Re: [MaxImDL] Re: Pulse Guiding and SBIG AO8 - Bump and Error Q's


Sep 28, 2012

 


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

#50936 Sep 28, 2012

1. If I use pulse guiding to guide my scope/mount, will my camera's AO8 "bump" still get to the mount when it reaches its tilt-tip limits? The AO8 setup in MaxImDL sort of implies that it operates only with the guider relays, but I may be misunderstanding.



2. Also, is there an easy way to confirm that a bump occurs based on the percentage error measured? Do I have to set up the bump, start guiding with the AO8, then manually slew the scope a couple of arcsecs to see if the bump happens? And is there a log I can check to see if and when the bump occurred, and how much it moved the mount?



3. Finally, can the AO8 produce a near-perfect guiding experience (assuming excellent mount alignment and PEC, local seeing quality notwithstanding), or does it always have an error offset? Control theory says a gain-only (or proportional-only, for theory purists) feedback loop will always have an error offset, which I believe applies to the AO8 response to error signals. Offset is usually reduced (or anticipated) by adding integral and derivative control to proportional control, but I don't think these enhanced control algorithms are part of the AO8 controller - or are they? If they are, the setup dialog boxes don't offer an opportunity to change I-D settings.



Just wondering about Q3... not really that important in the larger scheme of things.

Rod



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

#50939 Sep 28, 2012

On 2012-09-28 3:00 PM, Rod wrote: > 1. If I use pulse guiding to guide my scope/mount, will my camera's AO8

> "bump" still get to the mount when it reaches its tilt-tip limits? The AO8

> setup in MaxImDL sort of implies that it operates only with the guider

> relays, but I may be misunderstanding.



Yes. That is a relatively new feature (V5.13).

> 2. Also, is there an easy way to confirm that a bump occurs based on the

> percentage error measured?



You should see that the the AO Tilt position stays within the "Activation Level

+/-%" control at all times. If you want to deliberately move the telescope out

of the window then you should see it pull itself back again.

> 3. Finally, can the AO8 produce a near-perfect guiding experience (assuming

> excellent mount alignment and PEC, local seeing quality notwithstanding), or

> does it always have an error offset? Control theory says a gain-only (or

> proportional-only, for theory purists) feedback loop will always have an

> error offset, which I believe applies to the AO8 response to error signals.

> Offset is usually reduced (or anticipated) by adding integral and derivative

> control to proportional control, but I don't think these enhanced control

> algorithms are part of the AO8 controller - or are they? If they are, the

> setup dialog boxes don't offer an opportunity to change I-D settings.



AO is designed to operate fairly quickly, and there is no mechanical lag as you

will find with conventional telescope mount guiding. As a result any residual

is extremely tiny - less than the "noise" due to seeing etc.



Doug



--



Doug George

dgeorge@...



Diffraction Limited

Makers of Cyanogen Imaging Products

www.cyanogen.com/



100 Craig Henry Dr., Suite 202

Ottawa, Ontario,

Canada, K2G 5W3



Phone: (613) 225-2732

Fax: (613) 225-9688



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

#50943 Sep 28, 2012

Q1. Ah, so the bump signal follows the guiding configuration. Good.



Q2. So if I set an activation level of +/- 25%, I should see a bump response when the AO8 gets below 25% or above 75%. Then depending on how many 100 ths of a sec I set for the bump signal, I should see the tracking return to somewhere between 25% and 75%...



Q3. Um, ok - BTW, I wasn't complaining about the tracking, just wondering if I should spend time trying to get to a "zero" error on the graph. Since the RMS error is always on one side of the axis, which makes sense at least in RA, I wasn't sure if the offset was my guiding, my seeing limits, or an imcomplete AO8 setup. From what you wrote, it sounds like I-D control isn't required or implemented. Good enough.



Rod

--- In MaxImDL@yahoogroups.com, "Douglas B. George" dg@...> wrote:

>

> On 2012-09-28 3:00 PM, Rod wrote:

> > 1. If I use pulse guiding to guide my scope/mount, will my camera's AO8

> > "bump" still get to the mount when it reaches its tilt-tip limits? The AO8

> > setup in MaxImDL sort of implies that it operates only with the guider

> > relays, but I may be misunderstanding.

>

> Yes. That is a relatively new feature (V5.13).

>

> > 2. Also, is there an easy way to confirm that a bump occurs based on the

> > percentage error measured?

>

> You should see that the the AO Tilt position stays within the "Activation Level

> +/-%" control at all times. If you want to deliberately move the telescope out

> of the window then you should see it pull itself back again.

>

> > 3. Finally, can the AO8 produce a near-perfect guiding experience (assuming

> > excellent mount alignment and PEC, local seeing quality notwithstanding), or

> > does it always have an error offset? Control theory says a gain-only (or

> > proportional-only, for theory purists) feedback loop will always have an

> > error offset, which I believe applies to the AO8 response to error signals.

> > Offset is usually reduced (or anticipated) by adding integral and derivative

> > control to proportional control, but I don't think these enhanced control

> > algorithms are part of the AO8 controller - or are they? If they are, the

> > setup dialog boxes don't offer an opportunity to change I-D settings.

>

> AO is designed to operate fairly quickly, and there is no mechanical lag as you

> will find with conventional telescope mount guiding. As a result any residual

> is extremely tiny - less than the "noise" due to seeing etc.

>

> Doug

>

> --

>

> Doug George

> dgeorge@...

