Guiding Issues only in one directions

[Ted Terranova]

I posted this on Cloudy Nights and have been getting some great ideas and feedback back but so far no solution so I thought I'd post this here and maybe someone would have an idea of what is wrong.

I'm having issues with PHD2 when it starts guiding/calibrating for guiding. It almost seems like the mount's standard tracking turns off once PHD2 kicks in. I made a video last night showing how I can nudge the telescope around with manual controls no problem as the mount continues to track.

 

Video showing the issue -  https://youtu.be/wKhtKCwuzXs

 

Over a year ago I bought the Orion Magnificent Mini AutoGuider Package. I mounted it on my telescope, installed PHD2 but have not been able to track at all. I have an Explore Scientific 80 ED Triplet, Celestron CGEM Mount, and the Celestron Star Sense. I have the wifi dongle which I use to connect to my IPAD to the mount. I run Skyportal on the Ipad, connect and align. I run Backyard EOS connected to my Canon 80D. I was running PHD2 with my laptop connected to the Starshoot Autoguider and connecting the PC to the hand controller(i have a usb type controller). I am able to effect the mount with manual guiding so I know that the connection is working.

 

My scope will be tracking well but I can only get 30 second images so I start the guiding in PHD2 and it starts to calibrate. First is moves West and seems to get very far from where it was, then north and south seem ok, then it moves East and seems to keep going West. It's as if in one direction it can't move enough to keep up with the star. And so the tracking fails or the calibration says there's some problem. 

 

Have checked balance, tried different balances, sidereal is on, connections seem good, mount responds to manual commands, I've even turned off the wifi to let the hand controller track but the same issue occurs once PHD2 kicks in.

 

It was suggested that I run a star cross test. I've attached the results. They look like question marks sort of.

I've also attached some logs.

Why can I nudge the scope around no problem with the virtual hand controller through ascom. But when PHD does it, this seem to get out of control? Could I have set something wrong somewhere? I might try uninstalling and reinstalling PHD. Also maybe just try another laptop.

Thanks for any help.

[peter wolsley]

Ted,
I believe your issue is with pulse guiding.  I own an older CGEM...I don't use Star Sense.

When you test your mount by using the virtual hand controller you are sending slew commands to your CGEM.  These slew commands look like they work ok which is why you can align/calibrate your mount.  When you fire up PHD2 it will control your mount by using a totally different type of command called pulse guiding.  I would assume that you do not have a cable plugged into the Auto Guide port on your CGEM.  Your configuration appears to be using ASCOM pulse guiding which is great and works very well for me.  RA axis pulse guiding is not working for you.

Your calibration data tells an interesting story.  When PHD2 performs a calibration is always follows the same steps.

1)Send RA axis WEST guide pulses...in your case, PHD2 sends out 6 guide pulses which is ok.

2)Send RA axis EAST guide pulses.  These guide pulse typically cause the mount's RA axis to reverse it's steps and return to where it started.

3)Send DEC BACKLASH NORTH guide pulses.  These are large guide pulses that are intended to move the DEC axis thru it's backlash so that the DEC mechanism has taken up all of it's backlash in the North direction.

4)Send DEC NORTH guide pulses...in your case PHD2 sends out 7 guide pulses.

5)Send DEC SOUTH guide pulses...The intend is for your mount to reverse direction and return to where is started.  In practise this rarely happens because of the backlash in the DEC mechanism.

Your mount looks like it is able to execute WEST, NORTH and SOUTH guide pulses.  EAST guide pulses seem to act like WEST guide pulses that are twice as fast.  I suspect that when PHD2 tries to issue an EAST guide pulse that is when you see what looks like your tracking being turned off.  You can see this for yourself if, in PHD2, you select the Tools drop-down menu and select the Manual Guide Option.  Every time you click on the North/South/West/East buttons PHD2 will send a pulse guide command to your CGEM with a duration given by the "Guide Pulse Duration(ms)" control located below the North/South/East/West buttons.

All of the movements you are going to perform via these pulse guiding commands are very small compared to typical slew commands.  Because you are using 85% autoguiding gains your pulse guiding commands will look like slew commands at Rate 2 = ~15a-s/sec.  You should set your slew rate to 2 and press the slew buttons to get familiar with how slow these movements are.

When you press the West pulse guide button you should see your guide star move West at roughly 15a-s/sec.  When you press the East pulse guide button, a properly operating mount will move the mount in the opposite direction (East).  I think you will find that your mount moves West at roughly double the speed.

