-15 deg affecting all devices/connections or am i chasing a ghost?

[opticalpath]

Hope you find an answer with your last hardware checks. Let us know what happens!

Adrian

[graemlourens]

Did various hardwarechanges today, still same problems.
Took the computer home, and will be reinstalling everyting 1/1 the same as on the backupcomputer.

Murphys law that it will be the absolut last thing i test that will fix the problem - as usual

[Physopto]

One thing you can try is to unclip the memory and reseat it several times, pushing it in and out of the sockets. This tends to clean off any "varnish"  caused but the atmospheric dampness, dust and debris. You can do this with all connections on the motherboard, be they boards or connectors. You can also use a real rubber, "as in pencil rubber" to clean the plug in board connection strips. Just be a ware of static electricity. Ground yourself before and during the procedure. Touching the computer case will do as long as it is plugged into the mains BUT SWITCHED OFF !!!

We used to do this sort of thing routinely at sea on all connections many years ago. Marine atmosphere can play havoc with electronics. I still do it every so often on my computers to prevent any problems.

Also be aware that USB connectors on the computer can be at fault due to the repeated plugging in and out of connectors. it could be the reason for a particular socket causing intermittent failures. just check by using a different socket if you can.

As has been said hope you find the problem soon.

Derek

[graemlourens]

Uff, i think i have nailed the problem!

But will wait with sharing the solution before its run the whole night simulating imaging without errors tonight. Will write tomorrow if successful.

I am so excited....!

Kind regards, Graem

[graemlourens]

Good and bad News.

I have only partially killed the problem.

It turns out that the computer does not handle the tons of USB devices very well especially as the QHY5L-II is hogging all the USB traffic. I did a few reorganization of the USB ports, and now everything works except that the EFW2 in SGP starts hanging when changing filter after around 3-4 hours of operation. My suspicion is currently that i would have to sensibly distribute all USB devices (currently 11!!!) over the internal USB Buses, but first i'll have to find out which port is connected with which bus, something new for me.

Fact is for now i can safely rule out that it was a temperature problem, as currently we're above freezing again.

So I would like to thank you all for helping me out finding the problem. I will continue to pursue this problem to find out exactly what it is, but this will take me weeks/months as i was currently driving hundreds of KM just to replug a USB device and now i'll have to just do that at the next opportunity and not explicitly driving there for that.

My backupcomputer will take over the lead for now so i don't loose imaging time.

Kind regards, Graem

[Waldemar]

Maybe a SSD harddrive can solve your problem. They are far more resistant to low temperatures than the normal harddrives (no moving parts that slow down with low temperatures), and a LOT faster, so write/read  traffic will go easier.

Waldemar

[graemlourens]

Hi Waldemar.

Thx for your reply. My backupcomputer that works flawlessly has SSD, but i can pretty much rule this out as a problem, as the same error occured while temperature was higher again.

Currently experimenting a lot. Nearly there, but not quite yet, still one ghost to bust

Kind regards, Graem

[graemlourens]

Hi Everybody.

Just a quick update. I have been up and running again since around end of January. As the temperatures rose again to around 0 i can not pinpoint exactly what solved the problem, or if it really just was the temperature.

For sure the immediate solution that helped was to remove the QHY5L-II from the Hub and connect it to a totally different USB controller (at the front of the PC). This removed Timeout problems and since then everything is working. I suspect a combination of the QHY5L-II hogging all the bandwith (most say to not connect it to a hub) and temperature that affects the USB connections themselves. Its definitively not a software related problem.

Hope this helps anybody that is experiencing similar issues in future. I'll see what happens in 3/4 years when its back to far below 0

Kind regards, Graem

[StuartJPP]

On 08/01/2016 at 16:31, graemlourens said:

Murphys law that it will be the absolut last thing i test that will fix the problem - as usual

Once you have found the problem you don't need to search any further, so technically it should always be the last thing you test.

 

-15C is pretty extreme for consumer electronics but it is good that you found a potential candidate (the QHY camera), hopefully you have got it solved. Expecting 11 USB cables and devices to all work out of spec might be asking too much but we all do and given the conditions they fare pretty well.

[michael8554]

Perfrej, interesting ideas about USB connectors, particularly your advice to use mini connectors, only there seem to be many complaints on SGL about poor connections with mini USB's on cameras ?

Michael

 

[graemlourens]

Hi Everybody.

Its one year later! And guess what, first night with -20 degrees imaging, and some things start to fail again

I have had a perfectly running imaging automation for the whole of 2016 (starting approx march 2016 when the temperatures went back above -5 in the night). Yesterday was the first very cold night again (-20 degrees celcius) and again some ghost problems appeared. There have been no software or hardware changes, and it seems that its all concerning the usb cables / connections that do not like -15 / -20 degrees. Probably signals get lost and the drivers just are not used to handling that kind of signal loss.

I just wanted to let you guys know that is is definitively the extreme cold that creates these problems, but replacing usb cables etc doesn't help. These are all the same cables i had last year in the freezing cold, and they performed flawlessly now during spring/summer/autumn. So i'll just have to survive deep winter again and know to have an eye on it. For example i'm now not changing targets in the middle of the night when the equipment got very cold, even things like filterchanges etc i'm trying to avoid.

