Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Avalon M Due with absolute encoders?

  • Please log in to reply
25 replies to this topic

#1 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 6,060
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 05 March 2021 - 07:11 AM

Anybody know anything about this yet? Life’s events have kept me from imaging for the past six months. This may be just the thing to get me out of my funk.

https://www.avalon-i...rgo2-pro-detail

 

Derek

3AA17870-27B8-41AB-A464-F49DB3B5087D.jpeg



#2 Tom3

Tom3

    Ranger 4

  • -----
  • Posts: 363
  • Joined: 23 Jan 2018
  • Loc: Scottsdale, AZ

Posted 05 March 2021 - 12:35 PM

It looks like a great mount but I'm not yet ready to spend that kind of money and there are other mounts with absolute encoders to consider when spending $10 grand.

 

Tom


  • schmeah likes this

#3 psandelle

psandelle

    Soyuz

  • *****
  • Posts: 3,634
  • Joined: 18 Jun 2008
  • Loc: Los Angeles

Posted 05 March 2021 - 12:57 PM

If they turn X-Solver into an all-sky modeller for unguided (if it's possible with this mount), then it becomes more interesting to me. I am intrigued by their evolving mounts, though. I'm all for it!

 

Paul


  • schmeah likes this

#4 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 6,060
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 06 March 2021 - 08:51 AM

Price seems more than competitive, relative the others (Mach2 and 10 Micron) considering the T-pod is included. I’m always intrigued by Avalon’s innovations. Hopefully we’ll know more details soon.

 

Derek



#5 psandelle

psandelle

    Soyuz

  • *****
  • Posts: 3,634
  • Joined: 18 Jun 2008
  • Loc: Los Angeles

Posted 06 March 2021 - 09:45 AM

Price seems more than competitive, relative the others (Mach2 and 10 Micron) considering the T-pod is included. I’m always intrigued by Avalon’s innovations. Hopefully we’ll know more details soon.

 

Derek

Only one encoder though, if I read it aright.

 

Paul



#6 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 6,060
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 06 March 2021 - 10:16 AM

Only one encoder though, if I read it aright.

 

Paul

So it seems. Still a lot to learn about here. I presume a high resolution absolute encoder on the RA axis will substantially reduce the large PE inherent in these mounts, reducing the work of guiding, although I never had anything other than excellent guiding with my M Uno. RA drift was an issue on occasion, and hopefully this will be eliminated. They don’t mention unguided imaging capability, which everyone likes to feel they can do with their mount, regardless of whether or not it leads to any improvement in their imaging. I like the increased capacity and ability to run two imaging setups with no flip.

 

Derek


Edited by schmeah, 06 March 2021 - 10:25 AM.

  • psandelle likes this

#7 psandelle

psandelle

    Soyuz

  • *****
  • Posts: 3,634
  • Joined: 18 Jun 2008
  • Loc: Los Angeles

Posted 06 March 2021 - 02:11 PM

So it seems. Still a lot to learn about here. I presume a high resolution absolute encoder on the RA axis will substantially reduce the large PE inherent in these mounts, reducing the work of guiding, although I never had anything other than excellent guiding with my M Uno. RA drift was an issue on occasion, and hopefully this will be eliminated. They don’t mention unguided imaging capability, which everyone likes to feel they can do with their mount, regardless of whether or not it leads to any improvement in their imaging. I like the increased capacity and ability to run two imaging setups with no flip.

 

Derek

Yeah, my thought, too (I have M-Zeroes): they guide very smoothly, why the need for the encoder (and without all-sky modelling with refraction input, it's not really unguided grin.gif )? Still, Avalon keeps evolving their mounts, which is a great thing.

 

Paul


  • schmeah likes this

#8 KenS

KenS

    Apollo

  • *****
  • Posts: 1,088
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 06 March 2021 - 03:49 PM

The StarGo2 is Raspberry Pi 4 based so that opens the possibility for sky modelling to be developed - either by Avalon or open source.