When you press the North pulse guide button you should see your guide star move North at roughly 15a-s/sec.  You may not see any North movement at all until you have pressed the North guide pulse button several times(be sure to wait for the GuidePulse Duration to finish between each pressing of the guide pulse button).  The lack of movement is because of your DEC backlash which I estimate from your PHD2 calibrate as being roughly 100 arc seconds...I hope I am wrong.  When you press the South pulse guide button you should see your guide star move South at roughly 15a-s/sec.  You may need to press the South guide pulse button several times until you move your DEC axis thru it's backlash.

Your big issue is to determine how your mount's RA axis pulse guiding is behaving.  I think you should also verify that the North/South pulse guiding is working correctly.  You may find that backlash is huge but you really want to confirm that you can move the DEC axis in both directions using only pulse guiding.

Once you perform these tests report back to us so we can offer more help.  I am not an expert on PHD2.  I have a lot of experience with pulse guiding my CGEM and I like to study guiding logs.  I suspect I made a few assumptions that are not totally correct but I believe the Manual Guide tool in PHD2 is what you should work with right now to learn more.

Peter

[Ted Terranova]

Hi Peter,

Thanks for all the information. That all makes sense.

I did do the star cross test, which is run by PHD2. I would assume that PHD is using pulses to perform this pattern? Would that be similar to me using the manual guiding buttons in PHD? I would think that's a good indicator of how my mount is responding to the pulses. Can you see the attached images on my post? The stars look like ? marks, not crosses.

link to test description -  https://openphdguiding.org/man-dev/Trouble_shooting.htm#Star_cross_test

For the record I have used the manual guiding buttons in PHD and see movement but I didn't really pay attention when doing this. Also that was when I was guiding through the guide scope with ST4. Once I tried ascom, hoping that would fix the problem, the virtual hand controller buttons appeared and they were much more responsive, as you explained.

Would an ST4 guide pulse be the same as the ASCOM pulse? I could try going back to the ST4. Also my hand controller seems to be newer with a USB connector. I bought it used so not sure if this is standard. Also it being used, maybe the previous owner set something up with the pulses that is flipping them? Just shooting in the dark but maybe that info will help?

I'll try firing up PHD2 and using the manual guiding in PHD2. I believe I could even just do that during the day if I focus on some far away object, right? Weather around here has been bad lately.

If I haven't answered anything that would help resolve this please ask away. I'd love to narrow this down and figure out the cause of the issue. Thanks!

[bw_msgboard]

Hi Ted, sorry you’re having trouble.  Have to say, this is a pretty unusual problem.  What we see is that the mount won’t reverse direction correctly going from west to east.  Let’s look at how your calibrations typically go, looking only at the west and east measurements:

 

Direction, Step#, Dist

West,0,  0.000

West,1, 0.651

West,2, 6.046

West,3,-11.730

West,4,-17.233

West,5, 22.920

West,6, 28.552

 

East,5, 28.552

East,4, 41.071

East,3, 50.381

East,2, 59.785

East,1, 74.368

East,0, 79.191

 

The third column shows the total distance the star has moved from its starting point.  Notice how that distance keeps increasing even though the guide commands switched direction from going west to going east. 

 

Here’s an example of how it should look (different image scale but the basic picture should be similar):

 

West,0,  0.000

West,1,  2.262

West,2,  4.433

West,3,  7.041

West,4,  9.260

West,5, 11.498

West,6, 14.079

West,7, 16.125

West,8, 18.672

West,9,  20.764

West,10, 23.021

West,11, 24.911

West,12, 27.577

 

East,12,  27.577

East,11,  25.285

East,10,  23.115

East,9,    20.708

East,8,    18.490

East,7,    15.997

East,6,    13.709

East,5,    11.458

East,4,      9.363

East,3,      7.003

East,2,      4.860

East,1,      2.153

East,0,      0.417

 

You can see how the East guide pulses push the star back close to its starting position.

 

As Peter pointed out, there’s a big difference between pulse-guide commands and the slewing operations that are done by either real or virtual hand-controllers.  The problem you see here with calibration – that guide-east commands don’t work – also explains the peculiar star-cross test results you got.  The easiest way to generate manual pulse-guide commands is to use the Manual Guide tool in PHD2.

 

I don’t know much about Celestron mounts or the many connection options you’re using.  But my first guess is that this isn’t likely to be a mechanical problem with the mount.  I’d suggest reducing to the simplest possible test arrangement:  no wi-fi, no planetarium programs, no other applications attached to the mount.  Make sure that nothing is plugged into the guide port on the mount and nothing plugged into any of ‘aux’ mount ports if you have any of those.  Then initialize the mount however Celestron wants you to and immediately try using the Manual Guide tool to see if you can get things going.  

