 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Firmware Upgrade fried USB ports
| Author | Message | ||||
James_From_Canb Senior Member  Joined: 19/06/2011 Location: AustraliaPosts: 265 |
I have a powered MBeat 13 port USB hub plugged into my laptop. The Maximite USB is connected to this hub. When I recently updated the firmware on my Maximite the port on the hub failed when I pressed the boot load button. I tried another port and it also failed. Now those ports won't recognise the Maximite or a flash drive, and the port on/off switches don't work. I've just re-read Geoff's instructions and they say to connect the USB AFTER pressing the boot load button so it's my fault for not following the instructions. The firmware upgrade worked successfully when I connected the USB after pressing the boot load button so all is well, except for the 2 dud ports. What bothers me is that I've never had a problem updating the firmware until now. I've always had the USB cable connected to the computer when I press the button. The only difference I can think of is that the 13 port hub is a new addition. Has anyone else had a similar problem? Thanks James My mind is aglow with whirling, transient nodes of thought careening through a cosmic vapor of invention. Hedley Lamarr, Blazing Saddles (1974) |
||||
Geoffg Guru  Joined: 06/06/2011 Location: AustraliaPosts: 3194 |
It must be something else other than the bootload button. All it does is pull an input to the PIC32 up to 3.3V - the current draw would be about 0.02mA. Negligible when the hub should be able to supply 500mA. The reason I said to plug in the USB after pressing the button is because the bootload firmware only looks at the bootload button on power up - so the button needs to be depressed when power is applied. I was not trying to warn you of any damage. I would suspect that a faulty connector has damaged the socket on the hub. This happened to me, the connector pushed some if the pins into the back of the socket where they made an intermittent connection. Another possibility would be that you had the power jumper set up to connect BOTH the USB and external power at the same time. The MM would then try to power the hub and that could easily blow something in the hub. Quite often hubs have re settable fuses so try it again after a few minutes. Geoff Geoff Graham - http://geoffg.net |
||||
bigmik Guru Joined: 20/06/2011 Location: AustraliaPosts: 2914 |
Gday James, I was about to reply the exact same as Geoff, Powered Hubs are often not designed well (I have some that feed the +5V back into the PC thus connecting 2 5V supplies together.... not good)... How is your MM getting its power? If via the USB then it should not have caused you any grief.. if via an external 9V then I wonder if you have it somehow connecting both to USB and external power (shouldnt be possible) Mick Mick's uMite Stuff can be found >>> HERE (Kindly hosted by Dontronics) <<< |
||||
| James_From_Canb Senior Member Joined: 19/06/2011 Location: AustraliaPosts: 265 |
Hi Geoff and Mick. Thanks for the replies. The Maximite is jumpered to get its power externally, not from the USB. I looked for any obvious solder bridges and couldn't find any. I agree about the quality of the powered hubs. When I originally bought the MBeat I connected it to my PC (not the laptop) and the PC would not boot. I recall there was a BIOS message about too much power draw on the USB port. That's even though the hub had a power pack. I connected it to my laptop and it boots OK. Maybe the ports are failing because it's a marginal hub. I've changed the USB cable and I still get the same problem. I turn on the power for the Maximite and the light on the hub port goes out, and the port fails. That's 5 failed ports now. It also means that the problem isn't associated with the firmware upgrade, it's associated with powering on the Maximite. I disconnected the MBeat hub and connected a generic unpowered 4 port hub. I connected the mouse and a flash drive without incident, but when I plugged in the Maximite there was a message about the USB hub power being exceeded. Does this mean the Maximite is drawing power from the hub? That's what it implies to me. Should it? I then connected a power supply to the 4 port hub and was able to power the Maximite on and off without apparent damage to the port. It seems that the likely culprit is the 13 port hub, which is a shame because it wasn't cheap, and had more ports than I was ever likely to use. What could be causing the ports to fail? How could I check? Sorry for the dumb questions, but any suggestions on debugging this problem will be gratefully accepted. My mind is aglow with whirling, transient nodes of thought careening through a cosmic vapor of invention. Hedley Lamarr, Blazing Saddles (1974) |
||||
| paceman Guru Joined: 07/10/2011 Location: AustraliaPosts: 1329 |