  • schmeah and psandelle like this

#9 Wiser

Wiser

    Sputnik

  • -----
  • Posts: 32
  • Joined: 10 Dec 2018
  • Loc: Utrecht, Netherlands

Posted 09 March 2021 - 07:05 AM

Recently, I equipped my Avalon Uno with a TDM Drivemaster encoder. I posted a detailed description of the development route on the Avalon forum (link).

The combination works quite well fully loaded (C11 and accessoires ca. 15 kg), producing an RMS >0.14" and allows precise unguided imaging. For sky modelling, I am experimenting with The SkyX and T-point. The SkyX offers a special ascom 2x mount driver that treats the mount and the T-point model as one single entity.

 

The adapters were designed and CAD drawn by the inventor of the TDM on my request and are quite straightforward to make. The advantage is that the system can be fitted to existing mounts. I asked Avalon if they were interested to manufacture them as an upgrade offering but didn't get an answer yet. Alternatively, a few adapters could be made independently for those interested.


Edited by Wiser, 09 March 2021 - 10:41 AM.

  • psandelle and Manitu like this

#10 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 6,060
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 09 March 2021 - 09:10 PM

Recently, I equipped my Avalon Uno with a TDM Drivemaster encoder. I posted a detailed description of the development route on the Avalon forum (link).

The combination works quite well fully loaded (C11 and accessoires ca. 15 kg), producing an RMS >0.14" and allows precise unguided imaging. For sky modelling, I am experimenting with The SkyX and T-point. The SkyX offers a special ascom 2x mount driver that treats the mount and the T-point model as one single entity.

 

The adapters were designed and CAD drawn by the inventor of the TDM on my request and are quite straightforward to make. The advantage is that the system can be fitted to existing mounts. I asked Avalon if they were interested to manufacture them as an upgrade offering but didn't get an answer yet. Alternatively, a few adapters could be made independently for those interested.

That looks very impressive. I had inquired about this in a previous thread:

https://www.cloudyni...-avalon-mounts/

I don’t quite understand what you mean, however, when you state that it has. +/- tracking error of <0.75” but an RMS<0.14. And the TDM overrides the native RA guide pulses except when dithering? Most folks get better guided results when using encoders because the encoders reduce guider requirement. This is then essentially always unguided? So if I typically get a guided RMS of 0.5”-0.6”, will my results be no better, or worse if using the TDM and getting a tracking error of 0.75”. I don’t really have any great desire to image unguided. The major reason I wanted to consider the TDM was to eliminate the significant RA drift that I see with my mount which risks losing a guide star during cloud cover. But I wouldn’t want to trade this for overall performance / results. Also since I run all my cables through the RA/polar opening, I imagine this might be a problem. Anyway, I find this very intriguing and would like to learn more about it. Also, what might be the total cost for this adaptation? I suspect Avalon would be less interested in manufacturing the adapter now that they have the M Due with encoder already onboard.

 

Derek

 

Derek


Edited by schmeah, 10 March 2021 - 06:53 AM.

  • psandelle likes this

#11 KenS

KenS

    Apollo

  • *****
  • Posts: 1,088
  • Joined: 10 Jan 2015
  • Loc: Melbourne, Australia

Posted 09 March 2021 - 09:43 PM

I think the 0.75" is a peak to peak measurement and represents 95% of the distribution which implies +/- 2 std dev. So one std dev would be around 0.19" which is the same as RMS. (not a statistician so happy to be corrected). So a slight discrepancy with the 0.14" RMS but not much. As I understand it, the TDM corrects the Avalon via the ST4 port with guide pulses. To correct the motors directly would most likely require a firmware change. But you should still be able to control the mount, including guiding, via the USB port



