Celestron Advanced VX mount

[saiph]

Hello everyone!

I recently had an opportunity to finally "get down to business" with my AVX, and train its PEC. However I stumbled upon some issues that prevented me from succeeding with this operation.

After more research, I realized I had to switch the COM port of the USB-serial adapter to a smaller value (by default it was installed on COM13, then moved to COM14 as I changed the physical USB port), since the Celestron utility can presumably only communicate via COM 1..4.  But that didn't work out either.

So, right now, in a day or two, I'm having a second chance to go for it, and I'm still clueless as to what should I do to have it communicate and upload PEC data into the mount controller.

All I managed to do so far was to have the guiding software see my webcam (finally...  - an MS LifeCam) and get some activity going on the screen. That probably means I'll only be able to record some PE data. But how can I go further afterwards and upload the correction data into the mount?

 

Does anyone have a clue, please?

I'm using a Hama 00053325 USB-Serial cable adapter, btw, that uses the Prolific platform (not the FTDI one, which I noticed on another forum as recommended connection accessory - I'd like to know why is there such a difference between them and how can I use what I already have), on a Win7 laptop. PEC Tool is version 1.0.18, PHD v.2, MetaGuide 5.2.6 and GuideDog can assist as needed. As primary scope I use a C80ED.

 

L.E.: Can anyone post a complete checklist, step-by-step, with pre-requisites, how to perform a successful PEC Training session, along with comments on what to keep an eye on, what not to miss, what to use and what not to? as I could only find bits and pieces of information, with lots of steps forgotten or skipped intentionally, and things should be a little clearer than that imho when it comes to - I'd say rather - delicate matters such as the one mentioned in this topic.

 

Thanks and clear skies!

[Shelster1973]

Am going to redo my PEC as have had the mount in pices and it is something I need to get done.

You arrre right, there is a lot of confusing and sometime contradictory information on the web about this, but guess that is par for the course for the tintyweb.

What I plan on doing is the following

• Setup scope on mount
• Polar Align
• Carry out Plate Solving with sync to get mount star/goto aligned
• Start PHD to guide mount
• Run PECTool over several worm rotations to get average of PEC
• Upload PECTool output to mount
• Start playback of PE
• Image away while guiding

The one thing that I have taken away is that the recording you upload is held in the mount when you power down, but it needs to be set to play when you re-power the mount otherrwise it will not be applied.  You will also need to re-write it to the mount if you update the firmware as it gets overwritten.  You can use PECTool to save what is in the mount prior to upgrading and then upload again.

I did not record my PE before I carried out the works, so do not have a record of what it was and what it is now, which is mildly annoying but am not going to die in a ditch over that.

Also read that you can run PECTool without guiding, but this is where the confusion comes in I suppose.......

[saiph]

So, what you're saying is that I have to run both PEC Tool and PHD (or Metaguide - I recently discovered PHD won't see my guide webcam right) and record PE data into PEC Tool only while guiding?

If I save a log while guiding with Metaguide and load the PEC Prep log file with the corrections applied by Metaguide into PEC Tool and then upload it into the mount's controller, isn't that going to do the trick?

 

Last night I managed to measure the PE in tracking mode, over only two cycles : +/- 23".  When Metaguide started to apply corrections it went down to under +/- 4".  I'd say I'm happy with that number, provided that i'll be able to get that kind of residual error every time I playback the PEC; going for another run tonight, so I hope i'll get some good data over 5-6 cycles to average out random error values, and then upload it into the AVX.

 

[Shelster1973]

As far as i understand yes.  PECtool is what is recording the PE error and then subsequently uploading the requried corrections.

The corrections that metaguide applies are not the same as what the PE is.  These are correcitons for other parameters that effect the tracking accuracy of the mount (PA, imbalance, wind, small furry animals etc).  The PE is a constant that occurs due to the imperfections of the worm and gear box.

I have to admit that I find it odd that you need to guide when doing PE recording as, as I have just mentioned above, the PE is going to be the same due to it being in the gears, but the guidance (even from Celestron) has that you do it that way, so is best to stick to the guidance.

Suppose there is no harm i doing it both ways (one with guiding and one without) and then seeing what the ourtcome is.  The ultimate goal is to capture the best data, so go with whichever method gives this at the end.

[saiph]

Ok, so I managed to get some PE data using Metaguide last night, with my AVX mount

I've been digging in the log file I saved and got the attached graphs. The amplitude of the residual periodic error looks low enough for what I intend to do with the mount. But first I need to let the controller know how to quiet down the mount's PE; so, here's:

 

- To all experienced AVX users who succeeded in training their mounts PEC - how do I get these numbers up into the mount's controller now?

 

I'm interested in the actual way, step by step, with recommendations; not only "use that software", or "use that cable".

