When measuring with an oscilloscope, keep the probe as short as possible. For example, short the ground to the probe tip and place the probe near the switcher. If you still see the switching spikes, then your probe ia too long. A ground wire of 3 or 4 inches is too long. As I said before, electrolytic capacitors are no good for switchers because their ESR is way too high. Ceramic capacitors with a low ESR will clean up your power supply noise and make a big difference. Even tantalum capacitors are no good for switchers generally speaking. If you look at most vendor reference designs, they always use ceramic capacitors. In fact, some switchers won’t work unless you use ceramic capacitors.
Regards, John > On Dec 15, 2015, at 9:04 PM, Morgan Redfield <[email protected]> wrote: > > Hi everyone, > > I probed the 5V line of my setup today, with and without a BBB attached to my > cape. > > With no BBB attached, the unloaded startup looks like this: > <BenchSupplyUnloadedStartup.bmp> > There's considerable noise here, but it doesn't seem to exceed the limits of > the TPS65217. > > > When the BBB is connected to the cape, startup looks like this: > <WallWartLoadedStartup.bmp> > > > So I don't think the issue is with startup transients. > > During normal operation, the 5V rails looks like this: > <LoadedNoise.bmp> > There are 200mV spikes at 150kHz due to the switcher. I haven't seen those > spikes much larger than 200mV, but if they were then I could see them > damaging the TPS65217. > > The capacitors I've been using on my power supply are electrolytic. I can add > some ceramic capacitors as well. > > Does this all make sense? Any other ideas for what may have happened to my > BBBs? > > Thanks, > Morgan > > On Mon, Dec 14, 2015 at 8:44 PM, John Syne <[email protected] > <mailto:[email protected]>> wrote: > What is the part number for C2/C13? I would recommend ceramic with a low ESR. > Tantalum for Electrolytic will have a higher ESR and will not be suitable as > you will see considerable ripple/spikes. Regarding the relays, without the > diodes, it will only affect the transistors and not the power supply. You > want a diode across the relay to prevent the transistor collector voltage > from rising above the relay supply voltage. When the relay switches off, the > reverse EMF will cause the voltage on the transistor to exceed the relay > supply voltage and potentially damage the transistor. With the diode, the > energy from the reverse EMF will be dumped into the relay supply. If you have > low EMF ceramic capacitors across the relay supply, the supply will remain > clean. > > Regards, > John > > > > >> On Dec 14, 2015, at 4:57 PM, Morgan Redfield <[email protected] >> <mailto:[email protected]>> wrote: >> >> Thanks for the advice, guys. >> >> I actually had a couple of relays attached to this without flyback diodes, >> so that may be causing voltage spikes on the 5V input line. >> >> I'll take a look at the un-loaded startup of my regulator tomorrow and see >> how it looks. >> >> The regulator portion of the schematic is: >> <image.png> >> >> Best, >> Morgan >> >> On Fri, Dec 11, 2015 at 9:15 PM, Graham <[email protected] >> <mailto:[email protected]>> wrote: >> Morgan: >> >> It is likely a transient voltage spike that can come out of your switcher. >> >> The BBB does not turn on it's power supply until it thinks the incoming >> voltage is stable, which means that your 12V to 5V switcher is starting up >> without a load. If it overshoots badly in that start-up period, it could >> kill something. Or if it overshoots when the BBB load is finally applied >> >> I would start by repetitively starting up your 12V to 5V switcher, without a >> load on it, and watching what the output does on a storage (memory) >> oscilloscope, so that you can see the worst case startup condition. Then >> repetitively add a load equal to the BBB and all its input capacitance, and >> watch what happens. >> >> What were you controlling with the BBB/Cape? Things like relays or stepper >> motors generate inductive spikes that can easily kill semiconductors, if the >> spikes are not managed correctly. >> >> --- Graham >> >> == >> >> >> On Friday, December 11, 2015 at 7:14:38 PM UTC-6, Morgan Redfield wrote: >> I think I managed to burn out the TPS65217 on the BBB using a custom cape >> that I designed. The cape has a DCDC switching regulator that I'm using to >> drop a 12V supply down to 5V for the beagle bone. I have the 5V from that >> switching regulator connected to pins P9.6 and P9.5. >> >> I've now had two BBBs fail while powering them from the board. I left both >> on for a couple of days, and at some point the BBB just died. After that, >> the BBB don't boot at all, even with the cape unplugged. >> >> When I apply 5V from a benchtop supply to P9.6, I only see 1.1V on P9.7 >> (system 5V). >> If I hit the power button (S3), then the voltage on P9.7 will jump up to >> around 2.5V before falling back to 1.1V over around 20s. >> >> I'm not sure what's going on here, since the power supply I'm using looks >> pretty clean to me. It's an average of 5.14V with max 150mVpp noise. It's >> rated to 2A current draw. Switching frequency is 150kHz. >> >> Does anyone have any idea what might be happening here? Any ideas about what >> I should try next? >> >> Thanks, >> Morgan >> >> -- >> For more options, visit http://beagleboard.org/discuss >> <http://beagleboard.org/discuss> >> --- >> You received this message because you are subscribed to a topic in the >> Google Groups "BeagleBoard" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/beagleboard/hpHmvGR3cGU/unsubscribe >> <https://groups.google.com/d/topic/beagleboard/hpHmvGR3cGU/unsubscribe>. >> To unsubscribe from this group and all its topics, send an email to >> [email protected] >> <mailto:[email protected]>. >> For more options, visit https://groups.google.com/d/optout >> <https://groups.google.com/d/optout>. >> >> >> -- >> For more options, visit http://beagleboard.org/discuss >> <http://beagleboard.org/discuss> >> --- >> You received this message because you are subscribed to the Google Groups >> "BeagleBoard" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] >> <mailto:[email protected]>. >> For more options, visit https://groups.google.com/d/optout >> <https://groups.google.com/d/optout>. > > > -- > For more options, visit http://beagleboard.org/discuss > <http://beagleboard.org/discuss> > --- > You received this message because you are subscribed to a topic in the Google > Groups "BeagleBoard" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/beagleboard/hpHmvGR3cGU/unsubscribe > <https://groups.google.com/d/topic/beagleboard/hpHmvGR3cGU/unsubscribe>. > To unsubscribe from this group and all its topics, send an email to > [email protected] > <mailto:[email protected]>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. > > > -- > For more options, visit http://beagleboard.org/discuss > <http://beagleboard.org/discuss> > --- > You received this message because you are subscribed to the Google Groups > "BeagleBoard" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] > <mailto:[email protected]>. > For more options, visit https://groups.google.com/d/optout > <https://groups.google.com/d/optout>. -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
