SW EQ5 guide erratic behaviour: what to inspect? (guidelog attached!)

[Dave_D]

oops...

[rofus]

Well at least you get some time guided...I seem to not get it at all!

Tried now installing Prolific usb-serial drivers in Windows XP under Parallels...drivers installed, no luck to get ASCOM to communicate to the COM created at the moment I connect the cable, I followed all configurations but even the Toolbox Comm test says Timeout...as well as the EQMOD if I launch it  

[Dave_D]

Well at least you get some time guided...I seem to not get it at all!

that's what's so odd about this.... the vixen gp is a hell of a good mount and i'm brutally overloading it with a 10" newt. if the issue was with the mounts themselves i would expect the problem to show right from the get-go, and  your setup is far lighter than mine.

one thing i've noticed (and i have no idea if this is related or not), but the Orion versions of PHD show a much cleaner image with far more guide stars available than the standard version of PHD. tbh i haven't tried the orion v2 but there *may* be something quirky going on between the orion autoguider and the standard version of PHD that has been 'tweaked' in the orion version of PHD. i may try it sometime. i can't see why though as it's just a rebadged qhy5 after all.

[rofus]

I finally got the mount to talk to EQMOD and PHD on Windows to connect to the mount through ASCOM...I forgot to select PC DIRECT on the remote...for controlling the scope from Mac with SkySafari I did not need it!

Once PHD is connected through ASCOM I can manually move the scope from PHD (and from EQMOD window), so it indeed connected and working through pulseguide (and tracking because as soon as it connected obviously the mount stopped tracking and I had to enable tracking again at sideral speed from EQMOD).

Anyway....now clouds gathered...will wait a bit more to see if I have a spell of clear sky.

Looking through the Syscan configurations I also noticed Autoguidespeed is 0.5x....now....is it worth trying as well again the ST4 connection with this setting at 1x ?

[rofus]

Today I prepared everything for what could be another night of guiding tests:

1) I updated the Motor Controller Firmware and the Handset Control Firmware to the latest versions from Skywatcher website.

2) I changed the autoguide to 1x on the handset, to try later and see if that makes the ST4 guiding working (with my low focal 200mm guidescope and 500mm telescope I think I need a more powerful guiding set on the Syscan and then fine tweak in PHD max settings).

3) I checked Ascom/EQMOD on my Windows XP installation, all working. From EQMOD panel itself, PHD1 and 2 and Cart Du Ciel I can connect and control the scope. I attach my settings in the screenshot.

Later tonight I'll first try again a ST-4 guiding now with the 1x autoguide speed setting. If not working I'll connect it all through ASCOM/EQMOD in Windows and will use PHD through EQMOD and Pulseguide instead of ST4.

[rofus]

Interesting immediate positive thing happening after motor control firmware update: motors slewing is now MUCH SMOOTHER, reading the changelog they made the acceleration/deceleration much smoother (no jumps/noises at start and end of slewing) and fixed a bug that always caused an additional loud noisy jump every time DEC motor started.

I was then analysing guiding graphs of yesterday...first one GUIDED (only RA), second one UNGUIDED (I disabled guide and in the screenshot there's no grey/motor action displayed), both from same calibration.

Basically it seems the unguided one was BETTER (!!) than the guided one..!

[ollypenrice]

I prefer AstroArt because the guide parameters are well laid out, logical and can be altered while guiding to see what they do. Pressing Here Dummy is genuinely great when it works, but when it doesn't I'd rather have AstroArt...

Olly

[ChrisLX200]

Interesting immediate positive thing happening after motor control firmware update: motors slewing is now MUCH SMOOTHER, reading the changelog they made the acceleration/deceleration much smoother (no jumps/noises at start and end of slewing) and fixed a bug that always caused an additional loud noisy jump every time DEC motor started.

I was then analysing guiding graphs of yesterday...first one GUIDED (only RA), second one UNGUIDED (I disabled guide and in the screenshot there's no grey/motor action displayed), both from same calibration.

Basically it seems the unguided one was BETTER (!!) than the guided one..!

Well unless I'm being thick (very possible) from what you are showing in the graphs, there is no significant difference between having guiding enabled or disabled. Commands are being sent during guiding but not being acted upon by the mount. Maybe your cable is duff....

Does this mount actually calibrate ok?

ChrisH

[Thalestris24]

Hi Rofus

How have your calibrations been? Reasonable number of steps? N/S, E/W movements ok?

Posting a debug log might be useful.

Louise

[rofus]

Well unless I'm being thick (very possible) from what you are showing in the graphs, there is no significant difference between having guiding enabled or disabled. Commands are being sent during guiding but not being acted upon by the mount. Maybe your cable is duff....