 

Hope you can track it down quickly,

 

Bruce

 

 

---

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[peter wolsley]

Ted,
I looked at your star cross test photos.  They seemed to indicate a similar problem that you are having with PHD2 calibration.  I have never performed a star cross test.  After this I will perform one to confirm what PHD2 is doing and what my mount does.

Ideally, you should be using ASCOM pulse guiding.  Once you have confirmed what the problems are with ASCOM pulse guiding.  You could re-configure your mount to use ST4 pulse guiding and repeat the test.  The ST4 pulse guiding uses a completely different hardware path to control your mount.  You may find that ST4 pulse guiding works fine which would get you going for now. 
It's worth a try researching the USB hand controller...maybe there is an issue with an old mount/new handcontroller set-up..I really don't know anything about the newer USB handcontroller.  Do you have the original hand controller?...something else worth trying.

You may also want to try running the mount without the wifi module.  Bruce's suggestion of simplifying your set-up to try and identify the component that is messing things up is definitely in your future.  If you find that the problem doesn't go away then you should look into updating the handcontroller firmware and the motor control firmware.  I use Celestron programs called HCUpdate and MCUpdate to do this.  Even if your firmware is up to date you could try reloading it...maybe it got corrupted along the way.  My gut feeling would be to update/reload the firmware in the handcontrolller and motor controllers.  You should also make sure you are using the latest Celestron ASCOM driver and the latest ASCOM platform.  Follow Bruce's lead...start with the simplest set-up and work your way up.

Peter

[Ted Terranova]

Thanks guys for all the advice. I REALLY appreciate it. I was originally using ST4 through the guide camera, and that's where the problem started but I can try ST4 again as well.

Here are the steps in order that I'll try them. 

1 - unplug everything from the mount except the hand controller and USB corder - try manual control through PHD2

2 - try a different cable and try again

3 - try another laptop. I installed everything on another laptop. I thought maybe the install or driver might be bad on the laptop. Easy to try.

4 - try plugging in the guide camera to the guider port and the USB cable to the guide camera. Unplug the USB going to the hand controller. Try manual guiding with PHD2 using ST4.

IF it still shows the same symptons, I'll try updating the hand controller and motor controllers and go back to step 1.

Sound like a good plan?

Is this what I need for the update? https://www.celestron.com/pages/drivers-and-software   ---> Motor Control Firmware Updates v.2.2.4  AND  ---> Hand Control Firmware Updates v.1.1.12 I believe I can just run on my laptop and it'll update through the USB connection to the hand controller?

The second laptop should have latest ascom drivers and all. Not sure about my current laptop, I'll check. And update if needed.

Thanks again and let me know if there's anything I left out.

[peter wolsley]

Ted,

So you believe that you first experienced problems with using ST4 for pulse guiding.  It's still worth a try if only to prove the problem is the same regardless.  The firmware updater programs sound like the right ones...they are from the Celestron site so that it good.  I checked my versions and they both are newer.  The versions on the Celestron site are public release versions which are considered stable.   The motor control firmware version to load into your CGEM is called CGEM_Series_0651.  I have a old handcontroller so it's firmware version is called GEM 4.21.  Your USB handcontroller is newer so your best bet is to look at your handcontroller as it powers up and make note of any firmware versions that appear on the display.  When you use HCUpdate you will see a list of possible choices.  You want to see if you can find the version you are using.  Once you find it then you can look to see if there is a newer version of that firmware paying attention to the first letters in the firmware name.  Those letters indicate the specific handcontroller model.

The next few nights are perfect for this kind of experimenting.  You want to get to the bottom of this before the new moon arrives.

Peter

[peter wolsley]

Ted,

FYI

The MCUpdate and HCUpdate programs both need access to the internet so that they can download the latest database of available firmware. 

Peter

[JohnS]

As an aside - not sure whether you know it but the latest Celestron utility for updating nexstar hand controllers and mounts is the Celestron Firmware Manager (CFM) which can be found here https://software.celestron.com/updates/CFM/CFM/.  The last time I looked the MCU and HCU updaters were well out of date - 2008 I think.  This may not matter and I know that CFM works with the CGX - not sure it works with CGEM.

Hope this helps and I haven't confused things.

John S

[peter wolsley]

John,

Thanks for your info.  I guess my CGEM is pretty old.  I have heard about CFM and it might be that CFM is the way to go for Ted.  I probably should look into it myself.  I am just assuming that because my CGEM came with MCUpdate and HCUpdate, that's the only way to perform an update on a CGEM.

Peter

[Ted Terranova]

Oh thanks for the info, John, I'll try the CFM first. You'd think Celestron would spend an hour just redoing their website to make things clearer :) The support download section isn't that easy to navigate. Should I just update things first? I'm really tempted to, but don't want to throw more factors in the mix.