#12 Wiser

Wiser

    Sputnik

  • -----
  • Posts: 32
  • Joined: 10 Dec 2018
  • Loc: Utrecht, Netherlands

Posted 10 March 2021 - 03:28 AM

Hi Derek,

KenS is right, the +/- value is a peak to peak measurement, while the RMS as used in PhD2 for example is a statistical value (Root Mean Square deviation, see link for explanation).The effect of the TDM is shown in the attachment: in the left half of the picture the encoder only reads the PE, which in my M-Uno oscillates +/- 12" in long cycles but also contains high frequency noise of about 1.5". In the right half of the picture the TDM corrections are switched on and almost nullify the long cycled PE; the high frequency noise remains, even at 5 Hz corrections. Due to the continuous character of the corrections, I would expect the guided tracking to be a bit improved. However, I have limited experience with the system up to now and am in the process of further evaluation during the sparse clear nights here. Overall, the TDM has a good reputation with other brands of mounts and this is why I took the leap to embark on this project.

 

The system is fully compatible with guiding but that needs to go through the ST4 port. A full explanation, the advantages and limitations of the TDM can be found here. Since I currently image mainly with a C11-Hyperstar4 setup, the exposure times range between 10" and 120" and I don't use guiding. This may change with projects needing longer exposure times, e.g. longer focal lengths. 

 

The TDM is quite expensive in my opinion, it sells in Europe for around Euro 1400 incl. tax. The Avalon adapters will add several hundreds to that and yes, I also miss routing the cabling through the polar axis as well... But you can't have it all.

Best, Willem

Attached Thumbnails

  • IMG_3989.jpg

Edited by Wiser, 10 March 2021 - 05:44 AM.


#13 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 6,060
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 10 March 2021 - 07:11 AM

Thanks Willem for that info, and for reporting your progress on this interesting project. I will be most interested to see your results combined with long exposure guiding, because that is predominantly how I use my setup. I have gone as long as 1-2 hour exposures for fun shooting narrowband on small PNs. So at that price point it would probably make more sense for me to sell my M Uno and opt for an M Due.

 

Derek



#14 Wiser

Wiser

    Sputnik

  • -----
  • Posts: 32
  • Joined: 10 Dec 2018
  • Loc: Utrecht, Netherlands

Posted 10 March 2021 - 12:20 PM

Fair enough, Derek, I like the Due too. It is heavier though, for those interested in mobile work at dark places. Wonder if it would take a RASA 11 (19 kg).



#15 Manitu

Manitu

    Explorer 1

  • -----
  • Posts: 56
  • Joined: 04 Mar 2020
  • Loc: South Germany

Posted 12 March 2021 - 07:57 AM

They say weight is 22,4kg.

Anyone knows if that is overall weight including the T-Pod weight or it is the weight of the mount head only?

Because it doesn't look like it could be 7kg heavier than the M-uno mount head without T-Pod (which is around 15kg)



#16 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 6,060
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 12 March 2021 - 07:10 PM

Yeah, I was curious about the weight as well, as it doesn’t look much bigger than the M Uno, but I would think it it would be unusual to include the tripod in the weight statement.

 

One thing, the site now clearly states that there are high resolution absolute encoders on both RA and Dec axis.

 

Derek



#17 Jarno

Jarno

    Messenger

  • -----
  • Posts: 413
  • Joined: 22 May 2005
  • Loc: Netherlands

Posted 14 March 2021 - 06:36 AM

Yeah, I was curious about the weight as well, as it doesn’t look much bigger than the M Uno,

I superimposed the images of the M-Due and M-Uno. By using the head of the T-pod as a reference I estimate the M-Due is 25% to 35% larger than the M-Uno. Now of course weight doesn't scale linearly with size but 14.9kg + 25% already puts you at 19kg. This next bit is purely a guess on my part but I wouldn't be surprised if the absolute encoders require a stiffer construction. After all, what the point of measuring with sub-arcsecond precision if you have flex in the mount? Finally, the M-Uno weight is without counterweight and bar, this is not mentioned at the M-Due. So I find a weight of 22kg for the mount alone quite believable.

