Atik Cameras Forum

Support (Please note, this is primarily a user forum, for direct Atik support, please email support@atik-cameras.com) => 3- and 4-Series Cameras => Topic started by: GH-ATIK on January 03, 2016, 04:44:05 pm

Title: Atik383-L+ strange bias
Post by: GH-ATIK on January 03, 2016, 04:44:05 pm
Hi there,
in the last time. I got a strange bias-pattern ( see picture ) , the pattern ist observed in the final-picture. Is there anybody with a solution of this problem ( the bias is an average of 50 pics, -15°C )

Thanks a lot

Gerhard
Title: Re: Atik383-L+ strange bias
Post by: Max on January 04, 2016, 05:50:58 pm
Hello Gerhard,

my Atik383 has the same behavior.
Please have a look to the picture (master Bias: 50 frames, binning 1x1, -25°).
SW: TheSkyX Camera Add on, CCDStack.

Thanks
Max

Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 04, 2016, 07:29:51 pm
dear Max,

thanks a lot, I have two cams, one of them shows no interferrences, the other one showes this effect. unfortunately , these waves at the left side beats by the final picture.
I try out pixinsight and deep sky stacker for calibration, I think, there ist no mistake in the workflow.
Max, do you see this effect in your final picture, or ist this effect corrected with picturecalibration ?

Thanks a lot

Gerhard
Title: Re: Atik383-L+ strange bias
Post by: Max on January 04, 2016, 07:51:15 pm
Hello Gerhard,

yes – and it’s terrible, especially in the OIII images there is the effect strong disturbing.

Please have a look to the picuture.
All work were done with CCDStack – to be sure that there are no side effects.

The effect remains me on SBIG ST11k the banding problem.
There is a special SW for the SBIG banding effect: CCDBand-Aid.

Thanks
Max
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 05, 2016, 07:35:45 am
Hi there,

Yes, I have the same effect. the question is : is the cam defect or were a fault in workflow responsable ?
It seems, there is a problem in the electronicdevice during reading the electons from chip , or the chip is in incorrect condition.

Question : is the cam a case for atik-service ?

Thanks, Gerhard

Title: Re: Atik383-L+ strange bias
Post by: Max on January 05, 2016, 09:56:44 pm
Hi,

I made an experiment in four steps.

All Darks (40 x 900s) and Bias frames taken one with -25°C.
The Dark Master has the same behavior as the Bias Master (banding on the left side).

All light frames are also from one night taken with -25°C. (02/03-November-2015)

First Step:
The Dark Master has the same behavior as the Bias Master. So let’s do some pixel math.

Please have a look to the picture. DarkMaster-BiasMaster.jpg
PixelMathe in CCDStack: MasterDark – BiasMaster -> it looks got, no banding.

Conclusion:
Bias and Master should not be the problem (by this example).
Title: Re: Atik383-L+ strange bias
Post by: Max on January 05, 2016, 09:57:49 pm
Second Step:
CCDStack work: Calibrate (Darks, Bias, Flats), Register, Normalize, Data Reject and Stack.

Third Step:
Stack all frames from the night (43 x 900s frames)

Please have a look to the picture: Mean OIII3nm_allFrames-43x900s_.jpg
Funny: When I take all frames from the night – then I see the banding on the right side of the picture.
Title: Re: Atik383-L+ strange bias
Post by: Max on January 05, 2016, 09:58:50 pm
Fourth Step:
Stack only with the frames before the meridian flip (24 x 900s).
Please have a look to the picture: Mean OIII3nm_beforFlip_24x900s-Frames_.jpg
Funny: When I take only the pictures before the meridian flip – its looks fine, no banding!

Conclusion:
So it seems (by this example) it’s a problem caused from the light frame. I really don’t know why the key factor is the meridian Flip (by this example).

Maybe: It could be the moon. It was half-moon, that means after midnight (and so also after the meridian flip) the light from the moon was disturbing the frames.

Thanks
Max
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 06, 2016, 09:17:10 am
Hi Max,

thanks, good idea, I tryed out a lot of combination in Pixinsight, but no chance. I think, I know the reason why : I performed my lights with dithering, and so, all frames were a bit moved and there is no way to get a congruent stack , no wonder for interferrences.
In future, I try out a session without dithering.

Thanks a lot , Gerhard
Title: Re: Atik383-L+ strange bias
Post by: Max on January 06, 2016, 06:06:37 pm
Hi Gerhard,

I performed also all my lights with dithering.
Today I took a closer look to my lights – the behavior is not because of the meridian flip.

The banding in the bias frames are exactly the same as in the darks.
It seems that the banding in the light frames is changing over the night.

Thanks a lot,
Max
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 06, 2016, 06:25:32 pm
Hi Max,

than it seems difficulter, as expected. If the banding is not stable, it is nessesary to take in every session, a new block of bias.

I try to observe this behavoir, and in the meanwhile , I think, the cam ist not ok.

Obviosly, we are now the only one, with these effect. Perhaps, there are some readers, who judge, if this behavoir is for some cameras normal or not.

In the moment, I´m not satisfied, when the banding moves every day.

Thanks a lot, Gerhard

Title: Re: Atik383-L+ strange bias
Post by: Max on January 06, 2016, 10:58:55 pm
Hi Gerhard,

How old is your Atik with the banding problem and how old  is the other one without banding?
And what about the power supply?

I bought my Atik in December 2011.

