PHD Crashing After Dither

[John Pappas]

Hey everyone,

First time here on the forums. I've been having consistent issues with PHD2 and I haven't been able to find much info online about this specific problem, so I figured I would try this forum out.

PHD2 has been crashing regularly on me, pretty much every night now. Strangely enough around the same time as well. I know that whenever it crashes, it is always immediately after a dither. And there are no telling logs in the PHD2 logs or the NINA logs. The only potentially telling information is in the Windows event viewer, but even then there's not a whole lot going on. I was having the same issue on my previous computer which was running Windows 10, but I recently upgraded to a more powerful Beelink running Windows 11 and the problem remains. Fresh install of PHD2 does not work. Drivers are all freshly installed and up to date. The only 2 things I can think of are it's either my camera (don't know how this would cause the issue) or it is my PHD settings, potentially a bug with something I have enabled in the settings? No idea...

Anyway, I am using an ASI220mm to guide my AM5. PPEC for RA and ResistSwitch for Dec. I have attached both my PHD2 logs and the text that is in the Windows event viewer. Any help here is greatly appreciated. Thanks so much!

[John Pappas]

As someone kindly suggested, I needed to upload my debug logs as well. They can be accessed here:

https://openphdguiding.org/logs/dl/PHD2_logs_ECNN.zip

[Scott Devine]

Hello Everyone, 

Pretty much the exact same thing is happening to me four times and now consistently the last two nights although I'm not sure if it was necessarily after a dither.  We appear to have similar equipment as well so that may be helpful.  I am using a ASI174 guide camera and an AM5 mount running firmware version 1.5.2. I'm using the same guide algorithms as above.   I'm using NINA 3.1 HF2 Beta.  Last windows update was July 9th.  This has been happening since July 13th.  Everything had been working flawlessly for about the past month until the 13th when I got a crash at 3:14am (7/13), 2:46am (7/15), 2:19am (7/19) and again last night at 2:50am (7/20).  This was my Windows problem details on the 20th:

Description
Faulting application name: phd2.exe, version: 0.0.0.0, time stamp: 0x65860c38
Faulting module name: ntdll.dll, version: 6.2.22621.3733, time stamp: 0x86d1a81c
Exception code: 0xc0000005
Fault offset: 0x0004ebff
Faulting process id: 0x%9
Faulting application start time: 0x%10
Faulting application path: %11
Faulting module path: %12
Report Id: %13
Faulting package full name: %14
Faulting package-relative application ID: %15

I've attached last nights log and the debug file as well. For good measure

[Bobby Sapovits]

I had the exact same problem a year ago.  (Search the archives.)  Bruce finally figured out that the AM5 driver for PPEC had a problem.  ZWO was in the loop and fixed it.  But just for kicks and grins, try switching to Hysteresis and see if that helps.  Maybe your AM5 driver is older than you think.

Bobby

[Scott Devine]

I can certainly try that tonight.  

[Bruce Waddington]

It's probably not safe to assume these are caused by the same problem, whatever it may be.  And of course, these are the worst kinds of problems because the system has typically been running for hours, doing the same thing over and over, and then something fails.  Here are some things to try to perhaps isolate the problem.

1.  Upgrade to the 2.6.13dev5 release - that will allow me to more easily give you a diagnostics version of PHD2 if it comes to that.

2.  Try using Hysteresis instead of PPEC for RA guiding.  The tracking errors in strain wave mounts are usually not very periodic so you probably won't miss out on much by doing this.

3.   Create a new profile that uses the ASCOM version of the ZWO driver.  In Scott's case, there was some indication that the problem was camera-related.  PHD2 failed right after trying to start the next guide cam exposure.  Camera-related problems can definitely kill PHD2 because the camera driver is running in the same process space as PHD2.  

Also, take a close look at your Windows environment and protect against unwanted interference:

1.  Block automatic scans and installations of Windows updates.  With the Pro version of Windows, you can defer updates until a particular date in the future.  It isn't enough to just constrain the reboots, the process of scanning, downloading, and installing updates can create problems.  The fact that these problems seem to happen in the early morning hours is a curiosity for sure.

2.  Be sure you have blocked Windows from disabling USB ports that it thinks are idle.  This is most easily done in the Device Manager - find the USB-related controllers, look at 'properties', look at the 'power' tab, and tell Windows to butt out.

Regards,

Bruce

[John Pappas]

I had a suspicion this was the case... I found a single thread on this a while back and that was the issue. I am definitely on the latest drivers and firmware. I wonder if the issue is back from ZWO's side... Will be sure to try Hysteresis on the next clear night.

On Saturday, July 20, 2024 at 1:22:04 PM UTC-5 bobe...@equusbicycle.com wrote:

[John Pappas]

I can definitely upgrade to the dev release as well as trying a night of Hysteresis. I will try your point number 3 as well. Thank you for the response!

[John Pappas]

Update to this thread: I have not done anything other than switch from PPEC to Hysteresis and my crashing issues are gone. After crashing every night in a row for about a week, I have had 2 nights without crashes since switching off of PPEC. I wonder if the ZWO driver issue someone mentioned previously in this thread is back. I guess I should reach out to ZWO..?