Does this mount actually calibrate ok?

ChrisH

Actually unless it's the two of us, it's exactly what I see...even more, the unguided one is better!

Maybe is the cable...if so later I'll try eqmod/pulseguide and I should see pretty much different results. For sure I was able to manually guide through the ST-4, meaning moving the scope...so if it's the cable, it's half broken not completely.

CALIBRATION: putting 2500ms or 4000ms I get W calibrated in 6 steps, then the 6 to E don't put back the star where it was, sometimes it even continues to drift in same direction of W calibration. So there's then a lot of backslash calibration, N can arrive up to 60 steps. The guided you see yesterday (the one compared to the unguided) it's with DEC switched off.

[Thalestris24]

CALIBRATION: putting 2500ms or 4000ms I get W calibrated in 6 steps, then the 6 to E don't put back the star where it was, sometimes it even continues to drift in same direction of W calibration. So there's then a lot of backslash calibration, N can arrive up to 60 steps. The guided you see yesterday (the one compared to the unguided) it's with DEC switched off.

That doesn't sound right... You need 7 or more steps in each direction to calibrate. Around 10-15 is usually about right. PHD2 will calculate the step length for you. You should see the display move N/S then E/W (I think! It's been a couple of weeks, lol ). Possibly with a clear backlash step. Anyway, there's obviously something going wrong and you can't expect it to guide if it won't calibrate properly.

Have you balanced properly?

Debug log?

[rofus]

That doesn't sound right... You need 7 or more steps in each direction to calibrate. Around 10-15 is usually about right. PHD2 will calculate the step length for you. You should see the display move N/S then E/W (I think! It's been a couple of weeks, lol ). Possibly with a clear backlash step. Anyway, there's obviously something going wrong and you can't expect it to guide if it won't calibrate properly.

Have you balanced properly?

Debug log?

I'll try enabling the debug log as well to get calibration log. It's first W/E then N/S, but if I follow PHD2 calculated value (900ms) for calibration, it just won't calibrate saying the star did not move enough (my guidescope is 200mm and the scope is 500mm focal length).

Is all well balanced, with that same setup I can take usually 60 to 90sec (Nikon D90) unguided...something is wrong for sure, but it seems wrong with the guiding. I tried as well PHD1, and also PHD2 or 1 under Windows (always through ST4), always same results.

Tonight is clear, I update the motor controller firmware of my eq5, I updated the handset firmware as well, I changed the autoguide setting on the Synscan handst to 1x (it was 0.5x), so I'll try again after all this calibrating/guiding through ST4.

If it fails again, I also have EQMOD configured and tested...so I'll try PHD in Windows through EQMOD....hopefully these tests will explain something more!

[Thalestris24]

Hi

Presumably you are interfacing via ST4 and selecting 'on camera' for the mount? It sounds like maybe you have a ST4 cable / interface problem. Obviously if it won't calibrate with the calculated step size then there is something fundamental wrong.

[ChrisLX200]

I would also look at the mount's drive system itself, check if there is excessive backlash in the worm (meshing with the main gear, and also 'end-float' which is side-side movement of the worm). These clearances will be adjustable and will minimise the amount of movement the guider has to compensate for. Do not try to remove ALL clearnace and thereby leave it tight to turn or the motor will stall. You need to strike a happy medium with this adjustment.

ChrisH

[rofus]

Hi all, FINALLY PROBLEM SOLVED!

I knew before it was not the mount or the scope, balancing or polar alignment, because unguided I arrived sometimes to 2 minutes (depending on position in the sky), and this guiding was something else related only to the guiding. One of the following actions made it work:

1) update of EQ5 Motor Controller firmware

2) update of EQ Synscan Firmware

3) set the Autoguide speed on the Synscan hand controller to 1x

I tried in PHD2 on Mac, guided for two minutes without problems, 2000ms calibration time, all within 10/20 steps. So basically now ST4 works.

I then connected it directly through EQMOD, loaded PHD, Cart Du Ciel, and I guide perfectly till now...first minute it finds a balance with DEC, then it's flat line...tried pinpoint stars up to 4 minutes, trying now 6 minutes, will go on and I'll see what's the limit!

Will keep you updated...so maybe this will be useful also for others. My feeling is that setting 1x Autoguide speed on the controller and updating to latest Motor Controller and Synscan firmwares is a must to get perfect guiding. Motors slew now MUCH BETTER after the firmware upgrade.

Will post more later...