Thanks a lot
Max
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 07, 2016, 08:54:16 am
Hi Max,

The first one was bought in march 2011 , and was in service in summer 2014 , because the sensitivity was under 30% , the mainboard was substituted.
The new one was bougt in februrary 2014 , there is a lttle banding visible.

I use a sabilisized powersopply at 13,5 Volt.

Today I tryout a 12V DC-car-battery , suprised what happend .....

best regards Gerhard
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 07, 2016, 12:34:28 pm
Hi Max,

The last information was wrong :

It was a stabilized power-supply , adjusted to 12,5 V

here the to-day-old data :  n=50  1x1bin  -15°C
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 07, 2016, 12:35:21 pm
the same today n=20 :
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 07, 2016, 12:36:12 pm
n=20 1x1 bin 13,0 V DC
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 07, 2016, 12:37:31 pm
and the last : n=20  1x1 bin  12,5V DC  from a car-battery
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 07, 2016, 12:39:46 pm
it seems, that the voltage and the stabilyzing has an influence .

greetings Gerhard
Title: Re: Atik383-L+ strange bias
Post by: Max on January 07, 2016, 10:04:01 pm
Hi Gerhard,

there should be no problem with the bias frame – when there is the same behavior within the dark frames.

Please try this (easy to do in CCDStack):

1.) Open the DarkMaster
2.) Open the BiasMaster
3.) Take care that DarkMaster is selectet in CCDStak
3.) Process -> File Math -> Substract -> BiasMaster…
4.) Now the DarkMaster shout be clear – that means there is no Bias/Master banding problem.

Maybe:
The problem could be caused by the light frames – regarding the cooling unit of the early 383er.
The next days I will check other pictures.

Thanks a lot
Max
Title: Re: Atik383-L+ strange bias
Post by: Max on January 07, 2016, 10:16:28 pm
Hi Gehrhard,

An idea:
Tomorrow I will check the behavior of the fan.
1a) Fan speed with less colling.
2a) Fan speed with more colling.
3a) Fan speed with max colling.

1b) Fan speed with 12V.
2b) Fan speed with 12.8V.
3b) Fan speed with 13.8V.

When the fan speed is not constant than there is a different cooling for the electronic parts.

Thanks a lot
Max
Title: Re: Atik383-L+ strange bias
Post by: Max on January 08, 2016, 07:22:10 pm
Hi Gerhard,

I've done the tests.

Please have a look to the picture.
It was not so easy to find an analog volt- and  ammeter to see the rapid changes.
I have no oscilloscope at home…
The cable from the power supply to the ccd is only 1,5m (59inch) – to take care of the voltage drop.
The longer cable in the picture is only for the voltage measurement- it does not matter.
SW: TheSkyX with camera add on.

Tests 1a, 2a, 3a (power supply with 13V):
When the cooling is low (20% -> 0.7A) – than there is also less voltage drop (because of the current): 0.2V
When the cooling is medium (50% -> 1.2A) – than there is also medium voltage drop: 0.35V
When the cooling is high (75% -> 1.6A) – than there is more voltage drop: 0.5V
The speed of the fan is changes depending on the voltage drop.
When fan “coughs” than there was a fast change of the current – the regulation. (and there a lot of fast changes).
 
Tests 1b, 2b and 3b:
When the voltage is low - than the speed of the fan is also low.
When the voltage is higher - than the speed of the fan is also higher.
 
Conclusion:
There is no voltage regulation for the fan.
During the photo session the temperature of the night is changing.
Because of the change of the temperature the amount of current for the cooling unit is also changing.
This causes – because of the voltage drop to the camera plug – that the speed of the fan is also changing.
Maybe this effect could be the reason for the different banding behavior over the night.

I could make the measurements in the company with professional equipment – but I think it’s not necessary.

Thanks a lot,
Max
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 09, 2016, 07:30:41 am
Hi Max,


Well, a good idea, but only a theroretical consideration invalidated this unproven thesis :
the "aperture time" of bias were as short as possible, and temperaturfluctuation ( better voltagefluctuation ) in this short time,
is excluded. the whole bias-session as well.
but : the effect is , indeed , voltage-dependent, howsoever.

I will test grownding of the powersupply and the cam, today.

thanks a lot

Gerhard
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 09, 2016, 10:39:23 am
So, there is a solution :

I changed the powersupply , which is build for amateur radio with noise-reduction.
SPS 250-II
I measured the ground on minus, there was a connection.

on the other powersupply, there was no connection from ground to minus !

Ok, its not clear , if the voltage of 13,8V, the missing ground or noise-reduction was respunsible for the wrong bias,
but this result is now ok :


Thanks a lot

Gerhard

Title: Re: Atik383-L+ strange bias
Post by: Max on January 09, 2016, 08:02:54 pm
Hi Gerhard,

respect and congratulations!
I will also have a look to another power supply.

Thanks a lot
Max
Title: Re: Atik383-L+ strange bias
Post by: Max on January 09, 2016, 11:25:09 pm
Hi Gerhard,

I've looked at the specification of the Maas-SPS-200-II power supply: 13.8V.

Please have a look to the picture – also cleaner bias with my regulated power supply at 13.8V and grounding.
I have taken the bias at home – therefore only with -15°C.
That means: For clean bias 13.8V (in the range) are necessary.
As Atik said: No voltage over 13.8V - it should be OK.

Thanks a lot,
Max
Title: Re: Atik383-L+ strange bias
Post by: GH-ATIK on January 10, 2016, 02:41:51 pm
Hi Max ,


perfect win-win-situation  ;)

Thank´s a lot

Gerhard