Svbony SV505C astronomical camera test

Svbony company provided me with a Svbony SV505C color astronomical camera based on a Sony imx464 sensor for testing. This is an interesting sensor, which is a follower of the famous Sony imx462 with high sensitivity to near infrared rays (up to 1000 nm) due to the longitudinal increase in the size of the photodiodes. This camera may be of particular interest to those who like shooting Jupiter with a CH4 filter (890 nm) and shooting Venus. Of course, the camera is also suitable for general color photography of the Moon, planets, the Sun, comets and deep sky objects. There is no active cooling of camera.
Characteristics declared by the manufacturer

Sensor	SONY IMX464 CMOS​
Total pixels​	4.2 million pixels​
Diagonal​	9mm ​
Maximum resolution​	2712×1538 ​
Pixel size​	2.9μm* 2.9μm​
Image area​	7.9mm×4.5mm​
Maximum frame rate​	93FPS​
Time of exposure​	64us-2000s ​
Read noise​	2.9e~0.7e ​
QE peak​	80% ​
Full well charge​	12ke ​
ADC ​	12bit ​
Gain​	0-500 ​
ROI  ​	YES
Bin ​	1X1     2X2 ​
Digital noise reduction​	HCG mode noise reduction​
Compatible system​	Windows , Linux OS Raspberry Pi，Mac os, Chrome os​
Cache​	DDR3 128M ​
Guide satellite interface​	ST4 ​
USB interface​	USB3.0 ​
Camera interface​	1.25″/2″/ M42X0.75 ​
Back intercept​	6.5mm/12.5mm ​
Protective window glass​	AR antireflection film​​

 

Package list, overview
The camera is delivered in a cardboard box. A minimum of inscriptions, one sticker with the marking of the device.

In the box, I found the following items:
-Camera with a rubber protective cover installed
-T-adapter 1.25″
-T2 adapter ring (it was immediately screwed onto the camera)
-M42-CS adapter ring
-CS-C adapter ring (I did not find it separately. Perhaps it was immediately wound onto the M42-CS ring)
-ST 4 cable for autoguider
-USB 3.0 cable
-Mirror wiping cloth
-Manual (including in Russian).

The camera is made in the form of a metal “washer” 58 x 44 mm in size and weighing 0.6 kg. On the side there are several connectors – USB 3.0 and ST4. There is a 1/4″ thread on the back of the camera – it can be useful for mounting the device on a tripod or using it as an electronic polar finder. The matrix is visible in the front of the camera.

The sensor of the Svbony SV505C camera has a high sensitivity to near infrared rays, so the protective glass in front of the sensor does not have an IR-cut coating. This is done in order to be able to use the camera in several modes – color (with IR-cut filter) and monochrome (IR-pass filters from 685 nm). There are three camera options – without an IR-cut filter, with an IR-cut filter, or with a 685 nm IR-pass filter.

The package is rich enough. I lacked only a protective cover of 1.25 inches. In principle, it was also possible to put a CS lens in the kit for shooting meteors, it is inexpensive. I have no complaints about the quality of the camera and components.

Software and camera testing

The manufacturer claims that the camera will work under Windows, Linux OS Raspberry Pi, Mac os, Chrome os operating systems. Test computer specifications – Celeron J4125, 8 GB RAM, 512 GB SSD, USB 3.0 port, Windows 10 x64. Before checking the camera, I downloaded and installed the drivers from the official Svbony website. To test the camera, I used SharpCap version 4.0.9226 (64 bit) and immediately got an error:

The camera started only after copying the SVBCameraSDK.dll file from the SDK archive to the SharpCap folder. After that, there were no problems with detecting the camera in SharpCap. Perhaps in new versions of SharpCap this will be fixed.

The maximum image resolution when shooting is 2712 x 1536. The maximum frame rate is as advertised (93 fps), but please note that this will require a USB 3.0 port in the computer and a fast SSD drive. On devices with a USB 2.0 port and (or) HDD drive, the frame rate will be much lower. There is support for the SER format, as well as RAW8 and RAW16. A wide variety of camera settings are available, including per-channel white balance adjustment.

The camera was tested on the planets using the following equipment: Celestron NexStar 8 SE telescope, Barlow lens cell NPZ 2x, atmospheric dispersion corrector ZWO ADC, extension tube, filters IR-cut Svbony 1.25″, ZWO 850 nm, ZWO CH4 890 nm.

At first look, in color mode with a Svbony 1.25″ IR-cut filter, the camera works flawlessly – the white balance is adjustable, the recording speed is stable, and the color is quite correctly.

When shooting with an 850 nm filter, I had to set the color balance to 128\128\128 (by default):

But in a more detailed comparison with the QHY5III462C camera, I noticed a strange pixelation of the image.

With CH4 filter, the effect is more noticeable.

Below is one of the stacks taken with the Svbony camera using wavelets in Registax 6. Everything seems to be fine. However, upon closer examination, the square shape of the satellite is noticeable.