 

What I find most intriguing is the statement that the mount is Raspberry Pi based. Imagine if it were possible to run something like the Asiair software from that, it would (in my opinion) be the best mobile astrophotography solution ever produced. As in, "shut up and take my money" - good. bounce.gifdrool5.gif

 

Jarno


  • schmeah likes this

#18 Gleason

Gleason

    Ranger 4

  • *****
  • Posts: 300
  • Joined: 03 Jan 2013
  • Loc: SF Bay Area

Posted 16 March 2021 - 02:52 PM

50 lb head weight is a bit much for portable. Doable, but portable?  I came across this mount today and am very interested in single arm fork to hold a pair of FSQ's as a dual astrograph for survey work.  Because I do wide field survey work, the elimination of mount flip is a feature.  Still, I wonder about dampening in wind when considering that long RA arm.  Can anyone with an M-UNO comment?  The DUO appears larger by design.  

 

-j



#19 schmeah

schmeah

    Fly Me to the Moon

  • *****
  • topic starter
  • Posts: 6,060
  • Joined: 26 Jul 2005
  • Loc: Morristown, NJ

Posted 16 March 2021 - 09:40 PM

50 lb head weight is a bit much for portable. Doable, but portable?  I came across this mount today and am very interested in single arm fork to hold a pair of FSQ's as a dual astrograph for survey work.  Because I do wide field survey work, the elimination of mount flip is a feature.  Still, I wonder about dampening in wind when considering that long RA arm.  Can anyone with an M-UNO comment?  The DUO appears larger by design.  

 

-j

Over the years I’ve seen many people raise concerns over the wind sensitivity of Avalon mounts, though more often than not by folks that don’t own the mount. And the concerns relates more to the flex/compliance of the belts rather than the long RA arm. As I usually image at long FL and with subpar seeing to begin with, I typically don’t image in wind more than 10 MPH. On the few nights that I have though, even when the guide graph looks more like a seismograph, the stars have still been tighter than expected. I suspect that because of the absence of backlash and the excellent guide pulse responsiveness of this mount that even when knocked around by the wind the guide star snaps quickly back on pixel. Admittedly I have nothing to compare it to, but suspect the concern is exaggerated.

 

Derek



#20 Jarno

Jarno

    Messenger

  • -----
  • Posts: 413
  • Joined: 22 May 2005
  • Loc: Netherlands

Posted 17 March 2021 - 05:30 AM

50 lb head weight is a bit much for portable. Doable, but portable? 

I don't think it's that much, particularly considering the payload. It's about the same as the CGX but of course that doesn't include the CGX's counterweights. All in all the weight is about the same as my Linear including its counterweights and I find that quite transportable.

 

As for wind sensitivity, Derek is absolutely right. The belts aren't simple rubber bands but are made of high tensile strength material. Before the Linear I owned a Losmandy G-11 and that didn't fare any better in strong winds. Even if your mount is absolutely rigid then you still can (and probably will) have play between dovetail and saddle, between rings and OTA, maybe flex in the OTA and perhaps the focuser. But it'll be interesting to see if the encoders provide faster corrections to wind gusts.

 

Jarno


  • schmeah likes this

#21 Wiser

Wiser

    Sputnik

  • -----
  • Posts: 32
  • Joined: 10 Dec 2018
  • Loc: Utrecht, Netherlands

Posted 17 March 2021 - 06:43 AM

In my experience, the wind sensitivity depends on the payload and length of the tube. With my C11, I prefer it under 5-6 mph unprotected. With my RASA 8 more wind is acceptable. I have succesfully tested a c6 at speeds > 10 mph. When at home, I work from a balkony at 50 m height that is unprotected from the east, that's why I experimented a lot. Previously, I owned a well tuned CPC1100 and the wind sensitivity was very comparable to the Avalon Uno. The CPC series is worm driven.

 

Regarding the Due's HiRes encoders, I asked Avalon and got word that the technology is expected to be able to perform a better tracking and a better pointing accuracy as a first step. More features, like sky modeling and unguided imaging, can be available later. So those interested in unguided imaging may have to wait.


