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

ROI and FPS for ASI455 vs ASI533, also Newton's rings

  • Please log in to reply
2 replies to this topic

#1 Coconuts

Coconuts

    Viking 1

  • *****
  • topic starter
  • Posts: 742
  • Joined: 23 Sep 2012

Posted 11 August 2022 - 06:12 AM

Hi.  I enjoy solar observing, but will be new to solar imaging.  I have a Lunt Solar LS80THa that is double stacked, with a 12 mm blocking filter, a Televue 2X PowerMate, and an ASI6200MM camera which is overkill from a sensor size perspective but is great for deep sky imaging.  The LS80THa is f/7, so after the 2X PowerMate I would be critically sampled for the ASI6200MM's 3.76 um pixels at 656 nm and f/14.  So all good, but...

 

The solar full disc image will be 10.4 mm diameter, so it plus prominences would need ~2900 x 2900 of those 3.76 um pixels.  I have Firecapture, but am currently at a different property than the camera and scope, and so can't test the supported ROI flexibility.  The ASI spec sheet only mentions rectangular ROIs, and the only one of those exceeding 2900 in its shortest dimension is 6400×4096, which is slow at 7.44fps.  Does Firecapture support arbitrary square ROIs on the ASI6200MM, such that I could dial in 2900 x 2900, and if so, what would the fps be?  The smaller sensor ASI533MM has 3008 x 3008 pixels, pretty close, also 3.76 um, and it claims a full resolution frame rate of 19.88 fps.  Would the smaller ASI533MM actually provide a higher FPS than an ROI on the ASI6200MM, and if so, by how much?

 

None of these FPS would be anywhere near the 100 Hz range for a full solar disc, and I get that higher FPS improves the odds for lucky imaging, but 12 mm sensors at high frame rates are pricey; one such option would be a 25 GigE "Bolt" camera from Emergent Vision Technologies with the Sony IMX531 sensor (2.74 um pixels, 4504 x 4504 pixels, 12.3 mm square, up to 100 fps at full resolution, 112 fps @ 4000 x 4000; link below), but that seems like overkill and expensive, also a bit oversampled at f/14 (4.1X instead of 3X). 

https://emergentvisi...es/hb-20000-sb/

 

How much of a compromise would 10 to 20 FPS be, if I was ready to discard capture frames or sequences that had bad seeing in them?  Wouldn't I eventually "get lucky" even at these lower frame rates?

 

Last bit: I have read snippets here that adding a Barlow (so maybe also the telecentric PowerMates) will result in Newton's rings.  Is that true, and if so, what are my options?  Thanks in advance!

 

All the best,

 

Kevin


Edited by Coconuts, 11 August 2022 - 06:23 AM.


#2 MalVeauX

MalVeauX

    Voyager 1

  • *****
  • Posts: 10,822
  • Joined: 25 Feb 2016
  • Loc: Florida

Posted 11 August 2022 - 06:29 AM

Hi,

 

It's hard to guess at the potential FPS because it greatly depends on how the camera is configured for read out, the buffer size and memory speed and how you configure memory handling. Most of the really big sensors with huge arrays do not get much faster FPS with region of interest. I would not expect a huge increase in speed. It will always be limited to the USB3 interface's limits and mainly your bus speed.

 

20~30 FPS is fine for full discs. You can go for 30~60 seconds (if there's lots of activity maybe stay closer to 30 seconds) of total time per run before running into big problems with apparent motion on objects.

 

FireCapture allows you to drag a box over the FOV and make that your ROI without using a preset size/shape.

 

A barlow doesn't add newtonian rings other than adding more optics and potential pieces in the imaging train that can have tilt, which will make the light cone not orthogonal to the sensor. However, the main culprit of newtonian rings is usually the glass plate on the sensor itself and tilt and not the other optics (like a barlow). You can use any barlow you want. You'd be hard pressed to see a difference in any of them relative to your seeing.

 

I'd love one of those GigE interface cards, but overall I'm not fooling with them due to cost. I do my full discs around 30 FPS and my high res work between 160~170 FPS and am satisfied with that so far. I'll wait for the next interface change in USB and cameras to not have to get a specific PCIe card just to get access to those cameras (all of which are costly).

 

Very best,


  • rigel123 likes this

#3 Coconuts

Coconuts

    Viking 1

  • *****
  • topic starter
  • Posts: 742
  • Joined: 23 Sep 2012

Posted 11 August 2022 - 08:17 PM

Marty:  Thanks.  It turns out that FPS does go up a lot with large sensors as the ROI shrinks, as shown in the ASI6200MM specs below, but I am pretty sure that shrinking rows and columns aren't equal when it comes to faster fps, as the example below of 4096 x 2160 vs 3840 x 2160 shows: identical fps at 14.35.  But other ROIs show that it isn't a simple inverse second number.  I suppose I can wait until I get back to that property, and run FC with the camera to confirm if it supports arbitrary drawn ROIs with ZWO cameras, and what the fps will be at 2900 x 2900.

 

Max FPS at full resolution:
16Bit ADC
9576×6388 3.19fps
7680×4800 5.29fps
6400×4096 7.44fps
4096×2160 14.35fps
3840×2160 14.35fps
1920×1080 28.04fps
1280×720 41.12fps
640×480 59.67fps
320×240 108.72fps
more resolutions are in software, support customize resolution.

 

Very encouraging to hear that full disc lucky imaging can still be done at moderate fps.  The ASI533MM may well be faster at 2900 x 2900 px, but I doubt if the delta in speed is worth buying a second camera. 

 

I remember hearing somewhere that for less than full disc, there was a new smaller format, fast, small pixel, mono sensor/camera coming, but can't recall what that was.

 

It's also good to hear that a PowerMate (or Barlow) shouldn't introduce Newton's rings, but you indicate that the sensor glass could be an issue.  On the ASI6200MM, this glass is AR coated, something that isn't done with the RGB versions, which simply use an IR cut glass.  So maybe the ASI6200MM won't exhibit Newton's rings.

 

All the best,

 

Kevin


  • MalVeauX likes this


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