StackPro subexposure won't set to 1 second

On my new AC4040 (DLAPI 2.7.1.0, firmware 14), I cannot set the Stacking Sub-exposure in DL Config to 1 second. The Aluma AC manual (v.1.06, page 22) says that 1 second is possible, but no matter what I do, my DL Config refuses set a subexposure duration of less than 2 seconds. I can set 2 seconds and 10 seconds, but not 1 second.

For flats, I would really like to have StackPro subexposures of 1 second (or shorter if you can manage it!), as my flats are saturated at 2 seconds unless my flat panel is set to incredibly faint. Any suggestions?

 

Perhaps. A good idea. In fact, in High Gain I may as well take flats shorter than 1 second to keep the flat panel brighter.

To that end: do you know approximately the shortest exposure time the AC 4040's shutter will allow (in High Gain)? I would verify this experimentally for my own camera unit, but a starting point would be useful.

(On my earlier question: is the minimum StackPro subexposure actually supposed to be 1 second or 2 seconds? The answer is still useful for bright objects.)

 

All good to know. Thanks, Colin.

If taking flats much shorter than 1-2 seconds turns out to be advantageous, I'll report back here.

 

OK, I have my first working flat images on the new system.

OTA: PlaneWave CDK 20 f/6.8, no reducer.
Flat panel = Alnitak Flat-Man 24.
Camera = AC4040M front-illuminated, operated in High Gain mode and cooled to -15C.
Filter: Clear, of size 50x50 mm (Baader), in AFW-10-50SQ filter wheel.​

With panel at 80 (of 255), an exposure of 0.1 seconds gives 2500 ADUs at image center. (Panel brightness of 80 is about as low as I trust the flat panel to remain, well, flat.) Even the Sloan i' filter needs only 0.25 seconds for a flat of 2500 ADU at center (panel at max brightness).

Vignetting at corners is only about 18% from center (with all filters, all 50x50 mm square). A very good result.

Unfortunately, this is where 12-bit native hurts: to keep flat-masters from substantially degrading the signal-to-noise of my working photometry images (StackPro), I may have to take and average a lot of flat images, perhaps hundreds in each filter. Sigh. At least they get taken quickly.

One remaining question: With such short exposure times, all much less than the minimum subexposure for StackPro...is there any reason to use High Gain rather than High Gain StackPro? That is, is there any real difference in the images taken with High Gain mode vs. with High Gain StackPro of only one subexposure? It would be nice not to have to program mode changes to take flats and flat darks.

Thanks for the guidance. These initial operating conditions and results were not intuitive, but in the main they are encouraging.

 

Hi Eric.

FWIW….

Re LED flat panel brightness.

Many LED panels use PWM (pulse width modulators) to vary the brightness and at minimum brightness settings the LEDs are pulsed off for longer than they are pulsed on leaving the output light to vary or flicker at sub second frequencies.

If you use sub second exposure times when taking flats and low brightness settings for the panel then there can be considerable pixel ADU variation sub-to-sub between frames.

Luckily, there is a low cost solution for such panels and photographic studios, theatres, dance venues etc have been using lighting gels for years to modify and control light quality.

Specifically, neutral density polyester film gels placed over the panel allows the brightness control to be set high or full to avoid panel flicker while cutting the transmitted light levels considerably.

I use Rosco eColor neutral density polyester gel sheets in front of my LED panel to achieve the necessary brightness reduction while leaving the brightness control set to maximum. Depending on the brightness of your panel just one or two sheets of neutral density film is sufficient to reduce the output brightness to give you sensible exposure times for any camera acquisition mode.

Rosco eColor Neutral Density lighting gel sheets are very uniform, surface coated polyester film, quite tough, but can be scratched if mis-handled and are best used when sandwiched between two sheets of clear or opalescent polycarbonate or glass to give protection to the surface coating of the filter when used in the observatory.

The E.299. 1.2 Neutral Density lighting gel filter for example passes only 13% light while the E.298 .15 ND lighting gel filter passes 69% light. There are several other ND reduction filters in the range so its quite easy to find a single sheet suitable for your camera filters while leaving the panel brightness set to maximum for stability and flicker reduction.

The Rosco ND filter sheets are supplied in 1.2m by o.53m single sheets, large enough to cover most LED panels used in amateur observatories in a single sheet, cost is around 8 US dollars per sheet and are usually despatched rolled up in a postal tube with tissue paper on the coated side for protection during transport.

Check the Rosco website for more details on the eColor lighting filters and links to retailers:

https://emea.rosco.com/en/product/e-colour

HTH.

William.

 

Thanks, William, I'll certainly keep this in mind. Probably I'll first test the reproducibility of a series of flat exposures shorter than I'll need, perhaps of ~0.05 seconds. If the flat panel PWM runs at 1 kHz or faster, that shouldn't present a problem. We'll see.

 

So far (though I'm fighting a camera-disconnect problem when taking images at very high cadence, as with flats), my flats have indeed been reproducible down to 0.05 sec. This is with an Alnitak Flat-man 24 and the AC 4040M.

To keep the number of flat darks manageable, I'm planning on standardizing my flat exposure times to ~ 0.10 seconds, except 0.25 seconds for Sloan i'.

 

It is not Selective Suspend, it is not Power Management. Checked long ago, and regularly since.

My first USB problem is that my mount is too large to allow a USB 3 camera-PC cable as short as 10 feet. It absolutely has to be at least 12 feet, which is longer than the 3-meter spec. So I have installed a high-end USB 3 active repeater cable, but I may need to experiment with other such remedies. Not easy from 240 miles away.

 

Last night's log is attached--it shows five rogue disconnects, most of which I happened to witness (remotely). They all happened during rapid imaging cadences of one kind or another.

Disconnect #1 at 02:41:45: during series of panel flats (very short exposure times, rapid cadence)
Disconnect #2 at 02:42:47: during same series of panel flats.
Disconnect #3 at 02:43:28: during same series of panel flats.
Disconnect #4 at 02:54:15: during same series of panel flats.
Disconnect #5 at 07:08:43: during FocusMax autofocusing (also very rapid cadence)
(The disconnect at 10:50:11 was the scheduled end-of-night shutdown.)

• The visible error message, from MaxIm, was something like "Unexpected disconnection (-4)", if memory serves.
  
• The cooler was running at 60% early in the night, but the last disconnect happened at only ~45%, so I doubt that voltage drop is the problem (I'm using the power supply that came with the camera).
  
• It's possible that the laptop is not putting out enough power via USB for the active cable; I'm ordering an externally powered active cable to try.
• There is also some question about the USB 3 drivers in this 2015 (Intel i7) laptop, and the New Mexico Skies staff are trying a couple of things tonight. I am prepared to replace the scope PC if that's the problem.
  
• But if you see something I've not perceived in the log, for sure let me know.

One oddity: I'm not sure why (e.g., 01:58:45) the Guider is said to be running an algorithm (?) and calibrating when I'm not using the guider at all. It was powered on, but never used by me or by ACP.

Thanks for having a look.

 

Thanks. The repeater is a Tripp-Lite, very unlikely to be a crappy one, but it is USB powered. And I've used only industrial StarTech USB hubs in my observatory, for many years now already. They have never failed on me, not once, regardless of weather etc.

For now, I will try the USB 2 cable as a temporary fix. Much better to have slower downloads than disconnects. As a more permanent solution, I will be purchasing an externally-powered StarTech extender.

Thanks for the tips. Onward.

 

All ACP.