Date
1 - 20 of 35
USB and cold weather
|
Barry Megdal
Just a general comment about USB and cold weather. It turns that that low-cost USB gear (whether cards in the back of PCs or hubs, etc.) often use ceramic resonators rather than quartz crystals to produce the needed clock frequencies. USB is not very tolerant of errors in clock frequency, and ceramic resonators change frequency far more than quartz crystals with temperature.
So pay attention to specified temperature range for USB gear, and stick to commercial products where the given specs for temperature are likely to be more meaningful than with consumer gear.
- Barry
Dr. Barry Megdal
President Shb Instruments, Inc. 19215 Parthenia St. Suite A Northridge, CA 91324 (818) 773-2000 (818)773-2005 fax
Faculty (retired) Dept. of Electrical Engineering Caltech
|
|
|
|
On the topic of through mount USB3 and USB2.
USB 2.0 will work just fine, for a number of reasons. Most notably, there is far less communication actually happening between device and host, at a far lower speed, with far less 'protocol' communication. With most devices on the market, this is all you will ever need, and I could not make it fail in testing.
USB 3.0 is more tricky. As Roland mentioned in the thread, signal will bounce off of interfaces at the frequencies of USB 3.0, echoing in the line. The best way to deal with this is to ensure the least number of interfaces and shortest wire path between powered USB 3.0 ports. In the Mach 2 manual, we make some suggestions for USB 3.0 configurations that I have tested and can confirm to work perfectly. They all come down to having some kind of booster or powered hub right at the connector at the base of the mount, and a powered hub at the top of the mount. The internal cable is very high quality, but no amount of quality can prevent interface reflections.
An easy way to test your USB 3.0 setup is to get a USB 3.0 hard drive, and set it at the top of the mount. You can then use a USB speed tester program to see what quality of data transfer you get. I used USB Flash Benchmark with very good results. If the test fails, or Windows has a hard time detecting the hard drive, it means that the cables are too noisy to do proper communications and you need to reduce the cable length/connector interfaces between powered USB ports.
Astro-Physics
From: main@ap-gto.groups.io <main@ap-gto.groups.io> on behalf of Brian Valente <bvalente@...>
Sent: Monday, January 11, 2021 2:26 PM To: main@ap-gto.groups.io Subject: Re: [ap-gto] USB and cold weather >>>....when we had live bands at my bar.
Howard, you buried the lead here, interesting to know this about you :)
On Mon, Jan 11, 2021 at 12:10 PM Howard Hedlund <howard@...> wrote:
We have had pretty good results with the DeoxIT line of products for contact cleaning and moisture protection. I first saw DeoxIT D5 many years ago when we had live bands at my bar. The band roadies swore by the stuff.
|
|
|
|
>>>....when we had live bands at my bar. Howard, you buried the lead here, interesting to know this about you :)
On Mon, Jan 11, 2021 at 12:10 PM Howard Hedlund <howard@...> wrote: We have had pretty good results with the DeoxIT line of products for contact cleaning and moisture protection. I first saw DeoxIT D5 many years ago when we had live bands at my bar. The band roadies swore by the stuff. --
|
|
|
|
We have had pretty good results with the DeoxIT line of products for contact cleaning and moisture protection. I first saw DeoxIT D5 many years ago when we had live bands at my bar. The band roadies swore by the stuff.
https://caig.com/ The D5 is a great general purpose contact cleaner, and they have several products that create barriers to condensation. I'm not current on their latest and greatest, but you might want to give their products a look-see.
|
|
|
|
So I protected (DIY with foam) all USB3 contactors and tonight, with still 30 ° F I have no problems with my QHY294c and my complete installation as described.
I will try to improve his protections even further. Dominique
|
|
|
|
Micheal Fields Jr
All my hubs are 12v powered. And I have been only using one hub until the test.
|
|
|
|
Micheal Fields Jr
That was just an experiment to see what would break first. The idea being that if the worse case senario works but it still doesn't work through the mount, then I have isolated the problem. Also as Roland mentioned, maybe having extra power and signal integrity before it goes through the mount and after it goes through the mount will provide a more stable situation. So that is why I experimented indoors with using two hubs. That is not how I have been doing it up until then. I was just testing ideas.
