Hi jmh, i have been debugging it today after issues last night. So a summary; the tinker board is stable no issues so no more updates for it. I use it when im doing a serious imaging run as it works perfectly. The rpi4 is a build in progress.
So i had not updated the rpi4 since the end of February 2020 as it was working ok. After the issues last night (mainly with guiding) i decide to do a full update and then the qhy issue started. The qhy and the zwo asi224 (guider) are both connected via USB3.
This morning i tried again with no luck. So i connected the qhy to the USB2 port instead of the USB3 port and presto, it worked. When i connected back to the USB3 port, the thing works.. All power connections we the same as last night after the upgrade. I then ran everything (mount via wifi, guider, imaging camera with cooling etc and everything is working under both USB2 and USB3. Really weird, i cant explain it.
I have read elsewhere on the forum issues with using USB3 and 2.4G wifi (which is how i remote to the rpi4). I switched to 5G wifi with no issues, except my mount connects via wifi and the shywatcher wifi dongles do not support 5G wifi, so i have to stick with 2.4G or go back to USB eqdirect.
Anyways, ill give it another go tonight. Maybe it is just scared of the dark
Lots of trial and error, mainly error, but in hindsight, it was easy!!!
Takahashi FC100DF imaging scope
QHY5 on Orion 50mm guiding
Skywatcher HEQ5 Pro via Wifi
Baader modified Canon 550D DSO
ZWO ASI 224MC Planetary
DIY Astroberry focuser
Raspberry Pi 4 with KStars/Ekos
I use a TinkerBoard with NAFABOX system (Ubuntu 18.04LTS) both a QHY 183C and QHY 224C. I use also an RJ45 cable 30m long to connect the TB and my laptop. I don't use USB3 HUB and all work perfectly. The wifi of this card are too feeble and need to be replaced by a directional antenna to run well.
The only issue I have, is the loading of default camera settings (gain, offset, USB trafic and USB buffer). The right values are not loaded correctly for some or all the settings.
More the USB buffer setting don't appear in the XML file configuration of the camera.
The cooler regulation works pretty much but no more.
I used a ROCK64 SBC with a USB3 hub and could never get it really stable. The USB3 hub connected the camera and a 500GB SSD boot drive. I tried a number of USB3 hubs but experienced issues with all of them. Rather than spend more money on looking for a combination of parts that worked, I went with the NUC. It uses more power, but it much faster and I've not had an issue for 6 months. I also don't use WiFi as it isn't full duplex, 5G attenuates more quickly than 2.4Ghz, and the use case just isn't suited to wireless. Much better to go with 1Gbe wired.
The drivers are another issue. QHY needs to find a good engineer to help with the cooling problem - it shouldn't be that hard to fix. The settings is also an issue, but is not unique to CCDs - I have to manually adjust the guide rates for every session for my CEM60 mount as they are not saved. Perhaps there were some unintended changes when the github repositories were reorganised. Over time, they'll get fixed I'm sure. I do have to say that in general, INDI is much more stable and the usability of EKOS get improving too.
Yesterday night was an awful nightmare with Kstars-Ekos-Indi !
My setup : Celestron AVX, SW ED80, QHY 183C, QHY 224C. All device plugged on a TinkerBoard (with NAFABOX system). Connection between the TB and my laptop with a RJ45 cable. This setup works perfectly for the last two weeks. I do all the polar and stars alignment with the nextar. Kstars V3.4.2. On sunday night I could image M65-M66.
I began to have issue with the TB (?) after an update. The Indi server breaks some time to time, but I can take images until last Sunday.
First of all I don't understand at all how the mount reacts at the Ekos connection. The setting is Ekos updates all devices. The mount is not located on the Polaris in Kstars. I have to do a polar alignment with the nextar to obtain the right location. Why ? I Would think doing an Ekos polar and stars alignment was the right way. So I do an Ekos polar alignment. But the pointing during the stars alignment was absolutely wrong !
More, the park is not the same if I do it with the Nextar or with Ekos.
If the QHY 183C works well this night, it was not the case for the QHY224C. Impossible to capture images into the guiding module, internal or PHD2 as the cam was not connected. I change USB place on the TB, but the result was not convincing. So I connected all the device with an USB3 hub (not powered) to my laptop and the 224C works well.
One more time, the cam settings (gain, offset, USB trafic, USB buffer) were not loaded correctly. And even the mount settings were not correct. By example focal 1800 instead of 510, or all the focal and aperture settings were losted ! I don't think it is an issue with the QHY SDK in this case. I made several message about this issue with no answer.
Issue also with the internal guide. Often the calibration fails. Either on the RA or on the DEC. Sometimes the guiding starts and runs well ?????? It is more reliable with PHD2. My settings for the guiding are: cam gain 11, exposure 1 or 2 sec, dark used, bin 2.
I join all the log files of this awful night. thank you for your help.
1. Regarding mount, Celestron MUST be aligned from the handset before you can connect it to Ekos. That's true for almost all other mounts which the exception of SkyWatcher which uses EQMod.
2. From the logs, it appears 224C is working OK. Not sure about configuration not being loaded, if you go to Options --> Load Config, they don't get loaded properly?
The log does need contain guide logging, so can't tell much about what was the issue with guide. If you're guiding via the MOUNT, then you can test manually if sending pulses in the Celestron drivers causes a change in RA/DE. You can send pulses manually by going to the Celestron driver --> Guide tab and then enter some pulse (3000 ms) for example in NORTH, and then you should be the DEC value changes. For WEST/EAST, you should see a corresponding change in RA.
Thank you for your answer. About guiding, it is via the mount. Using the Indi mount driver, the mount moves well. What is surprising is that sometimes the guidance starts on an object and not on the next, with an error in RA or DEC, during the same night.
If the mount must be aligned before connecting Ekos, what about the Ekos polar alignment and mount model in this case? No use ?
About indi device configuration, it occurs for the mount, all parameters about focal and aperture are all loosed ! I send you previously , the QHY183C XML file and the screen capture with different values. Load config doesn't work right in my thinking. I'll try next session.
Until the last version of Kstars, the cooling worked enough well.
Since the release it is less true. The temperature measured and displayed grows up to more than 5000°!
Deconnecting and connecting Ekos seems to cure the issue.
This function is really not stabilized.
I have just tested with the latest nightly build of INDI and KSTARS.
I have a QHY163m and QHY 5L-II mono and there are no major issues or changes since mid-April. Cooling stays within 0.2 C within a minute or two. No issues with downloads or any other camera functionality. I'd like the cooling tightened a bit more, but I can work with it now. After 5 minutes, it more or less stays within 0.1 C with an odd excursion to 0.2 C. The QHY5L-II connects to a remote PHD2 instance using INDI and also work well. It's been stable now for quite some time.