It seems that either debayerization does not work quite correctly, or some additional processing is applied, leading to pixelation. Most of the image settings in SharpCap were by default – in any case, I did not encounter such a problem with QHY \ ZWO cameras with similar settings.

You can download test videos from these two cameras from the link below and explore the material in the “Jupiter (QHY5III462C vs Svbony SV505C)” folder for yourself.
https://cloud.mail.ru/public/YZoU/br7kCrFsQ

UPDATED: On November 21, 2022, I received another Svbony 505C camera, and on November 25, 2022, I also received a Svbony 705C on a Sony imx585 color sensor. After processing the videos, I carefully checked the stacks first of all – I did not find any “square” artifacts. True, I installed the new driver version and used a different computer. Comparison videos from the QHY5III462C, Svbony SV505C and Svbony SV705C cameras can also be downloaded from the link above in the folder “2022-11-25 Jupiter New camera test (505c and 705c)” .

 

Equipment:
-telescope Celestron Omni XLT 127
–Sky-Watcher AZ-GTi mount
-Barlow lens 2x NPZ
–ZWO ADC
-ZWO IR-cut filter
–QHY5III462C , Svbony SV505C and Svbony SV705C cameras.
Stacking of 800 frames with Autostakkert, wavelets with Registax 6.

Dark frames and “amp glow”.
I also tested the camera for the presence of the sensor’s own glow at long exposures – the “amp glow” effect is noticeable both at 10s and at 60s, despite the presence of a DDR buffer in the camera. As far as I know, the buffer is needed just to minimize the speed of reading data from the sensor, and not for the stability of data transfer. I did not find the buffer on/off settings in ShaprCap.

An example of a dark frame at a shutter speed of 60s and a gain of 309 (about 50%), with debayerization and auto levels in the AsiFitsView utility:

Perfectly noticeable “lantern” in the lower right corner, and a uniform glow on the left side, as well as a bright upper right corner. However, the self-luminance of the sensor lends itself well to calibration and is not a problem, but I’m more concerned about the “chubby” defective stars-like pixels, as well as the familiar pattern of square pixels. And if defective pixels are usually removed by calibration and shooting with movement, then what to do with such large and star-shaped pixels – I don’t know yet.

Test darks are available at the link in the folder “Svbony SV505 dark test”.
https://cloud.mail.ru/public/YZoU/br7kCrFsQ

Test according to M 57 in short exposure mode

As a test object, I chose the M 57 “Ring” nebula – it is a very bright and compact object. A Sky-Watcher BKP15075 telescope + a Sharpstar 0.95x coma corrector, a Celestron NexStar SE mount in AZ mode, and an Svbony IR-cut filter were used for shooting. The addition took 507 frames at 0.5s. Of the calibration frames, only dark (dark) shots were taken.

SharpCap settings:

[SVBONY SV505C] FrameType=Light
Debayer Preview=On
Output Format=TIFF files (*.tif)
Colour Space=RAW16
Binning=1
Capture Area=2712×1538
Pan=0
Tilt=0
Black Level=0
USB Speed=1
Frame Rate Limit=Maximum
Gain=522
Exposure=500,0670ms
Timestamp Frames=Off
White Bal (B)=224
White Bal (G)=128
White Bal (R)=157
Trail Width=3
Minimum Trail Length=100
Trail Detection Sensitivity=9
Remove Satellite Trails=Off
Background Subtraction=Off
Planet/Disk Stabilization=Off
Banding Threshold=10
Banding Suppression=0
Apply Flat=None
Hot Pixel Sensitivity=5
Subtract Dark=None
NegativeDisplay=0
Display Black Point=0
Display MidTone Point=0,5
Display White Point=1
Notes=
TimeStamp=2022-09-10T20:25:51.7780716Z
SharpCapVersion=4.0.9268.0
StartCapture=2022-09-10T20:25:51.4972591Z
MidCapture=2022-09-10T20:29:01.1022591Z
EndCapture=2022-09-10T20:32:10.7063047Z
Duration=379,209s
FrameCount=756
ActualFrameRate=1,9936fps
TimeZone=+3,00

Next, I debayerized and subtracted the dark frame in PIPP, and then stacked in DeepSkyStacker. You can download the original frames and the addition result from the link in the “Svbony SV505C deepsky test M 57” folder
https://cloud.mail.ru/public/YZoU/br7kCrFsQ

Yes, with such a short shutter speed “amp glow” is almost imperceptible, but the camera sensor shows good sensitivity potential for shooting bright deep-sky objects.

Summary
Somewhat later I will add the results of other shootings from the Svbony SV505C camera, but so far my verdict is that the camera still needs to be tested, and the software needs to be finalized. I reported all the shortcomings and oddities in the functioning of the device to Svbony company. Unfortunately, the camera has already gone to another tester, but I will try to ask for it again later for testing. At the same time, in terms of sensitivity in the IR, the imx464 sensor turned out to be very similar to the imx462.

UPDATED: On November 21, 2022, the camera came to me again for a test, but a different copy. I installed an updated driver on another computer – I did not notice any artifacts on Jupiter either on the video or in the stacks. So the camera is working and can be successfully used for its intended purpose.