>

> Diffraction Limited

> Makers of Cyanogen Imaging Products

> www.cyanogen.com/

>

> 100 Craig Henry Dr., Suite 202

> Ottawa, Ontario,

> Canada, K2G 5W3

>

> Phone: (613) 225-2732

> Fax: (613) 225-9688

>







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

#50951 Sep 29, 2012

Hi, I do not want to hyjack this thread, but I am also trying to set up pulse guiding with Maxim DL but on an Atlas. I have gotten in the past very good guiding using EQMOD. I like this system because as the other writer I want to set limits to prevent bumpting my new camera. However, it does not seem to get set up properly and I have had to use the guider relays.



Are there any tricks to get this to work? I selected ASCOM DIRECT as the method of guiding. Should I use one of the Shioesting options... I am connecting via serial port.

--- In MaxImDL@yahoogroups.com, "Rod" rodsrun@...> wrote:

>

> Q1. Ah, so the bump signal follows the guiding configuration. Good.

>

> Q2. So if I set an activation level of +/- 25%, I should see a bump response when the AO8 gets below 25% or above 75%. Then depending on how many 100 ths of a sec I set for the bump signal, I should see the tracking return to somewhere between 25% and 75%...

>

> Q3. Um, ok - BTW, I wasn't complaining about the tracking, just wondering if I should spend time trying to get to a "zero" error on the graph. Since the RMS error is always on one side of the axis, which makes sense at least in RA, I wasn't sure if the offset was my guiding, my seeing limits, or an imcomplete AO8 setup. From what you wrote, it sounds like I-D control isn't required or implemented. Good enough.

>

> Rod

>

> --- In MaxImDL@yahoogroups.com, "Douglas B. George" dg@> wrote:

> >

> > On 2012-09-28 3:00 PM, Rod wrote:

> > > 1. If I use pulse guiding to guide my scope/mount, will my camera's AO8

> > > "bump" still get to the mount when it reaches its tilt-tip limits? The AO8

> > > setup in MaxImDL sort of implies that it operates only with the guider

> > > relays, but I may be misunderstanding.

> >

> > Yes. That is a relatively new feature (V5.13).

> >

> > > 2. Also, is there an easy way to confirm that a bump occurs based on the

> > > percentage error measured?

> >

> > You should see that the the AO Tilt position stays within the "Activation Level

> > +/-%" control at all times. If you want to deliberately move the telescope out

> > of the window then you should see it pull itself back again.

> >

> > > 3. Finally, can the AO8 produce a near-perfect guiding experience (assuming

> > > excellent mount alignment and PEC, local seeing quality notwithstanding), or

> > > does it always have an error offset? Control theory says a gain-only (or

> > > proportional-only, for theory purists) feedback loop will always have an

> > > error offset, which I believe applies to the AO8 response to error signals.

> > > Offset is usually reduced (or anticipated) by adding integral and derivative

> > > control to proportional control, but I don't think these enhanced control

> > > algorithms are part of the AO8 controller - or are they? If they are, the

> > > setup dialog boxes don't offer an opportunity to change I-D settings.

> >

> > AO is designed to operate fairly quickly, and there is no mechanical lag as you

> > will find with conventional telescope mount guiding. As a result any residual

> > is extremely tiny - less than the "noise" due to seeing etc.

> >

> > Doug

> >

> > --

> >

> > Doug George

> > dgeorge@

> >

> > Diffraction Limited

> > Makers of Cyanogen Imaging Products

> > www.cyanogen.com/

> >

> > 100 Craig Henry Dr., Suite 202

> > Ottawa, Ontario,

> > Canada, K2G 5W3

> >

> > Phone: (613) 225-2732

> > Fax: (613) 225-9688

> >

>



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

#50955 Sep 30, 2012

Rod --



Since you're a controls guy, here's the simplified story: > Q2. So if I set an activation level of . 25%, I should see a bump response when the AO8 gets below 25% or above 75%. Then depending on how many 100 ths of a sec I set for the bump signal, I should see the tracking return to somewhere between 25% and 75%...

Right. I do wish there were mount axis "bump lights" that would flash when the

outer loop servo is bumping. Also the help is potentially confusing. The

outer/bump loop is a "bang bang" non-linear servo. > Since , which makes sense at least in RA, I wasn't sure if the offset was my guiding, my seeing limits, or an imcomplete AO8 setup. From what you wrote, it sounds like I-D control isn't required or implemented. Good enough.

I'll add to Doug's info a bit, just FYI. The X and Y errors on the inner-loop

(deflecting element) are not proportionally resolved into the coordinates of the

outer loop. However, the inner loop is really fast (in the tens of milliseconds)

so if the guide star is bright and the guide sensor is sensitive, the exposures

can be in the tens of milliseconds and the inner loop can operate at over 10Hz.

It will respond basically "instantly" to a bump, which might last over 500 ms.

Of course the better (more expensive) the mount, the fewer bumps, and the better

the image quality. And so it goes with the guide star; the brighter it is, the

faster the inner loop can operate, and the better the adaptive optics seeing

removal.



For simplicity, I recommend testing at sky position angle of 0 (north-up images)

or 180 (south up images) so that the X bumping axis is the same as the X

deflector axis. Easier to interpret. So... If you are seeing bumping in only one

direction ("the RMS error is always on one side of the axis") then I would

suspect the mount is "dragging" the whole thing along in RA. I interpret your

observation to mean that the bumps are always in one direction. Even if you

increase the bumping time so as to push the deflector past 50% towards the

opposite bump threshold, eventually it's going to drift back to that same

threshold and bump. If you never see bumps in the opposite direction, then I

would say that your mount has a secular drift in Right Ascension. This could be

due to polar misalignment or a "DC" rate error in the RA drive.



If everything is healthy, the MaxIm DL AO servo design is more then "good

enough" (= engineering perfection).



-- Bob









[Non-text portions of this message have been removed]







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

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