[mythtv-users] HDPVR intermittent failure

glen glenb at glenb.net
Sun Jun 28 16:49:38 UTC 2020


On Sun, 2020-06-28 at 09:45 -0700, DryHeat122 . wrote:
> On Sat, Jun 27, 2020 at 7:51 PM glen <glenb at glenb.net> wrote:
> > On Sat, 2020-06-27 at 18:46 -0700, DryHeat122 . wrote:
> > > On Mon, May 11, 2020 at 12:59 PM DryHeat122 . <
> > > dryheat122 at gmail.com> wrote:
> > > > On Mon, May 11, 2020 at 2:41 AM Stephen Worthington <
> > > > stephen_agent at jsw.gen.nz> wrote:
> > > > > On Sun, 10 May 2020 20:20:30 -0700, you wrote:
> > > > > 
> > > > > 
> > > > > 
> > > > > >On Sun, May 10, 2020, 7:58 PM Greg <gregl at nycap.rr.com>
> > > > > wrote:
> > > > > 
> > > > > 
> > > > > 
> > > > > >Ok thanks I will investigate that.  But it's been working
> > > > > fine with the
> > > > > 
> > > > > >current power source for like a year now, and others on this
> > > > > list use it
> > > > > 
> > > > > >too.
> > > > > 
> > > > > 
> > > > > 
> > > > > USB cables often fit badly into the sockets.  If so, the
> > > > > connection
> > > > > 
> > > > > generally gets worse over time as you get dirt or oxidisation
> > > > > on the
> > > > > 
> > > > > contacts.  So badly fitting cables will degrade with time. 
> > > > > The result
> > > > > 
> > > > > is that the voltage drop across the cable will increase
> > > > > markedly.  And
> > > > > 
> > > > > they can be too thin - the amount of copper in the wires is
> > > > > too little
> > > > > 
> > > > > and that causes high resistance and a big voltage drop across
> > > > > the
> > > > > 
> > > > > length of the cable.  Some (most?) USB cables are designed
> > > > > only for
> > > > > 
> > > > > data transmission, or to run very low power devices.  For a
> > > > > high power
> > > > > 
> > > > > device, you need a better (thicker) cable.  The high power
> > > > > devices do
> > > > > 
> > > > > a negotiation with device supplying the power and request
> > > > > high power
> > > > > 
> > > > > mode.  If the cable is not capable of high power, that
> > > > > negotiation is
> > > > > 
> > > > > not supposed to work and the device should either only work
> > > > > in low
> > > > > 
> > > > > power mode or it should turn itself off.  But USB cable
> > > > > makers often
> > > > > 
> > > > > make cables not capable of high power transmission that will
> > > > > allow the
> > > > > 
> > > > > high power mode negotiation to succeed.  So even though the
> > > > > device
> > > > > 
> > > > > supplying the power is sending high current, the voltage drop
> > > > > in the
> > > > > 
> > > > > cable means that at the other end, the voltage can be below
> > > > > the level
> > > > > 
> > > > > required for proper operation or to fully charge the device's
> > > > > battery.
> > > > > 
> > > > > 
> > > > > 
> > > > > I have had two notably bad experiences with USB cables.  One
> > > > > was a USB
> > > > > 
> > > > > DVB-T tuner, and it was very like your experience - it would
> > > > > go for a
> > > > > 
> > > > > number of days just fine, then suddenly stop.  If I unplugged
> > > > > it and
> > > > > 
> > > > > plugged it in again, it would usually work again.  When I
> > > > > finally
> > > > > 
> > > > > investigated properly, I found the cable was just a little
> > > > > loose in
> > > > > 
> > > > > the PC's socket.  I replaced the cable with one that fit more
> > > > > tightly
> > > > > 
> > > > > and the tuner was much more reliable.  It still occasionally
> > > > > caused
> > > > > 
> > > > > trouble, but only when I had bumped the cables (or in one
> > > > > case, after
> > > > > 
> > > > > we had a small earthquake).  So because of that and because I
> > > > > needed
> > > > > 
> > > > > more DVB-T tuners, I finally replaced all my DVB-T tuners
> > > > > with an 8
> > > > > 
> > > > > tuner PCIe card.