This is crucial for me to understand, especially with all the headache already caused by the entire process so far.

The raw log files have a header and then the data flows as a table. It covers four complete RA worm revolutions.

PEC Tool is only able to load ".CSV" files. Possibly of all four revolutions, possibly of one single averaged revolution (with random noise canceled out), I have no idea, as no one would explain the entire process thoroughly, nowhere.

 

What data should I copy out from which table / sheet in the attached spreadsheet in order to produce a valid CSV file for PECtool to upload to the AVX's controller, that would provide the correction data for its PE?

 

I also took a single-cycle CSV file simply using PECtool, but it seems rather truncated - the graph looked incomplete, although it took its sweet 10min time to gather the data from the worm, identified the index and so on. Even uploaded that to the mount (before saving it), but I wouldn't put my trust in a one cycle set of data anyway - I was just curious if communication worked this time (after re-assigning COM port value to serial-USB adapter cable and inserting it into the controller port, not the mount  itself!  everywhere in the available documentation it says you should connect or transfer the data to the mount - which is WRONG!  the controller needs to handle all the communication.  why is it so hard to write the right words in a document... create all this fuss and confusion just for fun?); and that is why I need to upload the averaged data from the set in the attachment.

I used PECPREP to extract and generate the datasets for one (smoothed), four (original), and 10 (simulated) worm revolutions. There's one also with the inverted data of the correction graph - perhaps that is how the errors actually look like, the correction is just a mirror image of that with respect to the center line.

 

Just to make it clear - PHD didn't recognize my webcam correctly, so the "why don't you use PHD instead" reply will not satisfy me   I don't have another more suitable guide cam yet, so Metaguide worked fine with what I already have, produced some data, now I need to put it to some good use. Btw, PECtool and any guiding software simply refused to work together at the same time, FYI. And it stops logging automatically after one revolution - therefore I don't quite get it, how can it average multiple cycles - unless I load a pre-compiled CVS file containing enough data to cover more than one cycle. Metaguide kept on logging until I stopped it; and splitted the log files by itself into one big log for four revs and one for the remaining excess time.

And, again - "quit PEC, go for guiding instead" is such a mood killer. Just don't bother serving me that condescending backhand slap, please  I know it's so true and it makes everyone feel so righteous, but.. it's just not for me, ok?  I'm tired of justifying my "love" for guiding..... Thanks.

 

Clear skies!

 

 

 

PEC-data.xls

[Shelster1973]

I feel your frustration.  Getting you head around these things is mightily frustrating.

Am sorry that I cannot help you any more with your current setup, but have you tried looking at PEMPRO.  I have not used it myself but am planning on getting the demo and trying it out.  It works with a vast number of different cameras, so may be useful to you.

 

[saiph]

Thanks Shelster  I appreciate your input.

Waiting for other fellow stargazers to check in - I've seen quite a few. I doubt no one has any idea about this issue, especially when I saw that they use their PEC'ed AVXs to take some pretty decent pictures.

I've been doing some more research meanwhile, and I think I'm on to something here. I used PecPrep to tune up the PEC curve by filtering out some high pitched noise in the default error curve, thus resulting a smoother correction curve, which I limited to -1..+1 values in XL, and then saved that column of data into a CSV file (as I noticed the file that PECtool produced is configured pretty much that way).

 

I'm going to make another attempt today, at uploading this file into the mount's controller. If it fails, i'll try some more tricks on the filtered data before saving it again into CSV, and upload again to the AVX hand pad. Hopefully, tonight i'll take some test exposures with the PEC'ed mount   It's supposed to take it down to  -8..+16" +5.7..-3.5" of residual error (if I'm reading the graphs right).

Further tweaking of the frequency filter yields a smoother curve and a more balanced residual error - cutting down below Mag8, and limiting low pass filter to ~150, I get some +/-6" of RPE, which is quite ok I'd say. I'm hoping this will work and the mount controller will take my correction file.

L.E.: Attached - the periodic error correction graph, after applying the above mentioned filters on the raw PEC Prep data. For PEC Tool it's pretty much the same graph, only the data are constrained to a -1..+1 interval (not truncated, but factored down).

 

 

 

[saiph]

So, the file is finally looking right for PEC Tool. 

After hours of refining, converting, rearranging and understanding data format and values in the so-called CSV file (honestly, there's no coma whatsoever anywhere in that file... only blank "space"s and "new line"s).

The software does load it now, but when i click to upload the data into the mount's controller, all i get is that error message  "Run-time error '5'. Invalid procedure call or argument" and then nothing happens, it just crashes and i'm left looking at the PEC data i have.

Oh, right - and why in heaven's name "88" PEC cells?  When there are 598 entries - one for each second of data..

Ok, the "7200" as end of file marker i get. It's all about programming. But "88"..?  What am i supposed to do with my data file - compress it to only 88 values, out of 598? Should i just throw away to garbage 86% of perfectly useful data??