[ChrisLX200]

Well that's some good news.   I don't think changing the guide speed from 0.5x to 1x would make much difference - that is just the speed the RA motor is being asked to turn at. In one direction the motor will be stopped entirely, in the other the speed will be doubled. There should be no reason to go above 0.5x for the modest corrections necessary (and less strain on the motor and mount by using a slower speed). I would try again at 0.5x and if it works use that, if not revert to 1x and live with it...

ChrisH

[rofus]

Well that's some good news.   I don't think changing the guide speed from 0.5x to 1x would make much difference - that is just the speed the RA motor is being asked to turn at. In one direction the motor will be stopped entirely, in the other the speed will be doubled. There should be no reason to go above 0.5x for the modest corrections necessary (and less strain on the motor and mount by using a slower speed). I would try again at 0.5x and if it works use that, if not revert to 1x and live with it...

ChrisH

Interesting, at the moment I connected it all through EQMOD and so I'm anyway bypassing that setting (but I have 0.5 RA rate and 0.9 DEC rate in EQMOD).

In the end is more practical this way because I keep in Windows Parallels all I need to guide/command the scope, and in Mac I only have the app that collects the pictures from the camera, and there's one cable less. Also EQMOD seems to offer more options...

[rofus]

Two pictures 10 minutes....pinpoint stars...stopped only because of a cloud... !

[Christopher Davenport]

Yaaay,

I can feel you grinning from ear to ear!

Well done mate, I love it when a plan comes together.

[Gib007]

This is excellent news, congratulations! I'm surprised you even have a controller there at all. Was the SynScan controller connected while you were also using EQMod? That would seem really weird to me, as EQMod is basically a replacement for SynScan and interfaces a laptop directly to a mount without SynScan. Mind you, it seems very likely that having a guiding speed override on SynScan would completely screw up the commands sent to the mount. In effect, PHD2 would detect an offset, send a correction command and then SynScan would alter it to a level that the mount wouldn't correct the offset fully. This can and clearly does have a run-away effect. 

[rofus]

So after an almost full night (clouds gathered at around 2am) of tests, with a non perfectly balance scope (did not rebalance the slight difference when taking out the 2" diagonal/eyepiece with the Nikon) and also some heavy wind, I can say...IT WORKS!   (yeah that's grinning from ear to ear and behind as well)

First of all...I guided basically all night without problems, apart some RA spikes when scope was almost at zenith, with focuser cable pulling, not balanced etc. 

Apart from that, I attach three screenshots here...guiding worked perfectly in PHD1 (PHD2 seems not as good, either on Mac or Windows). First screenshot is the guiding graph (that I think is quite good) that was going on for 20 minutes that way, and a test shot of 10 minutes on M51 (crop 100%, Nikon D90 on SW Startravel 102 500mm). And also the total settings I used (here is on Windows, on Mac was the same with Autoguide Speed x1 on the handset).

Below there a few important points, hoping it will be useful for others:

1) I'm using a Skywatcher EQ5 Pro Synscan, SW Startravel 102 and 150 refractors (500mm and 750mm), Nikon D90, Orion 90x50 Guidescope (200mm) with Orion Starshooter Autoguide camera, Mac

2) PHD works better than PHD2, both on Mac and Windows. Depending on your focal length, you need to set Autoguide Speed of your hand controller to 1x (I suppose because you need full/stronger correction with short focal length), and 1000ms or 1500ms calibration step ms). For same reason I had to increase Autoguide settings in EQMOD as you can see in screenshot. Maybe these can be tweaked more when I'll use a longer focal length to avoid over correcting, but as you can see they work spot on. Calibration happened all within 12 to 35 steps, depending on position of the object in the sky. 

3) For ST4 you need to make sure your Motor Controller Firmware and Handset Firmware are updated to latest version. To me this update and the 1x Autoguide Speed made the Autoguide work, and also motor movements much smoother, nicer and a lot quieter

4) Polar Alignment through the polar scope (with mount level and scope balanced) is much more than enough to guide for 10 minutes or more with 500mm or 750mm. I aligned my polarscope once when I got the mount (second hand!), set the hour discs, since then I use them to get a good position of where Polaris should be, I center it into the small circle with the alt/az bolts, ad because I put the mount always in the same place, it just takes a minute. Then I do a 3 star alignment with the Syscan, once done I take out the Finderscope and put on the Orion Guiderscope with the Autoguide camera attached, connect the Nikon prime focus 2", focus it with my Bahtinov masks (rough focus through liveview on camera and then fine focus with few test shots on computer), and I'm done. Usually 40 to 1h of setup all included and I'm ready.

5) EQMOD on Windows XP (Parallels) works smoothly, with a simple serial to usb cable. Just be sure to select PC Direct on the handset! For some reason is not needed when using a Mac to slew telescope, but you need to do so when using EQMOD.