|
|
|
|
Dominique I suspect that unpowered usb3 hub at the computer is your problem. All hubs in the chain need to be powered. Don Anderson
On Saturday, January 9, 2021, 05:18:42 p.m. MST, Dominique <d.h.durand@...> wrote:
I saw a problem like this 2 days ago in 28 ° F. My installation is complex with the Pegasus powerbox V2 installed in front of the plate, 1 USB3 Hub at the output of the mount wiring, 1 active 5m extension cable and 1 other USB3 Hub (not powered) connected to the laptop (see the attached diagram). As it generally works well and ceal makes my life easier, I have no reason to change. Can the di-electric silicone grease for the connections make things better? In the meantime I have covered all the USB3 connections and I will see tonight (it will still be cold) if it gets better. It would be a shame to do without USB3 especially when you have to do the flats with the QHY294M-pro used in 47M. Dominique
|
|
|
|
Michael Try running your set up with an external USB cable using your existing Startech hub. If your rig runs reliably at the low temperatures, you can pretty much zero in on cabling within the mount being the problem. If that is the case, then you can look into installing better shielded USB cables to see if that works. Don Anderson
On Saturday, January 9, 2021, 12:55:40 p.m. MST, Micheal Fields Jr via groups.io <mpfjr@...> wrote:
My camera unfortunately doesn't have a built in usb hub so I can't go from camera to hub. I am going to further experiment with this later today but at this point the only solution seems to be to not have through the mount cabling which was one of the factors that steered me towards this mount over its competitors. I hope I can get this working reliably.
|
|
|
|
That hub should work fine at 47F but those specs are just a general average. Your particular hub may not work reliably at those temps. The problem could also be running power cables along side data cables within the mount. Most USB cables are not well shielded. That could explain why you are getting better results with an external USB cable. Don Anderson
On Saturday, January 9, 2021, 12:51:21 p.m. MST, Micheal Fields Jr via groups.io <mpfjr@...> wrote:
This is the hub I am using. https://www.startech.com/en-us/cards-adapters/st4300usbm It claims it can go down to 32F as well. I'll check out the pegasus astro product. I still find it weird though that if I bypass the mount pass through it works fine.
On Sat, Jan 9, 2021 at 09:55 AM, Don Anderson wrote:
|
|
|
|
Joseph Beyer
Not sure this idea will help but my Nikon D800 uses a USB3 cable and I commonly had problems with connectivity, especially in the cold. Now I use the following: from the camera I use an 18 inch USB3 cable plugged into a powered Startech USB2 hub on the OTA, then down through my Mach1 using a 6 foot USB2 cable plugged into 16 foot active USB2 cable (Tethertools). I use either a USB2 or 3 plug on the computer. The raw images are roughly 50 Mb and download in about 1-2 seconds. I’ve not had a problem since adding the active cable into the line.
|
|
|
|
Joe Zeglinski
A bonus for using USB-2, besides 3x the distance, is that those
cables are “thinner”, by two extra wires in the sheath, than USB-3. So, the
cable bundle inside the mount is just a bit looser, especially twisting
when its cold.
Joe
From: Micheal Fields Jr via groups.io
Sent: Saturday, January 9, 2021 6:09 PM
To: main@ap-gto.groups.io
Subject: Re: [ap-gto] USB and cold weather I
do not use video for sure. Not with this camera. 3 seconds on
USB2? Interestingly I have an Icron Ranger 2304 which is a USB 2.0 over Cat5e which was allowing me to run long ethernet cable out to my mount in the back yard but use my in home computer to control everything. I would take about 30 seconds per download. You are right, I don't love your solution but it is either that or not use the through the mount cabling. Such a shame. I just did some experiments. Bottom line is if I bypass the mounts internal cable, I have zero issues at USB 3 speeds. You are right though. USB 2 port doing the same cables resulted in perfect no error images. Damnit.
|
|
|
|
In automotive applications I've seen people slop silicone grease all over a multi-prong connector and not have any issues. I do it myself for outside connectors, but for the most part those are only speaker terminals and low speed or low voltage power terminals (pool applications).
Thinking more about USB connectors, I'm not so sure. The way the USB connectors work you have relatively small surfaces making contact with each other with pretty low pressure spring force. The silicone greases are pretty viscous and very good insulators. So my concern is they might actually weaken the signal transmission a bit because when you slide the connectors together the pins might not have enough surface pressure to completely push the grease out of the way. I'm sure they'll still make contact, but it could be a little less contact area, which could slightly degrade a signal that's already nearing its limit. As it the temp gets colder the silicone grease gets thicker (harder to displace) and the spring force on the connectors likely goes down. So a silicone grease could make the situation worse, not better.
|
|
|
|
Sébastien Doré
Hello Michael,
As Roland pointed out, it also seem to me to be a matter of signal integrity at high freq. From your
video I counted 6 USB interfaces between your PC to your camera (1st hub input connector, 1st hub output, Mach2 TTM input, Mach2 TTM output, 2nd hub input, 2nd hub output), each of which are certainly affecting your USB3 signal waveform.
I would try removing one of the hubs from the signal path (hence removing 2 USB3 interfaces) and see if that helps.
Having 2 hubs to amplify signal is probably of little help anyway in your case given the short cables used between your PC and the camera. In fact, it probably does more harm than
anything by causing multiple insertion loss, reflections, frequency multiplication (harmonics) adding noise and cross-talks which all gets worst the higher you go in frequencies (i.e. in speed). More amplication (gain) isn't better when applied to a noisy
and heavily distorted (USB) signal unless the signal gets very efficiently conditionned (filtered) inside the hubs (which is unlikely).