Thanks again guys.

[JohnS]

I'd suggest maybe use the method you're comfortable with and try CFM during the daylight hours.  It may not add anything - or indeed work with the CGEM mount - but I see you've got the latest Nexstar+ handset with the USB connection and it definitely works with that.  I've looked at the interface and it definitely lists the CGEM models.  The lastest software version seems to be 2.3.7 and that was released in May this year - but that might be the CGX version.  I know somewhere along the line a 9th slew speed was added to Nexstar - but you may already have that too.

Good luck - hope it helps resolve the problem

[Ted Terranova]

I updated the handset. But apparently the cgem needs to update using the old skool method. I tried joining the CGEM beta site but still have not received a confirmation email :) So when I get that, if I get that, I'll get the motor control stuff off their site and try and update that.

But I did set up this evening. Connected nothing but the hand controller and ran PHD2. I was able to move the mount with the manual guiding in all directions. It was REALLY slow, and hard to see. I had to hit the directional button many times. But I had my telescope pointed at a transformer on top of a telephone pole so maybe it being so close had something to do with it. The weather is bad here. No real stars so I figured pointing at anything far away should work, right?

The mount did seem to be more responsive or faster in the North South direction. I couldn't tell if it was faster East or West. I did capture a movie of the process...a long boring movie LOL It's about 8 or so minutes.

https://youtu.be/aHI6WrSWEt4

I had the scope pointed at a dot on the side of the transformer but that's hard to see, but the crosshairs move about relative to the parts of the transformer and so that's how I detected the movement.

If you have time, and some coffee to keep you awake, take a look and let me know what you think. So I'm pretty positive I moved it in the 4 directions. What does this mean then? Maybe the hand controller update did something? Or is PHD2 just sending weird pulses for some reason. Weather looks to continue to be bad over the next few days but if there's a break I'll try and get out and try guiding.

Thanks again for all the help guys. Cheers.

[Ted Terranova]

I realized a sped up version might be useful and less tedious. Here's a link to it at 10x.

https://youtu.be/o_0aYM11Is8

I also highlight the buttons that are being pushed since it's a bit hard to see at times.

[peter wolsley]

Ted,

I looked at your video...the x10 version.  North/South look good.  East/West seem a lot slower.  You should look at your autoguiding gains.  I know when I load firmware into my CGEM there are values that are reset to default values.  You might find your autoguide gains, backlash, filter limits, custom rate 9 etc. values have changed.  The good news is that it appears that you are getting East and West movements.  All four movement should be the same speed.  Hopefully you can get out under the stars soon.

Peter

[peter wolsley]

Ted,

Another issue that is clouding the East/West problem in your video is the fact that you had to turn off tracking.  This makes the backlash in the RA axis more visible.  You may need to apply a lot of RA pulse guiding to verify that you have control and that the rate of control is the same as for your DEC axis.

Peter

[Ted Terranova]

Thanks Peter. "You may need to apply a lot of RA pulse guiding" I don't know what this means or how to do that. When you say apply a lot, does that mean turn up a setting in PHD or the hand controller or mount? 

I'm finding it confusing that there are different methods of rotating the mount and also different places where it decides how much to rotate. Like I know I can set a guide rate, I think it's .85 now...do I change that? 

Wish I could take a week off and just learn all this. Thanks.

[peter wolsley]

Ted,
There are only a few parameters to know about with regards to pulse guiding. The first is the Autoguiding Rates.  There are two of these(RA and DEC) and they are programmed using your handcontroller.  You have mentioned that your Autoguiding Rates are set to 0.85.  This rate is related to the tracking speed of your mount.  Sidereal tracking speed is roughly 15 arc-seconds per second.  0.85 translates to roughly 12.8 arc-seconds per second.  What all this means is that when your RA axis is tracking the night sky it is physically rotating the RA axis of your mount at a speed of roughly 15 arc-seconds per second towards the West.  When an RA axis pulse guiding command is sent from PHD2 to your mount the RA axis will momentarily turn faster(West pulse guiding command = 15 + 12.8 = 27.8 arc-seconds per second) or slower (East pulse guiding command = 15 - 12.8 = 2.2 arc-seconds per second).