6) I got basically same results guiding through ST4 and EQMOD. Everything is connected (guide camera, guide usb serial cable, nikon) to a USB3 hub on the scope, not powered and using a long extension cable to a USB3 port on my Macbook Pro Retina (I leave everything outside and I sit inside controlling scope, guide and Nikon camera -wifi connected- remotely). It works perfectly and I can switch from Mac/ST4 to Windows/EQMOD in a minute. In the end I possibly get better guiding through Mac/ST4 than EQMOD at the moment.

7) On Mac I use PHD1 (guiding), Skysafari 3 Pro (scope control/starcharts), Sofortbild (Nikon imaging)

8) On Windows XP SP3 (on Parallels Desktop 9) I use PHD1, Cart Du Ciel (scope control/starcharts), all through EQMOD, and I keep Sofortbild for imagine running in OSX)

My tests yesterday were done on objects relatively low on W and almost at Zenith, basically calibration takes more steps when at Zenith but then it guides perfectly. My dec usually starts slightly up, within a minute or so it's corrected down, it goes a bit too much down and then it's dead center. RA always starts well.

So as I said, upgrading the Motor Controller Firmware, Handset Firmware and setting Autoguide to 1x (and in EQMOD setting Autoguide speed a bit higher) solved all calibration and guiding problems. The autoguide settings probably were needed considering the short focal length of my guidescope (200mm), while the firmware upgrades just made the motors work much better and smoother and so helping again the calibration/guiding.

I'll add more details if I forgot something. The message is that with a normal/modest mount, without too much fuss to make it work better, with a normal polar alignment and normal balancing without worrying too much about payloads etc., if you get the software/firmware well configured and updated (I'm in IT engineer to be honest) and you're realistic about the focal length you can use to image from your setup, you can get 10 or more minutes autoguided images with a simple and relatively cheap setup like the Orion Magnificient Bundle.

[Dave_D]

For ST4 you need to make sure your Motor Controller Firmware and Handset Firmware are updated to latest version.

What is the latest btw? according to my handset i'm using 2.04.02 but the skywatcher site only lists it as 2.04 so i'm not going to mess about updating unless the latest is >2.04.02

[rofus]

What is the latest btw? according to my handset i'm using 2.04.02 but the skywatcher site only lists it as 2.04 so i'm not going to mess about updating unless the latest is >2.04.02

That's the latest if I remember correctly, not at home at the moment. I had the 2.01.02 and it was doing loud noises at start at stop, not smooth acceleration/deceleration, now between the new handset firmware (they changed alogs for alignment and slewing) and the updated motor, is a different story, feeling better also about the noise generated late night

You seem updated to the latest motor controller firmware version.

[Gib007]

Just remember you do NOT need to use the handset at all. I run my EQDirect cable from the mount straight into a USB port on my laptop. No problems with Autoguide speed because the handset is not even connected.

[rofus]

Just remember you do NOT need to use the handset at all. I run my EQDirect cable from the mount straight into a USB port on my laptop. No problems with Autoguide speed because the handset is not even connected.

I don't use an EQDirect cable (also because is just for Windows), and it was cheaper to use a normal serial/USB cable that connects directly to the handset. Otherwise yes with an EQDirect cable you don't need the handset.

I prefer to use the handset because when I'm setting up the scope I'm anyway outside/near the scope and I find it much quicker to do the alignment with the handset considering I still need to work on the scope itself and not on a computer. Once all is done/aligned/connected and imaging camera roughly focused, I just go inside and do all remotely through one USB cable leaving the handset where it is (serial connected like the rest to my USB hub). I sync Skysafari or Cart Du Ciel and I use those to slew to objects, always well centered and easy to frame better with manual remote slew from the software.

If I run it all from Mac I can use the handset together with the remote control, actually it works together (Skysafari directly through serial/usb cable, Autoguide through USB to camera and from camera to ST4). Then I can either use the handset to go to objects manually, or connect the serial/USB to the handset and (on Mac) I don't need to enable PC Direct mode. On Windows when connecting through EQMod I need to enter PC Direct Mode on the handet for it to work, and then I can decide if guiding through the camera ST4 or through the usb serial cable with PulseGuide.

In the end yes if you don't need to use the handset is better to use an EQDirect cable, but to save some money with a normal serial/usb and if you use the handset for all initial setup outside/near the scope then the usb/serial -> handset is better than an EQDirect cable.

Most of all if you are on Mac and you want to autoguide and control from Mac, you have no EQMod/Ascom and so you just need a usb/serial cable to the handset. And personally I found that guiding via the Mac in ST4 got better results than EQMod on my setup for some reason...