  • schmeah likes this

#22 Manitu

Manitu

    Explorer 1

  • -----
  • Posts: 56
  • Joined: 04 Mar 2020
  • Loc: South Germany

Posted 21 March 2021 - 10:59 AM

Hi Willem

 

I wonder why Avalon says there will be an option for unguided imaging later on.

You upgraded your M-Un with absolut encoders and you are able to do unguided imaging up to 120 seconds.

 

I would expect the same with the M-Due. 

With absolut encoders for both axis it should be possible to achieve the same results like you did with the M-Uno. Without any further upgrades like sky modeling.

 

What do you think about?

 

CS Tom



#23 psandelle

psandelle

    Soyuz

  • *****
  • Posts: 3,634
  • Joined: 18 Jun 2008
  • Loc: Los Angeles

Posted 21 March 2021 - 11:45 AM

Hi Willem

 

I wonder why Avalon says there will be an option for unguided imaging later on.

You upgraded your M-Un with absolut encoders and you are able to do unguided imaging up to 120 seconds.

 

I would expect the same with the M-Due. 

With absolut encoders for both axis it should be possible to achieve the same results like you did with the M-Uno. Without any further upgrades like sky modeling.

 

What do you think about?

 

CS Tom

I believe Avalon is talking about incorporating sky-modelling in X-Solver to use with the M-Due, so, like with 10Microns and the AP Mach 2, one can get 5 to 10 minute unguided without throwing any subs out.

 

Paul



#24 Wiser

Wiser

    Sputnik

  • -----
  • Posts: 32
  • Joined: 10 Dec 2018
  • Loc: Utrecht, Netherlands

Posted 21 March 2021 - 01:33 PM

Hi Tom

Avalon's answer to my inquiry was cautious. They didn't say sky modelling and/or unguided tracking will become available, they said it 'can be available later', in other words, it might be added later.

 

As I understand, the Due's high res encoders will provide improved pointing accuracy. Being absolute versions, the usual home synchronisation after startup will probably not be needed. Unguided tracking is a very different encoder application, requiring ultra high resolution, continuous closed loop rotational feedback and correction of the RA axis. The Due's specs don't currently disclose if the encoders have sufficient resolution for this purpose. It doesn't matter if the encoder is absolute or relative for such RA tracking, as long as it is extremely precise. The software controlling the feedback loop needs to be sophisticated, especially if the option of guiding is kept possible. Guiding at relativley long intervals (30" for example) may help to correct for some polar misalignment at longer focal lengths. It is my understanding that the iOptron EC controlled mount versions don't allow guiding for example when the EC is switched on, at least they didn't in the past. BTW, sky modelling may be offered anyway, even without the option of unguided tracking, similar to TSX T-point. The latter can build a sky model for any Ascom mount but only SB's own mounts are compatible with the build-in algorithm (ProTrack) to improve precise unguided tracking. 

 

The TDM Drivemaster is build around a very precise Heidenhahn encoder, resolution 0.125". It doesn't provide any additional pointing information, but secures the closed loop RA tracking feedback up to 5 times per second.This works very well for the Avalon Uno and enables unguided tracking much longer than 120", but I haven't come to explore limits. The TDM allows guiding as well, which I find attractive as an option for future projects. 

 

Best, Willem



#25 JethroXP

JethroXP

    Mariner 2

  • -----
  • Posts: 237
  • Joined: 12 Aug 2020
  • Loc: Duvall, WA, USA

Posted 22 March 2021 - 01:04 PM

Avalon have updated the description to indicate that the mount has Absolute Encoders on both RA and Dec Axis.  Also, in a FB thread on their official page they indicated that these are 21-bit encoders.  Not as high precision as others out there (21-bit is only 2M unique positions, which would give a resolution of about 0.62 arcseconds) but certainly better than none at all, and being absolute encoders means no sub-divisional error (SDE) that can occur with relative encoders.




CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics