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

ZWO EAF backlash

  • Please log in to reply
12 replies to this topic

#1 Narrowbandpaul

Narrowbandpaul

    Mariner 2

  • -----
  • topic starter
  • Posts: 287
  • Joined: 24 Dec 2013
  • Loc: Constrained by the Heisenberg uncertainty principle

Posted 27 October 2020 - 08:18 PM

Hi All

I was doing a visual check of the backlash of a ZWO EAF connected to a 3.5” FT. When I change directions it seems like the backlash is around 130 steps. This is on an esprit 150. Doing some measurements it seems like we’re 3.5um per step, so a backlash of around 0.4mm.

Does this sound typical. We don’t get the prettiest v curves but they do work.

Any tips for using the ZWO or can we reduce this backlash? Do you focus against gravity or with it?

Thanks

#2 Eric Horton

Eric Horton

    Mariner 2

  • *****
  • Posts: 259
  • Joined: 04 Dec 2011

Posted 27 October 2020 - 11:02 PM

Yes. High backlash is common on eaf. I have one on my newt. I think I'm around 100 to 120. Once set I've never had an issue with it. Just something to be aware of

#3 Narrowbandpaul

Narrowbandpaul

    Mariner 2

  • -----
  • topic starter
  • Posts: 287
  • Joined: 24 Dec 2013
  • Loc: Constrained by the Heisenberg uncertainty principle

Posted 28 October 2020 - 07:59 AM

Thanks Eric

Any way to reduce the backlash? Have you set the backlash value in the EAF driver?

How do your v curves look?

Thanks

#4 StephenW

StephenW

    Apollo

  • -----
  • Posts: 1,266
  • Joined: 04 Feb 2005
  • Loc: San Jose, CA

Posted 28 October 2020 - 09:23 AM

>Any way to reduce the backlash?

 

Shouldn't be necessary, as long as you set sufficient backlash compensation in your focusing s/w.

 

As an example, even if you had massive backlash (say 10s worth!), it would just mean your focusing routine is 10s longer each time it changes direction (usually twice per focusing run).


  • dswtan likes this

#5 Narrowbandpaul

Narrowbandpaul

    Mariner 2

  • -----
  • topic starter
  • Posts: 287
  • Joined: 24 Dec 2013
  • Loc: Constrained by the Heisenberg uncertainty principle

Posted 28 October 2020 - 10:07 AM

Thanks Stephen

So the backlash value so be a little larger than the true value, or should we try to measure precisely

#6 Narrowbandpaul

Narrowbandpaul

    Mariner 2

  • -----
  • topic starter
  • Posts: 287
  • Joined: 24 Dec 2013
  • Loc: Constrained by the Heisenberg uncertainty principle

Posted 28 October 2020 - 10:12 AM

What other focus motors are recommended?

#7 StephenW

StephenW

    Apollo

  • -----
  • Posts: 1,266
  • Joined: 04 Feb 2005
  • Loc: San Jose, CA

Posted 28 October 2020 - 10:19 AM

>So the backlash value so be a little larger than the true value

 

Exactly - as long as your backlash compensation is bigger than your actual backlash, you'll be fine.  There is no need to measure it precisely.


  • dswtan likes this

#8 Eric Horton

Eric Horton

    Mariner 2

  • *****
  • Posts: 259
  • Joined: 04 Dec 2011

Posted 28 October 2020 - 10:30 AM

Thanks Eric

Any way to reduce the backlash? Have you set the backlash value in the EAF driver?

How do your v curves look?

Thanks

Yes, I enter the backlash compensation.  Once i did that, the v-curve look normal.  Or normal for my seeing conditions.  I don't know or have found a way to reduce the backlash.



#9 Der_Pit

Der_Pit

    Surveyor 1

  • *****
  • Posts: 1,661
  • Joined: 07 Jul 2018
  • Loc: La Palma

Posted 28 October 2020 - 10:44 AM

>So the backlash value so be a little larger than the true value

 

Exactly - as long as your backlash compensation is bigger than your actual backlash, you'll be fine.  There is no need to measure it precisely.

Careful with that!  There are two different ways of BL correction.  One is to 'ignore' it, and make sure any focus change ends moving 'inward' (i.e., against gravity).  So if you want to focus X steps out, you actually first move X+Y steps out, then Y in again.  There, it doesn't matter how large Y is exactly, other than that it needs to be larger than the actual BL.

 

The BL setting in the driver works different, though.  There, if you change the direction of the movement, an additional distance Y is traveled, and then the counter is changed.  There, you need to enter the correct BL value exactly, else your numbers (and thus also the V-curve) are wrong....


  • NightBear likes this

#10 ryanha

ryanha

    Messenger

  • -----
  • Posts: 443
  • Joined: 05 Aug 2020

Posted 28 October 2020 - 10:58 AM

I was doing a visual check of the backlash of a ZWO EAF connected to a 3.5” FT. When I change directions it seems like the backlash is around 130 steps. This is on an esprit 150. Doing some measurements it seems like we’re 3.5um per step, so a backlash of around 0.4mm.

Does this sound typical. We don’t get the prettiest v curves but they do work.

Any tips for using the ZWO or can we reduce this backlash? Do you focus against gravity or with it?

Thanks

 

If this is a new unit I would suggest reaching out to the vendor and asking them the specs on how much backlash tolerance they warranty.  130 steps sounds like a lot.  I have heard others talk about these units being hit and miss wrt backlash.  Maybe you got a bad one.  Sure you can tune it out to an extent, but if you had the option of returning it for a tighter one, would you do it?

 

 

--Ryan



#11 NightBear

NightBear

    Vostok 1

  • *****
  • Posts: 149
  • Joined: 12 Nov 2019

Posted 28 October 2020 - 11:49 AM

I have asked ZWO about this recently and they said this amount of backlash is normal and just make sure your grub screws are locked down tight. The focuser works great for me, using backlash overshoot in NINA where knowing the precise number of steps is not needed. I previously tried to carefully measure the exact backlash using a dial gauge but that was a lot of trouble and the actual backlash seems to change depending on other conditions anyway (temperature? load? orientation?). Trying to nail down the exact value to use in backlash compensation produced inconsistent results.


  • dswtan likes this

#12 dswtan

dswtan

    Ranger 4

  • *****
  • Posts: 357
  • Joined: 29 Oct 2006
  • Loc: Morgan Hill, CA

Posted 28 October 2020 - 02:11 PM

I'm in the camp where "just enter a decent-sized amount" in NINA (can't speak for other apps), and all good -- @Nightbear, you have an SV80A with your EAF too? My same setup just needed a  "random" setting of 100 in Auto Focus Step Size plus Backlash Compensation = Overhsoot, and I get decent "V" curves (more U, but ok).


  • moonrider likes this

#13 NightBear

NightBear

    Vostok 1

  • *****
  • Posts: 149
  • Joined: 12 Nov 2019

Posted 29 October 2020 - 08:48 AM

I'm in the camp where "just enter a decent-sized amount" in NINA (can't speak for other apps), and all good -- @Nightbear, you have an SV80A with your EAF too? My same setup just needed a  "random" setting of 100 in Auto Focus Step Size plus Backlash Compensation = Overhsoot, and I get decent "V" curves (more U, but ok).

From what I understand, the autofocus step size should be 1-1.5x the critical focus zone of your optics system. See calculator here: http://www.wilmslowa...ormulae.htm#CFZ . Based on my setup that comes out to 20-30 steps and I use 5-6 steps for the autofocus routine. I get perfect results every time.




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