Sébastien
_._,_._,_
|
|
|
|
I use an Atolla 7port USB3.0 powered hub in my observatory, temp now is about 20deg F, no problems. It's cheap, about $25 on Amazon or Ebay. My ASI 1600mm camera has an internal hub adding the filter wheel and guide camera to a USB3 output that goes to the hub, the mount is connected via a serial to USB 2 converter, and the Shoestring USB focuser and a dongle for keyboard mouse goes in as well. So just the hub's single USB3 output goes to the computer, an Acer laptop. No issues save that the hub's USB3 output cable is only about 1 meter and USB 3 extension cables don't work well, whether active or passive. I can just get by with that length. The Pegasus looks great but this inexpensive hub has been working well for me for a year and it doesn't care about cold. Steve
Do you have any suggestions? I was considering the idea of a powered hub with a short cable plugged into the mount bottom and then another hub at the top but I believe there is a limit imposed by Windows on how many hubs you can have.
|
|
|
|
Micheal Fields Jr
I've never used them on USB connections previously but if there is a way to protect the connections and make sure they don't get moisture or start corroding I am going to do that. This will be in an observatory sitting for weeks at a time so once I have everything working reliably the cables will never have to come unplugged. (in theory)
|
|
|
|
Micheal Fields Jr
Not going to take the mount apart for sure.
Thanks for the info about the reflections. Way over my head for sure. Be interesting though to see if my experiment next week shows the same issue persisting or not.
|
|
|
|
Roland Christen
At the frequencies used in USB3, they begin to act more like light going thru various surfaces. At each connection there is a reflection of the signal just like the air-glass surfaces of a lens. So 2 more connections adds 2 reflections which lowers the signal amplitude and adds ringing of a steep rise-time pulse. If you have low signal to begin with, this may knock it down enough to cause the problem. There are hundreds of papers written on this if you care to search, but they are very technical if you're not knowledgeable of gigahertz stuff.
Running your own cable thru the mount would require taking the mount apart, which is highly not recommended and you could damage the internal wiring.
Rolando
-----Original Message-----
From: Micheal Fields Jr via groups.io <mpfjr@...> To: main@ap-gto.groups.io Sent: Sat, Jan 9, 2021 6:58 pm Subject: Re: [ap-gto] USB and cold weather Hi Roland. I just did a ton of experimenting. Xentex was right. If I simply plug into a usb 2 port the problems go away. Also if I bypass the mount entirely even with a 6 foot long cable, the problems go away. It is only if I go through the mount with USB 3.0 do I have issues. The Pegasus USB box is nice. Probably worth having anyways but it is also an expensive "test". I just bought a 3.3 foot female to female cable USB 3.0 to simulate going through the mount so I can see if it is simply having those extra connections or something else.
$10.00 with free shipping from Amazon will let me know. Lets say it turns out that it works great with that "pretend" through the mount female to female cable. What is my next step? By process of elimination it seems so far that the internal cable is not up to 3.0 standards. I am waiting to hear back from my second favorite vendor to see if they have the usb control hub in stock then I'll probably go ahead and order it anyways. Nice being able to turn off and on usb ports. -- Roland Christen Astro-Physics
|
|
|
|
Hi Roland. I just did a ton of experimenting. Xentex was right. If I simply plug into a usb 2 port the problems go away. Also if I bypass the mount entirely even with a 6 foot long cable, the problems go away. It is only if I go through the mount with USB 3.0 do I have issues. The Pegasus USB box is nice. Probably worth having anyways but it is also an expensive "test". I just bought a 3.3 foot female to female cable USB 3.0 to simulate going through the mount so I can see if it is simply having those extra connections or something else.
$10.00 with free shipping from Amazon will let me know. Lets say it turns out that it works great with that "pretend" through the mount female to female cable. What is my next step? By process of elimination it seems so far that the internal cable is not up to 3.0 standards. I am waiting to hear back from my second favorite vendor to see if they have the usb control hub in stock then I'll probably go ahead and order it anyways. Nice being able to turn off and on usb ports. Edit: I was recording some of my experimenting. Here is a link to a unedited video if any of you people are bored enough to watch. At about 7:50 I bypass the mount and everything works great. At the end I used Xentex idea and swapped to a usb 2.0 port and everything works that way too. https://youtu.be/AOXtrYJypmA
|
|
|
|
A silicone grease can be useful to prevent corrosion of contacts, and if you're plugging and unplugging them a lot or they get jiggled in their sockets it will help reduce wear. I use it all over outdoor contacts that are exposed to the elements. I haven't ever used them on a USB connector because I wouldn't expect a USB connector to be exposed to elements.
Regarding download times for a 47M image file, that should be about 1 second on a USB 2 interface.
|
|
|