ASIAir disconnecting in Plan mode.: Page 2

ASIAir disconnecting in Plan mode.

w7ay

Byrdsfan1948 I would then probably get the RST 135

There are other strain wave mounts, too, Al. Like the Hobym (also Korean, their CEO and RainbowRobotics (parent of RainbowAstro) CEO used to be partners) and the upcoming Sharpstar Mount. (Not sure what the Sharpstar is like yet -- most likely they are not using the gears from Harmonic Drive's subsidiary in Japan.) The Sharpstar is really heavy, though, just like their OTAs (the Askar line of refractors from Sharpstar are ridiculously heavy).

Just need to make sure that your software can control them.

Once you are spoilt by no counterweights and no need to balance, you will never be able to go back to legacy German mounts again. My Tak EM-11 mount has been idle ever since getting my RST-135s.

Based on the local weather forecast I have a solid week of New England cloud cover to get this done in

Based on historic climate, I won't have cloudless skies until April!

folding the cable from the mount and firmly attaching it to my ASIAIR was my first jury rigged attempt to eliminate any connector deflection at the ASIAIR...... It worked !!

Yep, whenever I read of people with sporadic, unexplained disconnects, the first thing that occurs to me is that there is a loose connector.

That being said, ZWO could easily monitor and recover from the problem without the user even knowing a disconnect has occurred (except losing part of an exposure that has to be rerun).

Chen

Byrdsfan1948

w7ay Hey Chen, Given all the time and effort you put into helping me with this on going problem I thought I at least owed you an update. During the four or five days of clouds here in New England I was at least able to secure all my cables. As well as all connections. Nothing as elegant as you have done but at least secure enough to hopefully assure no connection failures. Unfortunately that has not resolved the problem. Last night was very interesting in that both times that my Plan mode was interrupted by the "Connection to Mount" failure notice, the ASIAIR continued to run the rest of the Plan as normal, with the exception that the mount was disconnected. The mount continued to track the object but of course guiding stopped. When it came time to slew to the next object in the Plan, the mount did not do that. The Plan however did pick up as if it had slewed but continued to image the same object it was on, as if the mount had slewed to the new object in the Plan........... The images were incorrectly saved as if they were the new object Finally at the end of the Plan imaging session the ASIAIR followed the instructions to turn off the camera cooling, and shut down......... It of course did not return to the home position. I was able to return the mount to the Home Position using the hand controller....... So basically the problem remains... I'll keep digging in on this problem maybe more concentrated specifically on the actual ASIAIR to Mount circuitry. I've changed cables to new ones..... My next thought would be to change the current ASIAIR USB port I have to Mount connected to to another one. But I already have tried my back up ASIAIR and had this same Plan disconnect failure. I will also try some old fashioned Autorun sessions just to make sure this is Plan Mode isolated..... Just an update Al

w7ay

Byrdsfan1948 Just an update

OK Al, check your log to see what occurred.

It sound like your mount is still disconnecting, but we need to know if it is just disconnecting sporadically, or while it is attempting to move. Or did it happen when the ASIAIR was communicating with a different USB device (like an EAF, or filter wheel). That could tell us a lot.

If it occurred while doing a meridian flip, then connectors are almost surely the problem -- but it could be extra draw from the power supply too (the RST-135 sips power, but some mounts could draw an an extra 1 or more amps while slewing). You can test this even indoors by making ASIAIR execute successive GOTO between a location just west of the Meridian to a location just east of the meridian, to see it it stays alive. And also measure the 12V supply right at the mount's power connector to see if it drops while the mount is slewing. Best to do it in the cold, too, but you can do that while it is cloudy.

Make darn sure not to use the USB hub on your camera to connect to the mount. Anything USB from ZWO is poorly designed, and I have an ASI183 hub that would sporadically fail even at room temperature, indoors, and stationary, when I tested it.

There are also multiple ways that your mount could be the culprit.

It is also possible that your mount is sending some response to ASIAIR that it does not recognize, causing that thread in ASIAIR to crash. ASIAIR uses multiple threads (main camera, guide camera, mount, etc) and it is possible for one thread to crash while the rest is still running. But usually, when that happens (which you read a lot about too), you cannot connect back to the mount without a reboot. Well written code will look for bad responses and ignore and try to recover, instead of crash, but not well designed code may just crash.

It is also entirely possible that your mount is back powering the USB -- you can check this with the cheap USB power meters -- see if the Mount's USB hub is putting out any power. If so, your solution may be as simple as adding a USB hub before the mount.

Did you take care and add a ground bus between the mount and ASIAIR? I seldom see astronomy hobbyists do it, but I think that I have mentioned a ground wire that runs between the ground of the ASIAIR and to the ground of my RST-135. This way, I can be sure that the thin USB wires are not the ones that carry the ground return of the 12V supply to the mount. I didn't go as fas as to use a braided bus wire, and simply use 18AWG wire.

It is just weird that so many people are having problems with the ASIAIR.

I still have no opportunity to take the released v1.8 ASIAIR for a spin. Heck, we had our first snow last night. About 1.5 inch of accumulation at our elevation.

Chen

Byrdsfan1948

w7ay Hi Chen, First I have to say that as an old engineer, I really do enjoy trying to resolve a technical challenge. Although I am concerned that my rust is starting to show on this one :-((

 I can answer a few of your questions here, a few more I need to check. On the ones I can answer.

From checking the log, and also from comparing what I can see in the images that would correspond to when the disconnect failure occurred, both disconnects last night occurred while imaging. Not during any slewing or during a Meridain Flip. BTW my mount, Ioptron CEM25 EC and my ASIAIR seem to enjoy doing MF's because they are usually done perfectly.

The first disconnect occurred between images 17 and 18 of the first target. I can see the results of the disconnect in the images before and after that, with perfectly round stars up to 17 but becoming slightly elongated image 18 and after. I was taking 5 minute subs. The mount of course had continued to track, but was no longer being guided. Since this was my first target, I saw this unfold in front of my eyes so I also know in this disconnect occurred in the middle of an image while tracking and guiding and not when the mount was attempting to move otherwise, Slew,etc.

The second disconnect happened latter in the evening but looking at the images and comparing them to the Log suggests exactly the same thing. Just an addition comment...... I have never had any disconnect problem with any Go To or Centering after a Go To. Or as I mentioned during any MF which I have recently done many of.

The only other USB device I have other than that main and guide cameras is an EAF, and neither disconnect happened during a focus adjustment. So I would call this a "sporadic disconnect" and not when the mount is moving, other than "normal tracking"

I have not measured the actual current draw, but last night in an attempt to rule out any possible power source issue I powered the set up using a 120 outlet, and two 12 volt 5 Amp bricks. I usually use my Jackery Explorer 240 power pack to power my set up. Up until this recent issue with the mount disconnect that has never been any problem at all.

My dew heaters were totally isolated to one brick. The Mount and ASIAIR to a second separate one. FWIW I have tried to power my mount completely separately using the it's power brick. But when I try that the ASIAIR will not find the Mount. If I use that same power Brick ( 12V, 5A ) to power the ASIAIR and then from one of the 4 power outputs on the ASIAIR to power the mount it is then found. This is typically how the mount is powered and was so last night. I just mention that to you the EE ........ Maybe that means something to you that I as an ME am missing???? "-)))
Sorry no, I have not added a ground wire as you describe. Could you elaborate on exactly what you did there? I will certainly do that as well if possible.
It's not a good night here for imaging but I will run a Autorun test to see if I get a failure in that mode as well...... The next few days will be cloudy but as you suggest there is no reason additional test cannot be run........... Which I will do

Yes all these ASIAIR problems are very concerning. As well as the fact that I never hear anything back after sending in my logs except a thank you....... I recently canceled my order for the 2600MM camera for this reason.
Al

w7ay

Byrdsfan1948 The mount of course had continued to track, but was no longer being guided.

This is good information. Tracking is a mount property (don't need ASIAIR to be alive). Guiding is an ASIAIR property.

Keep in mind, while you are searching for problems that the ASIAIR guide pulses are done by sending very short slews (in tens to hundreds of milliseconds) commands to the mount, which is set to slew at usually 0.5 sidereal rate. There are mounts that accept pulse commands for much more precise guiding, but I have not seen ASIAIR use them. So, while not strictly "slewing," the ASIAIR is constantly, micro slewing the mount while autoguiding is turned on. (Strictly speaking, the ASIAIR sends a slew command to the mount, run a timer for N milliseconds, and turn the slew off.)

So the ASIAIR can still lose connection with the mount when not large scale slewing to a different target.

Could you elaborate on exactly what you did there?

The best is to see if there is ohmic contact between the metal shell of the mount's power connector to the mount's body. If so, find a bolt on the mount, sandpaper away any paint around it, and use an eye connector to connect a heavy gauge wire to it, and use as short a ground cable as possible, connect to the metal box of the ASIAIR. If you don't want to scape the paint, try to splice the ground wire as close to the mount's power connector, as possible.

Shielded USB cables can also provide a good low impedance path for ground through the shield-- not as good as a thick ground strap, but better than nothing. The flat ribbon cables that ZWO ships should be outright banned. Even cheap Amazon vendors don't stoop that low. They obviously don't teach the same EE classes that are taught in the schools that we attended :-).

