Re: APPM with Dome Questions


Shane Ramotowski
 

When using DeviceHub, if you are connected to both DeviceHub's telescope and dome drivers, then when you issue a telescope slew, DeviceHub knows the target RA and DEC and can start a dome slew to the azimuth that the telescope will end at.

If you cannot connect to the DeviceHub telescope and/or dome, then DeviceHub uses a polling scheme to ask where the telescope is and then send the dome to the correct azimuth.  The polling period, unfortunately, seems to have a minimum of 5 seconds.  So, during an APPM run, APPM will tell the telescope to slew.  Sometime less than 5 seconds later, DeviceHub will retrieve the current RA/DEC from the telescope and start the dome slewing to that position.  For longer slews, the telescope may still be moving.  5 seconds after the dome arrives at the indicated position, it finds that the telescope has moved again (in acutality, the telescope never stopped moving, simply completing its original slew.  So DeviceHub issues a second slew to get to the final position.

So for the Passive settings in APPM, I have to use timeouts long enough to account for 2 polling periods plus 2 slew times to ensure that the dome is really in position before imaging starts.  I can't let APPM catch the dome in a stopped state between two dome slews. I hope that makes sense.

I ran a large model last night using DeviceHub.  I lost 5 points in total, which isn't that bad.  Three of them were because of the bodaciously bright almost full moon and the other two were bad luck in the timing of long slews from the end of on string of points to the start of the next.  I was concerned that DeviceHub would not position properly for the counterweight-up points, but it did so correctly.  The model run took a lot longer than I expected since I was waiting 40 seconds (usually unnecessarily) for each point to give the dome time to get into position in case it was a long slew.

I really don't know if supporting DeviceHub's telescope driver will work.   I'm assuming that if the SideOfPier property is set correctly when the slew is issued, DeviceHub will go to the correct location even if it ends up counterweight-up.  But I haven't tested that and I really don't know how--well I guess I could write a little program to do that.  I will do so if you are considering adding such support and the results of that test would help.

As far as the APPM specific commands, could you allow a connection to DeviceHub to send the info necessary for the slew and then send the "special sauce" directly to the AP V2 driver?

- Shane

On 2/28/2021 8:10 AM, Ray Gralak wrote:
Hi Shane,

One more question, Ray: In APPM, is there a way to connect to the
DeviceHub telescope or is it locked to the AP ASCOM driver?
You would not want to connect APPM to the DeviceHub telescope. APPM uses some commands that the DeviceHub would not be able to pass through to the AP V2 driver.

That said, what is the use case for this?

-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf Of Shane Ramotowski
Sent: Sunday, February 28, 2021 3:52 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM with Dome Questions

One more question, Ray: In APPM, is there a way to connect to the
DeviceHub telescope or is it locked to the AP ASCOM driver?

Thanks - Shane

On 2/27/2021 8:09 PM, Shane Ramotowski wrote:
Ahh, I think it understand. I use DeviceHub to do the slaving, but
still connect to the DeviceHub dome with APPM. I will try that and
see what happens.

Thanks - Shane

On 2/27/2021 5:27 PM, Ray Gralak wrote:
Shane,

OK. I'll wait for the ASCOM folks to add the APIs that you want before
I try to use "Active" mode again.
I think you misunderstood. ASCOM is not planning to add those APIs.

For the second part of my reply, I guess I did not make it clear enough
that I was using DeviceHub not allowing APPM to connect to the dome,
since APPM cannot send the dome to the correct position an DeviceHub
can. _I am simply trying to understand what I'm doing wrong when
setting
up "Passive" parameters in the "Dome Settings" box.__
In passive mode, APPM only needs to poll the Dome's slewing state to
work. If you can't do that then APPM won't be able to tell when the
dome is done slewing.

If Device Hub won't let APPM connect to the Dome driver you can use
the older ASCOM POTH application, which allows multiple applications
to connect to an ASCOM driver.

So can you please tell me what settings I can use for "Delay before
staring dome slew check" and "Settle time after dome stops slewing" to
ensure that I end up with at least 30 seconds between the end of the
telescope slew and the start of imaging?
First, the two parameters only have meaning when you are connected to
a Dome driver. If you are not connected they won't be used.

"Delay before starting dome slew checking"

The number of seconds before APPM will start checking the dome
driver's slew state. This delay allows the dome driver time to
initiate dome movement that will be reflected by reading "True" from
the Dome driver's "Slewing" property. If the value is False when
read, then dome slewing will be considered completed.

"Settle time after dome stops slewing"

The number of seconds to delay slew completion after the Dome
driver's "Slewing" property reads "False".

-Ray





-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On Behalf
Of Shane Ramotowski
Sent: Saturday, February 27, 2021 2:16 PM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM with Dome Questions

OK. I'll wait for the ASCOM folks to add the APIs that you want before
I try to use "Active" mode again.

For the second part of my reply, I guess I did not make it clear enough
that I was using DeviceHub not allowing APPM to connect to the dome,
since APPM cannot send the dome to the correct position an DeviceHub
can. _I am simply trying to understand what I'm doing wrong when
setting
up "Passive" parameters in the "Dome Settings" box.__
_
So can you please tell me what settings I can use for "Delay before
staring dome slew check" and "Settle time after dome stops slewing" to
ensure that I end up with at least 30 seconds between the end of the
telescope slew and the start of imaging? Assume that I did not change
the default telescope slew of 900X. The log entry I posted is from a
session that used DeviceHub with "Delay before starting dome slew
check"
set to 30 seconds and "Settle time after dome stops slewing" set to 10
seconds. You can see that the time between the start of the telescope
slew and the the start of imaging is about 12 seconds. So, clearly, I
am not understanding how these two parameters work.

Thank you - Shane


On 2/27/2021 2:16 PM, Ray Gralak wrote:
Shane,

APPM's "Active mode" sends the telescope's destination
coordinates to the dome ASCOM driver. In this
case, it is the responsibility of the driver or intermediate
application to translate the RA/Dec coordinates to the
appropriate dome position.
Hmmm, so how does that work? What part of the ASCOM dome API
would that
be?
<https://ascom-standards.org/Help/Developer/html/T_ASCOM_DriverAccess_Dome.htm>
Again, APPM does not do dome calculations (nor is there a plan for
it to do that). IMO, the ASCOM Dome API
is missing a way to send scope Ra/Dec/pierside to the Dome driver.
ASCOM makes such a big deal about
abstracting away telescope applications from the hardware, but the
incomplete Dome API pretty much forces
the use of a third-party application.
I think each dome manufacturer should each create their own drivers
that should take into account the
telescope and mount dimensions. It would have made life much easier
for each application developer that has
had to implement dome coordinate transformations in their applications.
> In "Passive mode" APPM waits for the dome to finish slewing by
polling the dome's driver (or intermediate application).

That is exactly what I expected. What should I set the two passive
values to to ensure that the imaging doesn't start for, say, 1 minute
after the scope stops slewing? Or am I not understanding the
meaning of
those two fields and their documentation in the manual?
To use passive mode the dome driver or application must be slaved
to the telescope. In this case APPM just
polls the slewing state until dome slewing stops. If slaving can't
be done, there is an option in APPM to pause
after each point, which would allow you to position the dome opening.
The total time between the start of the point's processing and
start of
it's imaging is about 12 seconds. Shouldn't that be at least 40
seconds
to allow the dome delay and settle times to happen? What am I
misunderstanding?.
There are no dome messages in your post so I can't tell you what
happened. Are you sure that you had
APPM connected to your Dome application?
BTW, the problem might be that you are using the ASCOM Device Hub
instead of the older POTH dome
handling. The behavior is dome behavior is different in Device Hub,
and it sounds like the author left out some
old functionality.
-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On
Behalf Of Shane Ramotowski
Sent: Saturday, February 27, 2021 11:32 AM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] APPM with Dome Questions

Ray, thanks for the reply, but I don't really see anything that
will help.


APPM's "Active mode" sends the telescope's destination
coordinates to the dome ASCOM driver. In this
case, it is the responsibility of the driver or intermediate
application to translate the RA/Dec coordinates to the
appropriate dome position.
Hmmm, so how does that work? What part of the ASCOM dome API
would that
be?
<https://ascom-standards.org/Help/Developer/html/T_ASCOM_DriverAccess_Dome.htm>


I am definitely returning false fort the CanSlave and Slaved
properties,
since there is no hardware-based slaving capability in the dome
controller. And I don't see way to "send the telescope's destination
coordinates" to the ASCOM driver.