> > > > > 
> > > > > 
> > > > > 
> > > > > The second bad experience was my Samsung Galaxy Tab S2
> > > > > tablet.  Its
> > > > > 
> > > > > USB charging cable was supplied with it by Samsung, so I
> > > > > assumed it
> > > > > 
> > > > > was a good one.  But right from the start, the tablet took a
> > > > > long time
> > > > > 
> > > > > to charge, and the time gradually got longer and longer and
> > > > > the
> > > > > 
> > > > > battery life on one charge was getting less and less.  And
> > > > > then I
> > > > > 
> > > > > started to have to jiggle the cable in the socket to get it
> > > > > to charge
> > > > > 
> > > > > at all.  I actually called the Samsung help line about this,
> > > > > and they
> > > > > 
> > > > > said it sounds like a bad cable.  So I bought a expensive
> > > > > (NZ$30)
> > > > > 
> > > > > Pudney & Lee charging cable, which was a fair bit longer than
> > > > > the old
> > > > > 
> > > > > Samsung cable, but fit very tightly at both ends and was
> > > > > significantly
> > > > > 
> > > > > thicker - it has more copper in the wires in the cable.  Then
> > > > > suddenly
> > > > > 
> > > > > the battery charging times were what was specified for the
> > > > > tablet,
> > > > > 
> > > > > rather than three times as long.  And over a number of
> > > > > charging
> > > > > 
> > > > > cycles, the battery life came back again.  So the original
> > > > > Samsung
> > > > > 
> > > > > supplied cable was clearly bad from the start - it is
> > > > > probably less
> > > > > 
> > > > > than the specification required to charge the tablet properly
> > > > > as it is
> > > > > 
> > > > > too thin and has too much voltage drop even when the plugs
> > > > > fit
> > > > > 
> > > > > properly.  So definitely NZ$30 well spent.  But I am
> > > > > surprised that a
> > > > > 
> > > > > reputable company like Samsung would supply a bad cable with
> > > > > an
> > > > > 
> > > > > expensive top-of-the-line product like my tablet.  But they
> > > > > did - so
> > > > > 
> > > > > now I always suspect any USB cable I get and keep an eye on
> > > > > how well
> > > > > 
> > > > > it is working.
> > > > > 
> > > > > _______________________________________________
> > > > > 
> > > > > mythtv-users mailing list
> > > > > 
> > > > > mythtv-users at mythtv.org
> > > > > 
> > > > > http://email.mg.glenb.net/c/eJxNjkEOwiAURE9TdhL-LxRZsHDjPT4FWhKgpqCJt7caE01mM29mkvF2juSVZMniqAAwOq2CjmDOTiNMQjo1ChkBlQaBIhrNZyohD1IsOVTHa-hstQjehAkhSEXGqzkQSQQhKOIkndAs27X32zBeBrweyqn1xsuzr_3Bt305UKGUC9VvmGrc3vDTON1b2Bvb7b89Hvz2L-CbPz8
> > > > > 
> > > > > http://email.mg.glenb.net/c/eJxFjcsOgjAQRb-G7mw605csunDjSr-BUJhCY0GFQePfS-LC5K5OzsntQ5fa3hqRA2oLgCl6Sz5BfYwewSkTrVYmAVoPClWqvezaiUpl1FBojnImFmNwYNErU8cumnRE5zTolsgBJddjR6KEkflR6VOF533vfMty-vDIL3lfhp1c21zyPDSXvHJDnJ8bMZNYws86bCst6_76j74P1DtC
> > > > > 
> > > > > MythTV Forums: http://email.mg.glenb.net/c/eJxFjDEKwzAMRU8TbzWyItX24KFL72ElVlJI0uI4hd6-gQ6FP7zHgz-mQfPIZB4Je3YOVTwXry4G8eiuQMI9kDpk7wBBo7dDXsvSEUxL2cRupZk5CUTCnIFFmVwIKBTDSCcPnkouZklza6-9628d3s_psx6rXT9tbm_7rJOp6SeXYy91P-__7QsiqTJl
> > > > 
> > > > Thanks for all the suggestions.  I will investigate all of
> > > > them.  Something I do not see in here is an opinion that some
> > > > software problem could have evolved.  I was kind of thinking of
> > > > that as a possibility given that it worked fine for so long
> > > > then seems to be degrading.  OTOH I never thought of the
> > > > possibility that the USB connection could be degrading.  I'm
> > > > going to try Greg's connector, assuming I can get parts during
> > > > the Apocalypse.  Also, his link specifies Radio Shack parts. 
