

> >

> > The second plot was actually using a 60 seconds line fit to the data to remove drift (not 30 sec as I said in the post).

> >

> > Regards,

> >

> > -Paul

> >

> > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > >

> > > Paul,

> > >

> > > Which sheet in the workbook and which column did you plot in particular. The PE is given in "measures" in column D. And I dont see a -8 in there. But your first graph shows a -8.

> > >

> > > Ovision has used a graph of the raw data from the encoder to generate a trendline y=mx+b and then he has removed this drift line from the final values as calculated in column D of "measures". I also am not sure why there is a drift. But if you look at the other excel files in the files section for NS worms you will see some more pronounced drifts.

> > >

> > > Regards

> > > Fahim

> > >

> > > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > > >

> > > > Fahim,

> > > >

> > > > The only file I looked at was the FS one (yours, I believe).

> > > >

> > > > Here are the 3 cycles of data overlayed on one worm period:

> > > >

> > > > pk.darkhorizons.org/tmp/fahim.jpg

> > > >

> > > > The blue line (the bottom one) was cycle 1, the yellow line in the middle -- cycle 3, and the purple line, cycle 2.

> > > >

> > > > Then, I removed a 30 second drift from the data to produce this:

> > > >

> > > > pk.darkhorizons.org/tmp/fahim-drift.jpg

> > > >

> > > > Don't know where the drift is coming from, if it is real... The vertical scale is arcseconds in both plots.

> > > >

> > > > Regards,

> > > >

> > > > -Paul

> > > >

> > > >

> > > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > > >

> > > > > Paul,

> > > > >

> > > > > Thanks for your input, yes there does seem to be a change in p-p from cycle to cycle and no consistent pattern. This is what worried us all as all three seem to be similarly inconsistent.

> > > > >

> > > > > Can you verify which file you reviewed. As I put 3 up there end in FS, AB and PO. I think perhaps you looked at AB, FS and PO by my review have a P-p of almost 10 in the first cycle.

> > > > >

> > > > > Regards

> > > > > Fahim

> > > > >

> > > > > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > > > > >

> > > > > > Hi Fahim,

> > > > > >

> > > > > > I took a look at your spreadsheet, but I hope others will double-check what I found.

> > > > > >

> > > > > > The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.

> > > > > >

> > > > > > The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.

> > > > > >

> > > > > > What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).

> > > > > >

> > > > > > Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:

> > > > > >

> > > > > > cycle 1 is about 8 arcsecnds peak-to-peak

> > > > > > cycle 2 is about 5 arcseconds p2p

> > > > > > cycle 3 is about 6 arcseconds p2p

> > > > > >

> > > > > > Regards,

> > > > > >

> > > > > > -Paul

> > > > > >

> > > > > >

> > > > > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > > > > >

> > > > > > > Hi All,

> > > > > > >

> > > > > > > My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> > > > > > > Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

> > > > > > >

> > > > > > > tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

> > > > > > >

> > > > > > > FS is mine, PO is Peter's and AB is Andrew's.

> > > > > > >

> > > > > > > All three are rates as +/-4, all thre stat Simulated Seeing.

> > > > > > > Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

> > > > > > >

> > > > > > > Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

> > > > > > >

> > > > > > > In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

> > > > > > >

> > > > > > > Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

> > > > > > >

> > > > > > > We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

> > > > > > >

> > > > > > > Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

> > > > > > >

> > > > > > > Regards

> > > > > > > Fahim

> > > > > > >

> > > > > >

> > > > >

> > > >

> > >

> >

>







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

#42640 Apr 28, 2009

Hi Andrew,



Yes, that is a strange behavior, and I'm not at all sure that this drift is inherent in the worm assembly. More likely it is due to the mechanism that drives it (do we know if Ovision uses G11's to test their worms?) or due to a measurement error.



I can easily imagine some backlash in the worm/gear or in the axis/encoder couplings causing drift. Without knowing the exact testing setup it's hard to know. The fact that this non-periodic drift exists for multiple worms and multiple test runs would indicate that it's probably an issue with the test setup and not the worm itself.



Regards,



-Paul

--- In Losmandy_users@yahoogroups.com, "andrew.bundschuh" andrew.bundschuh@...> wrote:

>

> Thanks for taking a look and offering your help Paul. I'm one of the recipients of the worms as well, and I produced a graph similar to yours, although I didn't subtract the trendline, I just subtracted the average value to have all 3 cycles average zero. My first cycle is strange how it moves upwards at the end which is not present on the other two cycles

>

> andrew.bundschuh.googlepages.com/andrews-ovision-worm.jpg

>

> Regards,

> Andrew

>

>

>

> --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> >

> > Fahim, I used the DATA CODEUR sheet, columns "temps sec" and "reso 0.125", which are the same as "mesures" sheet columns A and B.

> >

> > I first interpolated the data to 1 second intervals, and then split at 240 seconds into three cycles. I then normalized the starting points for the three cycles to produce the first plot.

> >

> > The second plot was actually using a 60 seconds line fit to the data to remove drift (not 30 sec as I said in the post).

> >

> > Regards,

> >

> > -Paul

> >

> > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > >

> > > Paul,

> > >

> > > Which sheet in the workbook and which column did you plot in particular. The PE is given in "measures" in column D. And I dont see a -8 in there. But your first graph shows a -8.