 

Any piece of knowledge will be greatly appreciated, as this is getting more and more frustrating - i am so close to finally get this thing done, and i'm being held back by a glitch!?  

 

 

 

L.E.:  well, that's how it wanted to play, that's what i did - i manually rejected 510 data points, by choosing only those that matched the algorithm: every sixth division had a 5 data points gap instead of 6, and so 88 data points resulted, that were saved and then loaded into PEC Tool.  Although it recognized the file, with the "7200" at the end of it (and the "88" in the first row), it wouldn't upload it to the mount's controller; instead, i had to remove the EOF marker, save, load again (the error correction graph looked pretty much identical in the small viewport in PEC Tool now, finally, as it did in Excel and PEC Prep) and gave it one more try to upload - this time with success   played it back and it seemed to follow the line.  Now i only have to test it under the stars. Yey!..  wish me luck

[saiph]

Just one more curiosity:  

After downloading the PEC data into the mount, i thought i should back check that data, by downloading the PEC stored in the mount's controller. And guess what - they're not the same! Why??  See attached the curve as it looks before clicking "upload" (obtained from the acquired PE data and transformed so that PEC Tool can understand it - as in not arcseconds, but percentage of stepper spin rate), and what the same curve looked like after clicking "download".

In the end, how will the mount's PE be corrected? By following the first graph, or the second? And again - why?

There will be a bit of a huge setback in performance, if you ask me. Like it wasn't enough they crippled the PEC data to just 88 cells, now they also mess it up completely by doing.....  _this_ to it?!

Unless it's only a display glitch, although i doubt the PEC Tool is not able to show its own data right.. 

Helloooo!! Celestron team! Really, bruh..?...

I'll write them an e-mail, again, right now.  I feel we're starting to become "friends" already  lol

 

 

[Thalestris24]

Hi Saiph

Um, to the best of my knowledge there's no real advantage in PEC training if you are guiding. Periodic error takes place quite slowly but guiding corrections happen every few seconds. It's probably worth doing if you're not guiding and you're imaging at longish focal lengths.

Louise

[saiph]

I am pretty much aware of that, Thalestris  thank you.

However, i am not interested in guiding right now, as i mentioned already. It's not within my reach, to say the least.

I've seen users taking their unguided mounts (only PEC'ed) to pretty decent exposure times using focal lengths that i also have in my current setup (up to 600mm). So, i see no reason why i shouldn't be able to reproduce the same kind of result with virtually identical tools. See my point? Am i a special case, or what? What do others have and i don't? 

[Thalestris24]

Hi Saiph

Sorry - I should have read your previous posts! I can't help with PEC advice but hope you can get somewhere with it. Do you have a focal reducer/flattener? I think that would help.

Regards

Louise

[Droogie 2001]

Good luck Saiph with your PEC testing though I am not sure how long you are expecting to image without guiding?
With good Polar Alignment you can probably reach up to 4-5mins unguided with the AVX depending on the where the target is in the sky. I think got fair results between 2-3mins again the location of your target has alot of baring on how long you can go before star trailing.

Do you use a Polar scope?
Obviously you have mentioned your reluctance for additional 3rd party equipment though Celestron do make an all in one NexGuide:

https://www.celestron.com/products/nexguide-autoguider

This will require a guide scope but alleivates the need for a PC and alot of cabling. I hear mixed results with this device but maybe a way forward if you get PEC sorted.
As others have mentioned I looked at PEC but never really got anywhere with it so I moved straight onto guiding with a laptop.

 

 

[newbie alert]

Seem a lot of hard,headsratching work which will far exceed my capabilities..on a 80ed a finder/guider with a planetary cam sounds a far easier route.. what's the target goal in terms of exposure length? I've followed this as interested in pec training my avx but don't think I worry..far too technical..think I would of trashed it by now

[saiph]

Thanks for the recommendation with the Nexguider. Kind of like the Synguider, if i'm not mistaking.

I'll probably try one as soon as my funds allow me....

As for the C80 & AVX & PEC - yes, i use a 0.8x reducer, recently coupled to a 4 slot 2" manual filter wheel and it works pretty well. Except for the PEC thingy.

I suspect that the correction curve PEC Prep provided is exactly identical to the PE graph, as in not the correction graph - so instead of canceling the random motion, it doubles it. It's just a hunch - i inverted  the curve and trying to upload it again to the controller. Next clear sky window i'll also test this hypothesis.

For now, this is all i could get: an image with M27, LRGB, with around 2-3 usable frames out of 10 (60s each), but i used a little more for this test image - as you can see, stars look like anything but points....

The error in star position was somehow greater than in my former EQ5 mount (with no correction), which is embarrassing to say the least.