Hi James, I'm no expert but the MM schematic shows the +5v from the USB connector being always connected to the Vbus inlet of the microchip. The same +5v input is not connected to the LM1117 low dropout 3.3v regulator unless the power source jumper (J1) is linked to the "USB source" side. Presumably this means that the micro can't run without the link set to "USB" (because there won't be 3.3v supply) but there may be a current draw (??) because of the always connected Vbus line. Can't give a definitive answer for you though unfortunately. Greg |
||||
| James_From_Canb Senior Member Joined: 19/06/2011 Location: AustraliaPosts: 265 |
Thanks Greg. The MM runs with the link set to EXT, but I had to supply power to the USB hub. James My mind is aglow with whirling, transient nodes of thought careening through a cosmic vapor of invention. Hedley Lamarr, Blazing Saddles (1974) |
||||
| bigmik Guru Joined: 20/06/2011 Location: AustraliaPosts: 2914 |
Hmm thats an interesting observation Greg, I think that Vbus `maybe' should have been connected to the +ve side of C1 ie supplied by what ever source is supplying power... Comments Geoff? I am sure you might know why Vbus is always USB power? What exactly is the Vbus input? If no USB and external supply there would be NO +5V referrence to the Pic32 Regards, Mick Mick's uMite Stuff can be found >>> HERE (Kindly hosted by Dontronics) <<< |
||||
BobD Guru Joined: 07/12/2011 Location: AustraliaPosts: 935 |
I recall Geoff saying that +5v into Vbus is used as an indicator to the PIC firmware that the usb is active. Check the schematic, it can only be active from the usb no matter what power source. |
||||
| bigmik Guru Joined: 20/06/2011 Location: AustraliaPosts: 2914 |
Hi Bob, That is probably it... Also if you connected to +ve of C1 then we would have a clash of power sources (USB supplied and external) so that is most likey the correct reason. Thanks Mate. Mick Mick's uMite Stuff can be found >>> HERE (Kindly hosted by Dontronics) <<< |
||||
trippyben Regular Member  Joined: 26/06/2011 Location: AustraliaPosts: 91 |
James are you using switchmode plugpacks? Most of these have a small value capacitor (or two) for EMI purposes connected between the mains and the output. This means that around 120V ac with a very high source impedance can be present at the DC output terminals. The high impedance keeps it safe, but it may have enough energy to zap an unsuspecting port when you make the connection with the power supply on. |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 3800 |
Would that voltage be seen with a multimeter? Or if not, what sort of impedance would be needed to see it? John |
||||
| trippyben Regular Member Joined: 26/06/2011 Location: AustraliaPosts: 91 |
Most DMM's should see it. Just now I measure 93Vac between mains earth and the DC terminals of a GP 12V 1500mA switchmode plugpack. |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 3800 |
Thanks. I don't like the sound of that... is it safe for health of the important chips? If not, is there a way to bleed it off (or similar)? John |
||||
| James_From_Canb Senior Member Joined: 19/06/2011 Location: AustraliaPosts: 265 |
Ummm, I don't know, but they're the cheap power supplies that come with the USB hubs. They feel too light to have a transformer in them. I just bought an oscilloscope so I'll see if I can figure it out on Sunday when I get back from Sydney. My mind is aglow with whirling, transient nodes of thought careening through a cosmic vapor of invention. Hedley Lamarr, Blazing Saddles (1974) |
||||
| Geoffg Guru Joined: 06/06/2011 Location: AustraliaPosts: 3194 |
This is more of a response to Mick's question a few posts ago. As Mick noted, the 5V from the USB is connected to Vbus so that MMBasic can tell if the Maximite is plugged into a computer. When there is zero volts on Vbus MMBasic completely ignores the USB so that you get the maximum speed. Having said that looking after the USB only takes about 4% of the CPU... but every bit helps. Geoff Geoff Graham - http://geoffg.net |
||||
| trippyben Regular Member Joined: 26/06/2011 Location: AustraliaPosts: 91 |
I wouldn't hot plug anything to a PC that is powered by a switchmode PSU. In theory the power connectors on USB are supposed to mate first thereby saving the data part of the port..... As James has hypothesized, some brands of port are going to be more robust than others. Using an external hub is probably cheap insurance against destroying the ports on the PC. I originally came across this issue a few jobs ago, the resident engineer was destroying RS232 ports by hot plugging a device that was powered by one of these switchmode PSU's. You could probably bleed it off to mains earth (via the PC case?) with a suitable resistor - maybe 100k? I think it's just safer not to hot plug stuff to the PC when using them. |
||||
| JohnS Guru Joined: 18/11/2011 Location: United KingdomPosts: 3800 |
Thanks. I wasn't just thinking of USB or hubs (potential thread hijack lol), as lots of things use such power supplies. John |
||||
 Print this page |
