QHY5/PHD and Serious Noise Issue + Workaround

[Steve Drapak]

I have had an issue with my QHY5 guide camera for a while now (Using the ASCOM driver).  It turns out that updating past PHD 1.13.5 makes the screen get very noisy (see attached images), possibly an issue with the gain setting.  Setting gain to 100% helps but it's still pretty bad.

PHD 2.3.0 appears to be much more noisy even than PHD 1.13.x or 1.14.x. 

The older PHD versions can be downloaded here: http://www.stark-labs.com/OldVersions/Windows/

Hopefully the issue can be found and addressed in PHD 2 as I'd love to use it. 

Steve

[Bryan]

ve:

What happens when you move the screen gamma slider to the right?

Bryan

[SiriusOmegaBeta]

Which version of the QHY5 camera do you have?

Is it the standard QHY5 mono, or is it a 5V, 5T, 5-II, 5L-II, 5T-II or 5P-II?

Are you capturing darks?  Make sure you do as it will reduce noise.

I have the 5L-II here and I am running PHD 2.2.2a, I have attached a screenshot of the camera running in PHD 2.2.2a, 1 second exposure time after darks have been captured.

Ensure you have the latest driver from QHY installed.

[Astro Capella]

I have just started with PhD 2 and it is very frustrating with the noisy image and there is definitely something wrong when using the "qcam mono". All drivers are updated - I really hope it can be solved! 

Darks makes only a minimal difference. 

 

best regards

John

[Mike I.]

I'm having the same problem with my QHY5L-II M in both PHD and PHD2. The image brightens and darkens to various degrees with every exposure, with a great deal of noise, as if the camera or PHD was hunting for the right exposure or gain level. The noise and brightness problem frequently obscures the stars, so I can't guide. I've attached some screen shots to show what I mean. They look very much like Steve's screen shots. The problem occurs no matter what exposure duration or what version of PHD/PHD2 I use.

For hardware/software, I'm using an HP laptop with an i5 chip running Windows 7 Prof., 64-bit. The QHY5L-II M is being used in a Celestron OAG attached to an SCT. I'm using ASCOM 6, the most recent QHY drivers (ASCOM v11 driver and high-compatibility driver v14-12-12).

The camera connects without any problem in PHD (ASCOM Late Camera) and in PHD2 (using ASCOM Camera Chooser or the direct QHY5L-II Mono camera).

Stars are fairly well focused, and sometimes can be used to guide when they can be seen over the noise.

I've tried different usb cables, including the cable that came with the camera.

In camera properties, I've tried changing the gain or selecting 12-bit mode.

I've tried different exposure durations, between .5 and 4 seconds.

I've tried changing the screen gamma in PHD/PHD2.

I've tried using dark frames and noise reduction in PHD/PHD2.

Anyone else have this problem or can suggest a solution? Thanks.

Mike

[bw_msgboard]

There’s not really much info to work with here despite everything you’ve told us.  What is the SNR of the stars?  I don’t see anything alarming about the screenshots you sent, they just look like they might be too-faint stars.  Are you using a planetarium program to know how bright they are?  I can’t tell if you’re just setting up your OAG and perhaps aren’t familiar with it.  If you’re imaging with a long focal length SCT, you will almost always need to rotate the OAG and perhaps even move the scope a bit to get a suitable guide star.  It’s probably worth some time to get your own calibration of how faint you can go.  Set up first on a star you know will be bright enough – maybe mag 6 or so.  Make sure it’s focused well and get a sense of how long the guide exposure time needs to be.  Then move on to a fainter star, etc.  If you’re using some version of TheSky, this is pretty easy to do once you’ve constructed field-of-view indicators for your configuration.  If you’ve already done all this, that’s good, I just don’t know what you’ve done.  The darkening and brightening of the image display is normal if you’re dealing with a faint star right above the noise floor.  PHD2 is simply adjusting the screen stretch to display the full range of darkest to brightest pixels – if that range is small, the display can change quite a bit from frame to frame.  Watch the displayed SNR values to get a sense of whether the star is bright enough to be easily detected.

 

Good luck.

Bruce

 

 

---

--
You received this message because you are subscribed to the Google Groups "Open PHD Guiding" group.
To unsubscribe from this group and stop receiving emails from it, send an email to open-phd-guidi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Andy Galasso]

Mike,

It can be difficult to find guide stars when using an OAG imaging targets away from the plane of the Milkyway (like when imaging galaxies in the springtime.)   It might be useful to try pointing your scope into an area where there are more bright stars. That way you can determine if it a problem with the camera/settings, or, if it is simply lack of a bright enough guide star.

Andy

[Mike I.]

Andy and Bruce, thanks for replying. I've responded to Andy outside the forum with some additional info, which I include below, as well as some other comments:
The star in the attached screen shots is the same star, either Alkaid or Eta Herc., I can't remember which, as I had tried both of these  bright stars. The screen shots show the star well focused and clearly visible in one, while the same star is totally obscured by noise and brightness in the other. The screen shots were from a cold, clear night at a dark sky location. No dark frames were used in the screen shots. Tried darks on several occasions, and they did not help with this problem.  What I've been trying to convey, maybe inarticulately, is that the screen image in PHD and PHD2, even when a fairly bright star is used, varies in brightness and noise with each sub-exposure and produce images like the ones in the screen shots I attached.  It's as if the camera is hunting for the right brightness/gain with each exposure. I have tried all sorts of different gain settings in the camera properties setup.   It's not a defective camera, as the same night I tried a fellow club member's QHY5L-II M and got the same results. I posted this issue in the string of posts started by Steve Drapak last July, to show that I am having the same problem he did, although its not clear from his post whether he has the same camera.

[Sander Pool]

I am sitting here experimenting with Firecapture and my 5L-II. I just noticed that the first frame after changing gain is very noisy. It then calms down again. This is a far fetched idea but could it be PHD2 is sending gain and exposure settings to the camera before each frame when really there is no need to do that?

   Sander

[Bret McKee]

Interesting (and possibly quite useful) observation Sander.  

I'm heading out of town early tomorrow and can't do anything about it, but I took a quick look at the code shows that PHD2 does set the gain as part of every image capture...

Bret

--

[Andy Galasso]

We had some emails off the forum.  PHD2 is not setting the gain for every exposure, only when the gain is changed.

Andy

[Sander Pool]

It would have been a bit too perfect :)

Sander

[fly.r...@gmail.com]

G'day Steve,

I had very similar problem with Starshoot autoguider (same internals as a QHY5).

Just posted what I did, very much the same as you.

Basically I found PHD2 version 2.2.1 and SSAG driver v2 worked.
I think the version of PHD2 that had a listed update for the QHY5 might be where the issue started.
I have not yet tried PHD2 version 2.2.3 or 2.3.4 but plan to.

All the best,

Jamie