Kind regards & happy imaging!

Graem

[cardigan wearer]

I would check that with 11 devices on your usb that the +5 volts is actually not dropping due to high current. With everything connected accurately measure the +5volts at the PC power supply, and at the device. Or swap it out for a known good one/brand new.

Or use your backup PC in the winter and the main PC in the summer! But it would bother me that I know it has a reliability issue.

Keep us updated.

[Thalestris24]

On 3/11/2016 at 22:29, michael8554 said:

Perfrej, interesting ideas about USB connectors, particularly your advice to use mini connectors, only there seem to be many complaints on SGL about poor connections with mini USB's on cameras ?

Michael

 

Um, very sadly, Per is no longer with us though his knowledgeable posts live on.

Louise

[Thalestris24]

15 hours ago, graemlourens said:

Hi Everybody.

Its one year later! And guess what, first night with -20 degrees imaging, and some things start to fail again

I have had a perfectly running imaging automation for the whole of 2016 (starting approx march 2016 when the temperatures went back above -5 in the night). Yesterday was the first very cold night again (-20 degrees celcius) and again some ghost problems appeared. There have been no software or hardware changes, and it seems that its all concerning the usb cables / connections that do not like -15 / -20 degrees. Probably signals get lost and the drivers just are not used to handling that kind of signal loss.

I just wanted to let you guys know that is is definitively the extreme cold that creates these problems, but replacing usb cables etc doesn't help. These are all the same cables i had last year in the freezing cold, and they performed flawlessly now during spring/summer/autumn. So i'll just have to survive deep winter again and know to have an eye on it. For example i'm now not changing targets in the middle of the night when the equipment got very cold, even things like filterchanges etc i'm trying to avoid.

Kind regards & happy imaging!

Graem

Hi Graem

Have you checked that your usb cable connectors are being held tight at the laptop end? I've had problems with mine coming loose in cold weather. My workaround was to wedge the connectors with a small piece of card.

Louise

[graemlourens]

10 hours ago, Thalestris24 said:

Hi Graem

Have you checked that your usb cable connectors are being held tight at the laptop end? I've had problems with mine coming loose in cold weather. My workaround was to wedge the connectors with a small piece of card.

Louise

Hi Louise.

Do i understand you correctly that you're suspecting that due to the low temperatures the physical connection of the usb connectors (due to contraction of the metal) is not solid as it should be (even if those cables are never touched / moved) ?

It would be interesting. Not sure if the metal really can contract that much that it would affect the connectivity of the pins, but maybe that could be an explanation!

Kind regards, Graem

[VilleM]

Im using normal usb connectors in -25C without any issues.

-V

[graemlourens]

5 minutes ago, VilleM said:

Im using normal usb connectors in -25C without any issues.

-V

Interesting to hear. It puts be back to square one again though  

Its not only devices connected through a usb hub, its also directly connected devices such as the dome that cause random issues at low temps. The Weatherstation i have connected directly via Serial, and never had a problem anymore, also not in cold conditions. Its leading back to the obsy computer then again. Maybe i do try another computer and go through the effort of setting everything up just for a test!

This hobby never gets boring does it

[Thalestris24]

3 hours ago, graemlourens said:

Hi Louise.

Do i understand you correctly that you're suspecting that due to the low temperatures the physical connection of the usb connectors (due to contraction of the metal) is not solid as it should be (even if those cables are never touched / moved) ?

It would be interesting. Not sure if the metal really can contract that much that it would affect the connectivity of the pins, but maybe that could be an explanation!

Kind regards, Graem

I have a Lenovo Z50-70 laptop which is just over a year old. It has always had problems with the usb 2 connectors being on the loose side (even in the usb 2.0 ports) and they become very loose in the cold. Lenovo no longer offer this product... I don't know if it's a widespread problem with newish laptops generally, or with Lenovo, or just the model I have. Obviously you can easily check your system just by seeing if the usb cable connectors seem loose (side to side) or whether you get disconnects if you wiggle them slightly.

Louise

[graemlourens]

1 hour ago, Thalestris24 said:

I have a Lenovo Z50-70 laptop which is just over a year old. It has always had problems with the usb 2 connectors being on the loose side (even in the usb 2.0 ports) and they become very loose in the cold. Lenovo no longer offer this product... I don't know if it's a widespread problem with newish laptops generally, or with Lenovo, or just the model I have. Obviously you can easily check your system just by seeing if the usb cable connectors seem loose (side to side) or whether you get disconnects if you wiggle them slightly.

Louise

Hi Louise.

I will investigate that next time i'm at the obsy. To my last knowledge they are pretty tight, but on the other hand as i do not touch them, i also haven't got a reference to compare. I'll look into this!

Interestingly is also the fact, that when i have the problems with certain devices, if i restart the computer, the problems disappear, and will only appear a few hours later. This to me sounds either like memory issues of the system (that i can rule out as i did lots of tests last year) or to drivers that slowly get confused with more and more 'wrong' signals arriving.

Kind regards, Graem