Post by thread:[PIC] Having problems with 10LF322 going into brow

Good day to all. I'm working on a project that uses a PIC 10LF322 that runs from two NiMh cells in series. I've been having weird problems that I finally tracked down to the PIC going into reset at voltages less than about 2.20V. I figure that the PIC is going into reset because pins defined as output suddenly become high-impedance when the supply voltage drops below that magic value of about 2.20V. My current config word is __config _PWRTE_OFF & _WDTE_OFF & _FOSC_INTOSC & _CP_OFF & _MCLRE_OFF & _BOREN_OFF & _BORV_19 & _LVP_OFF I see another set of brown-out bits in the include files: _LPBOR_OFF EQU H'3DFF' ; BOR disabled _LPBOR_ON EQU H'3FFF' ; BOR enabled but I can't see anything that describes what that bit does in the current data-sheet. However, adding "_LPBOR_OFF" to the config word does not appear to stop the PIC from going into brown-out - and increases the supply current while in brown-out by 0.4mA or so. So I took that bit out of my config word. My goal is to have the PIC work down to at least 2.0V *and* to have the lowest-possible current consumption when the PIC goes to sleep. The PIC10LF322 is rated to work down to 1.80V. This project is to add low-battery disconnect to a board when its NiMh cells become discharged too deeply PLUS adding an idle timer that shuts the system down if it has not been used for a long period of time. Restart is caused by cycling the power into the board. The weird problems that I was seeing was that it appeared as if the PIC were never going into sleep. I would drop the supply voltage down towards the low-battery cutoff voltage (2.05V) but the current would never drop below about 0.22mA. And - I just happened to notice that the PIC pin that drives the power-switch MOSFET was going high-impedance instead of remaining low-impedance. Bringing the voltage above 2.20V causes the unit to resume normal operation instead of remaining in sleep. So: I'm guessing that I've done something stupid with the config bits but I just can't see it. I'm hoping that the collected wisdom of the PIClist will point me in the right direction. Many thanks! dwayne -- Dwayne Reid <spam_OUTdwaynerTakeThisOuTplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax http://www.trinity-electronics.com Custom Electronics Design and Manufacturing

At 06:00 PM 4/29/2012, you wrote: {Quote hidden}>Good day to all. > >I'm working on a project that uses a PIC 10LF322 that runs from two >NiMh cells in series. I've been having weird problems that I finally >tracked down to the PIC going into reset at voltages less than about 2.20V.. > >I figure that the PIC is going into reset because pins defined as >output suddenly become high-impedance when the supply voltage drops >below that magic value of about 2.20V. > >My current config word is > >__config _PWRTE_OFF & _WDTE_OFF & _FOSC_INTOSC & _CP_OFF & _MCLRE_OFF >& _BOREN_OFF & _BORV_19 & _LVP_OFF > >I see another set of brown-out bits in the include files: > >_LPBOR_OFF EQU H'3DFF' ; BOR disabled >_LPBOR_ON EQU H'3FFF' ; BOR enabled > >but I can't see anything that describes what that bit does in the >current data-sheet. However, adding "_LPBOR_OFF" to the config word >does not appear to stop the PIC from going into brown-out - and >increases the supply current while in brown-out by 0.4mA or so. So I >took that bit out of my config word. > >My goal is to have the PIC work down to at least 2.0V *and* to have >the lowest-possible current consumption when the PIC goes to >sleep. Internal brownout can't be guaranteed to not reset the part below 2.11V on the LF32x. (between 1.80 and 2.11V with nominal at 1.90V. If we make the assumption that you've set things up properly and the (dangerous) assumption that the chip you are playing with atm is nominal in characteristics.. then there is another possibility. If you have "noise" on the power supply, for example, from operating an intermittent load such as a beeper, motor, solenoid, pulsed LED or some such load, then if that noise causes the power supply voltage to drop below 1.90 V (nominally) for more than a microsecond or thereabouts, then the chip will go into reset and not begin to operate again until 1.925 to 1.95V, the way I read the data sheet. Worst-case it might not come out of BOR until 2.16V. Maybe it goes back into reset then if the load starts again. Does that sound like a possibility? Have you put a decent (>=50MHz bandwidth) scope on the power supply rails _right_ at the MCU GND/VDD pins? Best regards, Spehro Pefhany --"it's the network..." "The Journey is the reward" .....speffKILLspam@spam@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com

At 05:44 PM 4/29/2012, Spehro Pefhany wrote: >Internal brownout can't be guaranteed to not reset the part below 2.11V >on the LF32x. (between 1.80 and 2.11V with nominal at 1.90V. Thanks for that clarification. OK - that means that I can't use brown-out reset at all. However, I *thought* that I had the config word set up to disable the BOR module. {Quote hidden}>If you have "noise" on the power supply, for example, from operating >an intermittent load such as a beeper, motor, solenoid, pulsed LED >or some such load, >then if that noise causes the power supply voltage to drop below >1.90 V (nominally) >for more than a microsecond or thereabouts, then the chip will go >into reset and not >begin to operate again until 1.925 to 1.95V, the way I read the data >sheet. Worst-case it might not come out of BOR until 2.16V. Maybe >it goes back into reset then if the load starts again. > >Does that sound like a possibility? Have you put a decent (>=50MHz bandwidth) >scope on the power supply rails _right_ at the MCU GND/VDD pins? Yeah - I had a 200MHz DSO right at the power pins. That helped me fix one problem I was having (my bench supply going into current limit for a couple of microseconds when the PIC went to turn on one of those monstrously-high inrush-current LM2623 DC-DC converters I talked about a few weeks ago). I fixed that by momentarily connecting some nice, stiff NiMh cells in parallel with the power supply during the DC-DC converter startup. For tracking down this problem, I simply disconnected the enable lines to the DC-DC converters (they have pull-up and pull-down resistors so as to float safely) but was still observing this problem. There was no other load on the power supply except for the PIC chip itself, which wasn't driving any outputs. I mentioned that I had previously tried adding "_LPBOR_OFF" to the config word without having any effect other than increasing current while the PIC was supposedly in sleep. Unfortunately, I could find no description of what the LPBOR module was or how it was supposed to work in the datasheet, so I removed "_LPBOR_OFF" from the config word. What I don't remember is whether that was when I was using the beta version of the AC244045 Debug Adapter or whether it was with the chip all by itself. Anyway, I opened a support ticket with Microchip. The first suggestion they gave was to add that "_LPBOR_OFF" bit back into the config word. This time, it appeared to fix my problem. I'm working with the PIC all by itself at this point and indeed: dropping the supply voltage down to below 2.0V resulted in the chip NOT going into reset - the output pins remained as low-impedance. I didn't go much below 2.0V since I don't expect to be working at that low of a supply voltage. So: I look to be OK at this point. I did have problems with the beta version of the debug adapter not working at less than about 2.5Vdc but I'm attributing that to the beta nature of the adapter. I see that production versions of the adapter are now available, so I'll order some of those for future work with the 10F32x family. But the LF322 does indeed work down to less than 2.0Vdc. I appreciate the suggestions that were made. Many thanks! dwayne -- Dwayne Reid <dwaynerKILLspamplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax http://www.trinity-electronics.com Custom Electronics Design and Manufacturing