The call that the ASCOM driver receives from APPM is
"SlewToAzimuth()"
but I think you are basing that azimuth on the center of the dome
rather
than the position of the telescope, which can be significantly
different--consider the initial slew to zenith in which the RA axis
rotates west and level and the DEC axis point strait up. The
telescope
is several inches West of the center of the mount. From what I
understand, if you are going to use SlewToAzimuth(), you need to
consider this.

If you look at ASCOM's DeviceHub, written by Rick and Pete, who
answer
most of the dome-related questions posted to the ASCOM Driver and
Application Development Support Forum, it has a whole section for
dome
and mount geometry in it's setup section to allow slaving.

> In "Passive mode" APPM waits for the dome to finish slewing by
polling the dome's driver (or intermediate application).

That is exactly what I expected. What should I set the two passive
values to to ensure that the imaging doesn't start for, say, 1 minute
after the scope stops slewing? Or am I not understanding the
meaning of
those two fields and their documentation in the manual?

Here is a log excerpt from one of my attempts last night. Under
Settings->Dome Settings, Passive is selected, "Delay before starting
dome slew checking" is set to 30 seconds and "Settle time after dome
stops slewing" is set to 10 seconds.

0467221 2021-02-26 20:46:21.965: Info, State Machine,
Entering
State=SlewNext
0467222 2021-02-26 20:46:22.103: Info, SlewNext, Starting
Slew to Point 3 RA: 08h 30m 06.56s, HA: -01h 20m 00.00s, Dec: -10°
00'
00.0")
0467223 2021-02-26 20:46:22.105: Info, Slew Next,
East=True,
Dec=-10, HA=-1.33333333333333, MerDelay=-0.25, MerOffset=0
0467224 2021-02-26 20:46:22.105: Info, Meridian, Setting
Meridian Delay to -0.25 (-00h 15m 00.00s)
0467225 2021-02-26 20:46:22.144: Info, SlewAsync, Begin
UnSAFE Slew to RA/Dec: 8.501824 / -10.000000 ( 08h 30m 06.56s /
-10°
00' 00.0")
0467226 2021-02-26 20:46:22.215: Info, State Machine,
Entering
State=PreSlewing
0467227 2021-02-26 20:46:25.482: Info, State Machine,
Entering
State=Slewing
0467228 2021-02-26 20:46:30.722: Info, State Machine,
Entering
State=StartSettle
0467229 2021-02-26 20:46:30.741: Info, StartSettle, Starting
Settle wait time
0467230 2021-02-26 20:46:30.970: Info, State Machine,
Entering
State=WaitSettle
0467231 2021-02-26 20:46:32.974: Info, WaitSettle, Settling
Time Complete
0467232 2021-02-26 20:46:33.022: Info, WaitSettle, Finished
Slew to Point 3
0467233 2021-02-26 20:46:33.022: Info, WaitSettle, RA/Dec:
8.501833 / -10.000000 ( 08h 30m 06.60s / -10° 00' 00.0")
0467234 2021-02-26 20:46:33.022: Info, WaitSettle, HA/Dec:
-1.329942 / -10.000000 (-01h 19m 47.79s / -10° 00' 00.0")
0467235 2021-02-26 20:46:33.226: Info, State Machine,
Entering
State=StartImage
0467236 2021-02-26 20:46:33.260: Info, State Machine,
Starting
Exposure, Duration=5 LST=7.17191416666667 (07h 10m 18.89s)
0467237 2021-02-26 20:46:33.260: Info, State Machine, LST mid
image=7.1727475 (07h 10m 21.89s)
0467238 2021-02-26 20:46:33.260: Info, StartTakeImage, Sequence
Generator Pro: Binning=1, Subframe Type: 0, Duration=5,
IsDarkFrame=False

The total time between the start of the point's processing and
start of
it's imaging is about 12 seconds. Shouldn't that be at least 40
seconds
to allow the dome delay and settle times to happen? What am I
misunderstanding?

The dome controller and ASCOM driver do handle Slewing property
correctly, returning TRUE from the time that the slewToAzimuth() is
issued until the dome is stopped at the requested position, then
returning FALSE until the next slew is issued Sometimes,
DeviceHub or
SGP will end up issuing more than one slew to get the dome to the
final
position. The slaving software notices that the telescope has
moved and
starts an azimuth slew to where is while the telescope is still
moving.
Once the dome arrives at that position, another azimuth slew to
get to
where the telescope finally stopped. Last night, the longest time
that
such multiple slews took was about 20 seconds. That is why I set the
"Delay before starting dome slew checking" to 30 seconds. I though
that
would make APPM not even check for a dome slew until the dome was in
place and had stopped moving. Instead, it seemed to have made no
difference: the imaging started almost as soon as the telescope
arrived
at its position.