> > > > What's Radio Shack?!? ;-)
> > > 
> > > Well folks this is getting frustrating.  I addressed a possible
> > > ventilation issue.  I also unplugged the power cord I had from
> > > the Myth box USB and connected it to an old iPad charger rated at
> > > 5V/2A.  No help.  It wouldn't record anything, no matter how much
> > > rebooting and power-cycling.  So I concluded the HDPVR was hosed
> > > and got a new one (more accurately, a replacement circuit board
> > > for it).  Powered it with the same wall wart.  Recorded great for
> > > a day.  Just turned it on and I have failed recordings and it
> > > won't respond when I try to manually play a channel.  Reboot, and
> > > it works fine, same pattern as before.  
> > > 
> > > I finally got a molex connector and am still going to try what
> > > Greg suggested, but I really don't think it's a power issue. 
> > > Apply makes good electronics and that wart has the same specs as
> > > the power supply that comes with the HDPVR.  The HDPCR isn't
> > > overheating.  It's not the HDPVR itself.  That only leaves the
> > > software on the myth box.  Anyone have ideas how to troubleshoot
> > > that?
> > > 
> > > 
> > > _______________________________________________mythtv-users
> > > mailing listmythtv-users at mythtv.org
> > > http://email.mg.glenb.net/c/eJxNjkEOwiAURE9TdhL-LxRZsHDjPT4FWhKgpqCJt7caE01mM29mkvF2juSVZMniqAAwOq2CjmDOTiNMQjo1ChkBlQaBIhrNZyohD1IsOVTHa-hstQjehAkhSEXGqzkQSQQhKOIkndAs27X32zBeBrweyqn1xsuzr_3Bt305UKGUC9VvmGrc3vDTON1b2Bvb7b89Hvz2L-CbPz8
> > > http://email.mg.glenb.net/c/eJxFjcsOgjAQRb-G7mw605csunDjSr-BUJhCY0GFQePfS-LC5K5OzsntQ5fa3hqRA2oLgCl6Sz5BfYwewSkTrVYmAVoPClWqvezaiUpl1FBojnImFmNwYNErU8cumnRE5zTolsgBJddjR6KEkflR6VOF533vfMty-vDIL3lfhp1c21zyPDSXvHJDnJ8bMZNYws86bCst6_76j74P1DtC
> > > MythTV Forums: http://email.mg.glenb.net/c/eJxFjDEKwzAMRU8TbzWyItX24KFL72ElVlJI0uI4hd6-gQ6FP7zHgz-mQfPIZB4Je3YOVTwXry4G8eiuQMI9kDpk7wBBo7dDXsvSEUxL2cRupZk5CUTCnIFFmVwIKBTDSCcPnkouZklza6-9628d3s_psx6rXT9tbm_7rJOp6SeXYy91P-__7QsiqTJl
> > 
> > have you tried a conf file in /etc/modprobe.d setting this option:
> > options ir-kbd-i2c enable_hdpvr=1
> > 
> No.  does ir-kbd stand for infrared keyboard?  I don't have one of
> those.   

if you are using nvidia, recently, the drm driver for nvidia brought in
different dependent modules and this was one whether or not you have
hardware using it. check your loaded modules (lsmod) to see if it is
loading. i had this issue with no ir hardware and had hdpvr issues
until i changed that option. simple enough to add a conf file and see
if you get a difference or not.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mythtv.org/pipermail/mythtv-users/attachments/20200628/aad16ea3/attachment-0001.htm>


More information about the mythtv-users mailing list