[Half of my undergrad and grad profs can trace their doctorates back to Vannevar Bush from your old school :-). Bush's differential analyser was a genius device to solve differential equations, and even Claude Shannon (another product of your school) spent his undergraduate assisting Bush with his differential analyzer. My own advisor (RIP) traces back through Terman who was a student of Bush. Terman was the West coast Bush :-)]

Look for the shortest shielded USB cable that matches your needs.

I recently canceled my order for the 2600MM camera for this reason.

Other manufacturers have higher qulity cameras based on the same Sony IMX571. Just Google for the IMX571.

Chen

Byrdsfan1948

w7ay Chen Thanks for that information. I'll see if I can get this ground wire installation accomplished :-)

Additional Info.......... Clouds tonight prevented an extended test, but I was able to run a complete Autorun cycle equivalent to at least equal to what I tried to run last night in the first part of my Plan Mode session before the disconnect failure. Everything ran perfectly as it always has in Autorun.

I selected the target, pushed Go To and the mount did it's thing. It auto centered etc. I calibrated, and focused. Then ran the Autorun sequence........... It completed flawlessly. Then returned to Home and shut the ASIAIR down. As programmed. By then the clouds had covered enough of the sky so that an additional testing was not possible. I should mention that aside from an occasional problem, always resolve by re calibrating, my guiding has been outstanding through out all of this.
While tonight's Autorun test is hardly conclusive, it along with all the other data certainly suggests to me that ZWO has introduced, like the mutating Covid 19 virus yet another gremlin mutation in their easy as 1,2,3 miracle machine.

HAHA Yes Differential Equations!!! :-))) What would Mechanics of Vibrations, Fluid Mechanics, and Gas Dynamics been without them??? I cannot imagine what you EE's had to suffer through that way!!! I don't know about you but I learned more Calculus in my engineering classes than I ever did in the Calculus, Diffy, and Linear Algebra classes.......... Unfortunately virtually all now long erased form the memory banks due to time....... :-((((.................. But like you, who has so well demonstrated with so many of your responses here, I can at least still run tests, analyse results, and in my case anyway occasionally still solve a problem or two............
Al

w7ay

Byrdsfan1948 Everything ran perfectly as it always has in Autorun.

Al, if it continues to work on Autorun and consistently fails in Plan (I don't use Plan since I try to get as long an integration time that I could), then you would have solved on of ZWO's bug for them.

Chen

w7ay

Byrdsfan1948 I don't know about you but I learned more Calculus in my engineering classes than I ever did in the Calculus, Diffy, and Linear Algebra classes.

Hah, you can never stop using differential equations. I was fortunate enough to take a Partial Differential Equations class from Paul J. Cohen himself back in the 1970s. (He once tried to chase two of us off of his Number Theory class because the other chap was a Physics grad student, and Cohen didn't want to waste his time teaching Number Theory to non-mathematicians :-).

Chen

Byrdsfan1948

w7ay Hi Chen, A few comments: While Plan mode my not be of much interest to you, it is an excellent option for me, and provides several very valuable benefits. The first being that from my preferred viewing location, on my patio, both the side of my house and nearby trees, limit my FOV. I also have the side of a mountain to the East that delays when an object has risen above the mountain for viewing. As a result, I can only image a particular target through that FOV for a max of 2-3 hours each night. If I am only going to image a single target then call it a night, Autorun is fine. In fact before Plan Mode I used to use it ALL the time. I would get a few hours of imagining in on a single target until it moved beyond my FOV. Then get my scope back into the house early enough to get a good night sleep. But that also meant that the rest of a rare clear night was wasted.

Plan Mode was a game changer for me.  It allowed me to target whatever passed through that FOV all night long.   So I could program a series of 3 or 4 even more different targets to image each night as they pass one by one through that limited FOV.

Another great advantage is that with Plan Mode I can set a start time, ( I don't recall if that can be done with Autorun, I'll check latter ) .... 11 PM for example. So after I set up, do my PA and complete any other things like maybe run an Focus, I can then hit start. At that point the ASIAIR is on auto pilot. I can go to bed, the countdown "clock has started", and will count down to 11PM when the Plan will start running.
It's a beautiful thing when it actually works as it is supposed to, which it did for a while. If I just used Autorun to do the same thing I would obviously be up every few hours moving from target to target manually. In case you are wondering, I could image from a location with a better FOV... However the limited ASIAIR WiFi range, or the length of my Ethernet cables if I directly hard wire connect to my Home WiFi greatly limits those choices. As does my concern not to leave my Scope set up out in the middle of the yard all night for the world to see, an easy target for anyone interested in getting into this hobby on the cheap !!!

  After last night's test I still obviously cannot rule out a connection failure as the random culprit   However as a result of your advice I think I am far better off that way now than I was before.     My suspicions are increasing that this is just more bug in the ASIAIR software.