Thanks - Shane


On 2/27/2021 10:06 AM, Ray Gralak wrote:
Hi Shane,

APPM's "Active mode" sends the telescope's destination
coordinates to the dome ASCOM driver. In this
case, it is the responsibility of the driver or intermediate
application to translate the RA/Dec coordinates to
the
appropriate dome position.
In "Passive mode" APPM waits for the dome to finish slewing by
polling the dome's driver (or intermediate
application).
So, it seems that your dome's ascom driver (or intermediate
application) may not be correctly indicating
when
the dome is slewing.
-Ray

-----Original Message-----
From: main@ap-gto.groups.io [mailto:main@ap-gto.groups.io] On
Behalf Of Shane Ramotowski
Sent: Friday, February 26, 2021 9:19 PM
To: main@ap-gto.groups.io
Subject: [ap-gto] APPM with Dome Questions

Hi gang,

I am the proud owner of a brand new (well 2 weeks) Mach2 and
tried to do
my first APPM model tonight. I'm having problems with the dome
control
and hope someone can point in the right direction.

My observatory is a ClearSkys (I don't think they are around
anymore) 8
ft dome. The pier is centered and the bottom of the mount is
just about
even with the bottom of the dome. The rotation control is
homemade (my
COVID quarantine project) and works very will with both ASCOM
DeviceHub
and SGP. I'm using SGP for image capture and plate solve.

I can't figure out how to set up APPM to control the dome
properly. If I
use "Active" in the "AP Point Mapper - Dome Settings" panel,
APPM seems
to do it's calculations based on the center of the dome instead
of where
the telescope is. This is not surprising since I can't seem to
find any
where in the program or the documentation to set the mount and
telescope
geometry. I suppose, since APCC knows that I'm using a Mach2, it
_could_ already know roughly how far from the center of the
RA/DEC axes
intersection the bottom of my telescope is. But I don't see
anyway that
it could know where the center of my OTA is. I really don't
think it's
using the parameters from 3D view; I haven't set that up, but the
default is a much larger diameter telescope than mine.

Anyway, when I use "Active", I end up imaging across the
edge of the
slot because the slot is positioned for the center of the mount
instead
of where the telescope is. Most of the images ended up plate
solving
anyway, but since the stars are all diffraction spikes, I don't
know
the quality of those solutions. The initial slew to meridian is
probably the worst because the dome is in the exact opposite
position
than it should be and my slot just goes barely past vertical. I
checked the ASCOM logs on the PC and my ASCOM debug screen on
the dome
controller. The dome _is_ slewing to the position that APPM
requests;
that position just doesn't seem to be based on where the
telescope is.

Can anyone tell me what I'm missing? Where do I enter the
offsets for
the mount/scope geometer so that APPM can got to the correct
position?

So, after trying "Active", I swiched to "Passive" and tried with
two
different dome slaving programs: ASCOM Device Hub and SGP. For
"Delay
before starting dome slew checking", I used values ranging from
5 to 30
seconds. For "Settle time after dome stops slewing", I used
values from
1 to 20 seconds. I didn't see any effect when changing these
settings;
in all cases the the telescope slewed to the next position and
started
imaging a few seconds after it got there--many seconds before
the dome
was in position. None of those images solved since the dome was
in the
way. Again, I can't figure out what I'm doing wrong--it's like
I'm not
even changing the values. Is there another setting that I'm
missing
that enables these timeouts?

I'm sure someone else has successfully done an APPM run from an
automated dome! What am I missing?

Thanks - Shane

--
Shane Ramotowski
kor@cotse.net
https://www.kor-astro.net






--
Shane Ramotowski
kor@cotse.net
https://www.kor-astro.net







--
Shane Ramotowski
kor@cotse.net
https://www.kor-astro.net








--
Shane Ramotowski
kor@cotse.net
https://www.kor-astro.net








--
Shane Ramotowski
kor@cotse.net
https://www.kor-astro.net

Join main@ap-gto.groups.io to automatically receive all group messages.