|
Hi All,
Odd question, I am having issues with a device plugged into a USB hub of a touch screen PC, when it was connected to an old Win7 desktop via USB it worked with no problem. I'm guessing between 7 and 11 the USB driver changed (it would have to) and now the Win11 the device is unreliable and needs resetting. I wonder is it the device or the driver. It appears that the company does not have any specalised drivers so it's a standard com port.
Can this be an issue?
Glenn
|
|
|
|
|
If a hub is involved, my first suspicion is always power.
+5V ain't what is used to be (says an old time engineer).
You know where to put the multimeter prods...
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
I am a Linux fan, does Windows have an application to actually monitor / reed / guess the USB device power "in use " ?
Never seen such app in Linux, but did not bother to look for it.
|
|
|
|
|
|
Just had a dig and found my USB power meter!
|
|
|
|
|
As an electronics/hardware guy replying to another, I was suggesting a physical measurement, and comparison between the two physical environments.
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
I know, but if it's the bus power...
|
|
|
|
|
Slot in a powered hub?
Or I think they still make power injectors - signals straight through, solid +5 comes in the side.
Software rusts. Simon Stephenson, ca 1994. So does this signature. me, 2012
|
|
|
|
|
I will try that... have search my pile for an 9way RS232 cable, 'we don't have those things, why do you want one?'from IT... it gets worse!
|
|
|
|
|
It's sometimes hard to say what is windows versus what is a device manufacturer doing something in their driver.
I'm pretty sure I have had Windows 10 tell me I was pulling too much current on the USB and shut the port down temporarily (not a timer - it tells you to go turn it back on if it's OK).
|
|
|
|
|
Lost a couple of hours to that this afternoon!
|
|
|
|
|
Yup! 5.00 ish, Plugged into a dodgy hub... Interesting unit can crash and not appear as a com port...
|
|
|
|
|
Since I've been working with usbser.sys (>10 years), I've repeatedly encountered problems with the part. Since W10/11, power management has also been added to usbser.sys. This could be a cause of your problem; It's best to turn it off if possible.
|
|
|
|
|
Thank You! Got Senior to read this! with your reply I can prove it's not my program causing issues!
|
|
|
|
|
glenn, it's simpler than that... as soon as I get out of a meeting, I'll give you more details. Lived this for 20 years
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|
Your question is ambiguous. Bear with me.
- "when it was connected to an old Win7 desktop via USB it worked with no problem"
Explain this please. device was plugged in directly to the PC? If the PC was old, USB 2.0?
- is the touch screen a real PC running Windows 11? I want to be clear to make sure we're not talking about a WinIOT bastard child or something.
- Forget about any power issues. The first thing to check is what type of hub are you putting the old device into? USB is allegedly universal, but I've seen issues with older devices.
Note on the power comments: this absolutely used to drive me instance - Windows helping to save power by turning off devices. Its like MS lives in some other dimension. First thing I would do after plugging your device in is go to Device Manager. See if you have a driver issue. Your device may be sol old it's just not supported under Win11. I have a basket full of these things.
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|
|
Thanks for the link, could be a Microchip thinks it's an ATMEL
|
|
|
|
|
Unfortunately the most intersting link (http://www.microchip.com/forums/download.axd?file=0;538194[^]) is not longer available
That was an explanation of the low level stuff. Maybe I saved some parts of that link, I will search for it.
On the other hand, it was before >10 years. Maybe @charlieg has better information. I'm keen on his advices
|
|
|
|
|
Nuts, it would be useful, at least it's not my dodgy code
|
|
|
|
|
make sure you are checking EVERY return status. Since we have no idea what this USB device is (you can go ahead and admit it's a serial interface ), I have found that software people have far more faith than most devout <whatever> believers.
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|
Yes, its a dodgy com port ( ) I think people were expecting it run like native USB.
|
|
|
|
|
You have a serial port which is native a usb, but for your app it is still a serial port: What 'return status' you think you can check, other than the ones from the serial port? In such a case you simply rely on usbser.sys!
Sorry, but
Quote: I have found that software people have far more faith than most devout <whatever> believers.
is wrong and is just striking. It looks like you just want to appear omniscient here
|
|
|
|
|
Lord almighty no. However, I work in a very talented set of embedded developers, and I've watched over the past 15 years them making the same mistakes again and again. Carelessly not initializing variables; not checking return codes on ALL function calls. Hell, at least put up an assert or log something. Copying code cut/paste when one common source file would do... doubling and tripling the maintenance.
Omniscient? No. Scarred, bloodied and bruised? Yep.
One product we were working on uses ftp to shuffle files from the HMI device (Windows CE) to the custom controller board. They have a dedicated network - it's just the two devices. When we started working on upgrade testing, everyone just assumed that it should always work. Turns out that the ftp requests were failing about 25% of the time. Not checking status, not implementing retry code, etc. So, the people working in that area are now bloodied, bruised and much wiser.
Charlie Gilley
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
Has never been more appropriate.
|
|
|
|
|