 I have no idea what might have been involved for ZWO to alter their programming to add the Plan Mode.  But it seems to me that all Plan Mode is, is just the ability to string a number of Autoruns together, and add a start time.

Seems simple enough to me........ But since I am no programmer maybe there is some additional complexity that could be causing this disconnect issue?

It is interesting that so far no one else that I know of has indicated they are having a similar Plan Mode problem.   Which does suggest of course that the problem could be 100% all on my end. Although I do see many others who have complaints about various different disconnect issues on other forums, perhaps there is some common connection?  Also I suppose it is possible most people are like you, and have no real use for Plan mode and only use Autorun?

Anyway the investigation continues. I can still run some simulations the next few cloud filled days to see what I can learn. Al

Byrdsfan1948

w7ay WOW you certainly had some prestigious instructors. I had a few, but no big names like you. Perhaps my most interesting Professor was Dr Lee Chun ( yes Chun not Chen 🙂) ). I took Relativistic Physics as an elective under him. I actually had the highest test average he had had teaching that subject up to that point as Professor. He allowed me to skip the final as a result. I loved the subject matter so much I took the final anyway and almost blew my A as as result!! But I still pulled it out :-))) He got his Ph'D in that subject area at Princeton while Einstein was there in the early 50's. I don't know of any connection he had with Einstein other than that, but I like to believe that it was at least enough of any connection for me to feel like I got my knowledge in that area pretty close to first hand :-)) Yes we actually use Diffy all the time everywhere just never give it much if any thought Al

w7ay

Byrdsfan1948 start time, ( I don't recall if that can be done with Autorun, I'll check latter )

You can set the time delay until Autorun starts the first shot. But I don't use that either. Since my main interest nowadays is in instrumentation, four hours a night is more than enough for me to collect data. I just wait for the target to pop up over the trees, start the session, while monitoring everything. Sometimes, I let Autorun shut down by itself, but most of the time, I am there to move the mount to the next target. I try to stay away from automation as much as a car enthusiasts stay away from automatic transmissions.

I took Relativistic Physics as an elective under him.

I took a General Relativity course in the math department from M. M. Schiffer (he is one of the authors of an early text on GR -- you can still find it at Amazon books for an arm and a leg!). For me though, the best lectures from Schiffer was the class he gave on "perspectives in mathematics", a two semester course that connects the different fields of math together (e.g., how the Fredholm integrals leads to a natural linkage between Heisenberg's matrix method in quantum mechanics and Schrödinger's differential equations).

Amazon even has a two volume book on selected papers of Schiffer after he passed away.

One of Schiffer's traits was he would walk into the classroom holding just a new piece of chalk. By the time he is done with the chalk, all the blackboards are full of equations, and he walks out :-). These old profs don't use lecture notes. They have enough knowledge in their heads that they can take the subject line to whatever comes up at the moment.

I wonder what he did later when the dry erase pens came around!

Chen

Byrdsfan1948

w7ay One of Schiffer's traits was he would walk into the classroom holding just a new piece of chalk. By the time he is done with the chalk, all the blackboards are full of equations, and he walks out :-). These old profs don't use lecture notes.

HAHA YES!!! I had several like that. Two in particular stand out that way. Dr Saleeby, for Plasma Physics ( part of the flight propulsion dynamics classes ) and Prof Lobay a brilliant Russian engineer who I had for both my Strength of Materials classes. Saleeby would walk in barely acknowledge the class start on one end of the board and only stop when the chalk was gone. Basically the classes were one class of lectures, where he instructed and you wrote down anything you could keep up with. The next class was as describe above. It was our job to stop him to have him explain what needed better explaining......... Lobay scared the crap out of me.......Man he was an engineering genius but being a Russian just his accent made me feel so inferior. He just passed away BTW last year at over 100 years old. Man talk about complex mathematics when it came to solving loads on indeterminate beams etc. All calculus....... I clearly remember his heavy Russian accent as he would be writing on the board. 'Please to remember the DV/DX " :-)

Great times ahh Chen?? :-)) Al

w7ay

Byrdsfan1948 Russian just his accent made me feel so inferior.

Cohen had a Brooklyn accent, but Max Schiffer had a strong German accent.

The school had lots of foreign profs, although my advisor was California born. Ron Bracewell (solar radio-astronomer), also in the Radioscience Labs, was an Aussie with a strong Aussie accent. I took a Fourier Transform and an Interferometry class from him -- both still incredibly useful in my hobbies.

The CS department was different, mostly local borns. Don Knuth is native born (I took a "Concrete Math" class from him -- no, not mathematics of concrete, but basically numerically based math) in the CS department from him.
I still have my "Knuth Check" -- http://www.w7ay.net/chen/Others/More/knuth.html . Squirreled away in a folder so it does not fade in sunlight (many people frame them :-).

Chen

« Previous Page