> > >

> > > Ovision has used a graph of the raw data from the encoder to generate a trendline y=mx+b and then he has removed this drift line from the final values as calculated in column D of "measures". I also am not sure why there is a drift. But if you look at the other excel files in the files section for NS worms you will see some more pronounced drifts.

> > >

> > > Regards

> > > Fahim

> > >

> > > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > > >

> > > > Fahim,

> > > >

> > > > The only file I looked at was the FS one (yours, I believe).

> > > >

> > > > Here are the 3 cycles of data overlayed on one worm period:

> > > >

> > > > pk.darkhorizons.org/tmp/fahim.jpg

> > > >

> > > > The blue line (the bottom one) was cycle 1, the yellow line in the middle -- cycle 3, and the purple line, cycle 2.

> > > >

> > > > Then, I removed a 30 second drift from the data to produce this:

> > > >

> > > > pk.darkhorizons.org/tmp/fahim-drift.jpg

> > > >

> > > > Don't know where the drift is coming from, if it is real... The vertical scale is arcseconds in both plots.

> > > >

> > > > Regards,

> > > >

> > > > -Paul

> > > >

> > > >

> > > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > > >

> > > > > Paul,

> > > > >

> > > > > Thanks for your input, yes there does seem to be a change in p-p from cycle to cycle and no consistent pattern. This is what worried us all as all three seem to be similarly inconsistent.

> > > > >

> > > > > Can you verify which file you reviewed. As I put 3 up there end in FS, AB and PO. I think perhaps you looked at AB, FS and PO by my review have a P-p of almost 10 in the first cycle.

> > > > >

> > > > > Regards

> > > > > Fahim

> > > > >

> > > > > --- In Losmandy_users@yahoogroups.com, "Paul K" pkane2001@> wrote:

> > > > > >

> > > > > > Hi Fahim,

> > > > > >

> > > > > > I took a look at your spreadsheet, but I hope others will double-check what I found.

> > > > > >

> > > > > > The measurement appears to be done with an absolute high-precision encoder on the axis. In other words, it's not done with a star or a telescope, but by directly measuring the deviation from sidereal rotation rate by reading encoder position.

> > > > > >

> > > > > > The encoder resolution is given as 0.125 arcseconds. The simulated seeing was added in after the measurement, to make the data appear more realistic. There was no seeing error recorded during measurement, since the measurement didn't involve the atmosphere.

> > > > > >

> > > > > > What I find confusing is that it doesn't appear very periodic with the G11 240 second period... There are some similarities between the three worm cycles, but generally they are not the same in amplitude, or even in the direction of change. This either indicates a large random error in the drive mechanism or in the the measurement system (encoders).

> > > > > >

> > > > > > Because the error is not very periodic, an FFT or similar kind of analysis is probably not going to yield useful information. The error magnitude varies between the three recorded cycles:

> > > > > >

> > > > > > cycle 1 is about 8 arcsecnds peak-to-peak

> > > > > > cycle 2 is about 5 arcseconds p2p

> > > > > > cycle 3 is about 6 arcseconds p2p

> > > > > >

> > > > > > Regards,

> > > > > >

> > > > > > -Paul

> > > > > >

> > > > > >

> > > > > > --- In Losmandy_users@yahoogroups.com, "fahimshariff" fahimshariff@> wrote:

> > > > > > >

> > > > > > > Hi All,

> > > > > > >

> > > > > > > My friends and I recently got our Ovision worms and we each picked one out of a hat. I have posted the results in this folder under files.

> > > > > > > Note that you may have to change the Hex2dec function to hexdec, as it came from Ovision with hexdec but my version of excel 2007 did not recognise this function.

> > > > > > >

> > > > > > > tech.groups.yahoo.com/group/Losmandy_users/files/Ovision%20Worm%20Test%20Graphs/

> > > > > > >

> > > > > > > FS is mine, PO is Peter's and AB is Andrew's.

> > > > > > >

> > > > > > > All three are rates as +/-4, all thre stat Simulated Seeing.

> > > > > > > Can someone please explain what Simulated seeing is? One of the other users here posted his results called "Blue" has actual scope and camera information and the seeing data does not mention Simulated. So what is the difference? How is simulated data diffrent.

> > > > > > >

> > > > > > > Few things we noted, all three seem to have very poor first period results as compared to 2nd period.

> > > > > > >

> > > > > > > In AB's case the 2nd period is outstanding almost with 2arcsec error. But first and 2nd should about 6-7.

> > > > > > >

> > > > > > > Mine and Peters are more in line with first period error almost 10 and 2nd ad third about 7.5-8.

> > > > > > >

> > > > > > > We would really appreciate the help of those who understand these tihngs to explain them to us. Are we reading these results incorrectly?

> > > > > > >

> > > > > > > Also could someone help convert them to FFT plots? IT would be good if you could tell us how to do this, its been 10-12 years since i touched FFT for Antenna analysis. So I could use the help. I am wiling to have a go but would like some help.

> > > > > > >

> > > > > > > Regards

> > > > > > > Fahim

> > > > > > >

> > > > > >

> > > > >

> > > >

> > >

> >

>







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

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