The pulse guiding commands are issued by PHD2 to your mount...typically via your mounts SSCOM driver. The commands always contain two parameters.  The first is DIRECTION which is NORTH, SOUTH, EAST and WEST.  The second parameter is DURATION which is scaled in milliseconds (1,000 = 1 second).  The normal method for PHD2 to send a pulse guiding command is for it to send the command and then wait until the mount reports back that the pulse guiding command has finished.  When you press one of the buttons on the manual guide tool PHD2 will using the button name to decide the DIRECTION parameter and the "Guide Pulse Duration (ms)" value as the DURATION parameter.  If you look closely at these button you will see that they change color while the pulse command is active.  Once a command has finished you can send another command.  It's this waiting for the command to finished before sending another command that I referred to when I said "You may need to apply a lot of RA pulse guiding".

If you think about pulse guiding as DIRECTION and DURATION you will find that a pulse guiding command is actually a position movement command. For your mount, a WEST pulse guiding command with a duration of 1,000mS = 1 second will cause your RA position to move 12.8 arc-seconds to the WEST.  If 12.8 arc-seconds is not enough to see the camera view move in your x10 transformer video you will need to repeat the command with every repeat moving the RA position an additional 12.8 arc-seconds.  Of course, you could also just change the duration of the command from 1,000 to 2,000 which will double the movement to 25.6 arc-seconds per command.

You will need big pulse guiding commands to compensate for backlash and see movements for your video.  When your mount is guiding on a star with PHD2 the pulse guiding commands will typically be very small with small durations of 20 to 2000 ms.  PHD2 needs to move your mount fractions of an arc-second at times and 10s of arc-seconds when trying to correct for DEC backlash.  BTW, we never talked about this but you need to ensure that the RA and DEC backlash compensation values that you set via your handcontroller are all set to zero.  There are four values...two for RA and two for DEC. The values you set via your handcontroller are meant for visual observing.  When you switch to pulse guiding via PHD2 these backlash values interfere with PHD2's ability to control your mount.  PHD2 has it's own backlash compensation features which are designed for pulse guiding.

So...your autoguiding rate of 0.85 is fine...don't change it.  Now you are an expert in how pulse guiding works and you now understand that GOTOs and SLEWs are for BIG movements...pulse guiding is for SMALL movements.

Hope this helps

Peter

[Ted Terranova]

Peter, that's a great explanation. Thanks for writing that all out. So I decided to update the motor drivers. BUT with the CGEM you can't do that with STAR SENSE hand controller with the USB? Why? I don't know, so I needed to use the original controller and that odd cable converter USB to 6 pin setup. I did't have those cables so I ordered them. They came in today and I was able to update the drivers from 6.17 to 6.51. I was very excited that this worked.

As soon as that was done I wanted to do a test. I went back to the USB cable because PHD2 didn't see the scope through the phone line adapter wire, and I didn't care why :) I put some tinfoil over the guide scope, made a little hole in the foil and had a nice pinhole camera. I pointed it at my car and proceeded to manual guide with it. I made another video at 5X. This time, to my eye, the movements look equal in speed and reaction time. I have not tried tracking stars but this is really exciting. Here's a link to the video. I'd like to see if you guys agree about the movement. I REALLY hope this has done it. Thanks again for all the help thus far.

https://youtu.be/wBd20MW0xCs

[peter wolsley]

Ted,

A pinhole camera...that's a clever trick.  I do see movement in all directions and they look pretty even.  Fingers crossed that your pulse guiding issues are gone.  I don't have a star sense hand controller but I can only imagine that mixing old with new stuff creates issues.  I suspect that Celestron has only so many manhours available for software development so creating software utilities that work with every possible combination of hardware is a low priority.  The main improvement with V6.51 was fixing a bug in the PEC logic.  You may have had a corrupted firmware in your mount. There are two motor controllers in your CGEM. One controls RA and the other controls DEC so maybe the RA controller's firmware was corrupted. 

Regardless of the outcome please post a guidelog of your evening so we can see your calibration and guiding.  You should also run the PHD2 guiding assistant.  It can give you immediate advice to improve your guiding.

Peter

[Ted Terranova]

You're right. The folks at celestron and all the devs who work on this stuff do a great job. Especially those who do it for free and just have a passion for it. It seems like it's one man doing all the motor control drivers for the mounts at celestron! 

I think the updated drivers fixed it! I don't want to get too optimistic but it calibrated and then guided. I tried imaging for an hour or so and it stayed on the star. I did get some warnings and the guiding looked less than ideal. Lots of ups and downs...but it was GUIDING! I'm attaching the logs and the debug file so you can take a look and let me know what you think. I am SO excited that it seems to have worked. I know I may have a lot to do to get better guiding but knowing it might be possible is really exciting. Thank again to you and everyone else who helped me get here.

https://www.dropbox.com/sh/td8m45etu68f45s/AABfmSvsNA4rgZc6KmF7qQ0ua?dl=0

here's a link to the files. They were too big to attach.