Hi Mathias, I'm assuming you're using a coin cell for back up, so wondering if one of the diodes shouldn't be in series with the battery, with both cathodes to V_BCKP? Nigel GM8PZR I've updated the repository on github with the latest design files. The breakout board is now final as far as I'm concerned. I managed to implement all the requested features, this includes the backup battery and also the buffered PPS output with SMA footprint. Next step on my side will be to order a small batch of PCBs. There's not much on them and I guess I will end up with around 25? for 5 boards fully assembled, shipping not included. I eventually removed the I2C pull-ups again, thanks to Patrick Tudor for the hint. The footprints are still there, I only removed them from the BOM. Schematic and a screenshot of the top and bottom sides attached. BR, Matthias

Hello to the group Nigel is correct the way the schematic is drawn the Regulator will feed the battery and most likely cause it to burst. It should be 1 diode in series with the regulator and 1 in series with the battery so that the regulator can not charge the battery. Nice design you have created. Regards Paul WB8TSL On Tue, Dec 15, 2020 at 11:55 AM Nigel gm8pzr via time-nuts < time-nuts@lists.febo.com> wrote: > Hi Mathias, > I'm assuming you're using a coin cell for back up, so wondering if one of > the diodes shouldn't be in series with the battery, with both cathodes to > V_BCKP? > Nigel GM8PZR > > > I've updated the repository on github with the latest design files. The > breakout board is now final as far as I'm concerned. > > I managed to implement all the requested features, this includes the > backup > battery and also the buffered PPS output with SMA footprint. > > Next step on my side will be to order a small batch of PCBs. There's not > much > on them and I guess I will end up with around 25? for 5 boards fully > assembled, shipping not included. > > I eventually removed the I2C pull-ups again, thanks to Patrick Tudor for > the > hint. The footprints are still there, I only removed them from the BOM. > > Schematic and a screenshot of the top and bottom sides attached. > > BR, > Matthias > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. >

On Dienstag, 15. Dezember 2020 19:22:14 CET paul swed wrote: > Hello to the group Nigel is correct the way the schematic is drawn the > Regulator will feed the battery and most likely cause it to burst. > It should be 1 diode in series with the regulator and 1 in series with the > battery so that the regulator can not charge the battery. Yes, that's what I discussed with Nigel off-list. The two diodes in series create a voltage difference of 1.4V, so as long the battery voltage is above 1.9V, no current will flow through the diodes at all and the battery is safe. However, there is a greater-than-zero probability that a board will just be dropped into a drawer with the battery in place, fully draining it eventually. If it's then powered up again, chances are that a substantial current will flow into the battery and potentially cause a fire, explosion, etc. I've therefore updated the design and put a small Schottky diode in series with the battery. I selected a part with small leakage current ( 0.5µA typical). But of course a regular silicon diode like 1N4148WS will work as well, at the expense of wasting some of the energy of the coin cell. > Nice design you have created. Thanks - Hopefully it will also work :-) > Regards > Paul > WB8TSL > > On Tue, Dec 15, 2020 at 11:55 AM Nigel gm8pzr via time-nuts < > > time-nuts@lists.febo.com> wrote: > > Hi Mathias, > > I'm assuming you're using a coin cell for back up, so wondering if one of > > the diodes shouldn't be in series with the battery, with both cathodes to > > V_BCKP? > > Nigel GM8PZR > > > > > > I've updated the repository on github with the latest design files. The > > breakout board is now final as far as I'm concerned. > > > > I managed to implement all the requested features, this includes the > > backup > > battery and also the buffered PPS output with SMA footprint. > > > > Next step on my side will be to order a small batch of PCBs. There's not > > much > > on them and I guess I will end up with around 25? for 5 boards fully > > assembled, shipping not included. > > > > I eventually removed the I2C pull-ups again, thanks to Patrick Tudor for > > the > > hint. The footprints are still there, I only removed them from the BOM. > > > > Schematic and a screenshot of the top and bottom sides attached. > > > > BR, > > Matthias > > _______________________________________________ > > time-nuts mailing list -- time-nuts@lists.febo.com > > To unsubscribe, go to > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > > and follow the instructions there. > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and follow > the instructions there.

Yes you did the right thing. I am a drop it in the drawer person. Hate when those batteries make a bang. On Tue, Dec 15, 2020 at 5:06 PM Matthias Welwarsky <time-nuts@welwarsky.de> wrote: > On Dienstag, 15. Dezember 2020 19:22:14 CET paul swed wrote: > > Hello to the group Nigel is correct the way the schematic is drawn the > > Regulator will feed the battery and most likely cause it to burst. > > It should be 1 diode in series with the regulator and 1 in series with > the > > battery so that the regulator can not charge the battery. > > Yes, that's what I discussed with Nigel off-list. The two diodes in series > create a voltage difference of 1.4V, so as long the battery voltage is > above > 1.9V, no current will flow through the diodes at all and the battery is > safe. > > However, there is a greater-than-zero probability that a board will just > be > dropped into a drawer with the battery in place, fully draining it > eventually. > If it's then powered up again, chances are that a substantial current will > flow into the battery and potentially cause a fire, explosion, etc. > > I've therefore updated the design and put a small Schottky diode in series > with the battery. I selected a part with small leakage current ( 0.5µA > typical). But of course a regular silicon diode like 1N4148WS will work as > well, at the expense of wasting some of the energy of the coin cell. > > > Nice design you have created. > > Thanks - Hopefully it will also work :-) > > > Regards > > Paul > > WB8TSL > > > > On Tue, Dec 15, 2020 at 11:55 AM Nigel gm8pzr via time-nuts < > > > > time-nuts@lists.febo.com> wrote: > > > Hi Mathias, > > > I'm assuming you're using a coin cell for back up, so wondering if one > of > > > the diodes shouldn't be in series with the battery, with both cathodes > to > > > V_BCKP? > > > Nigel GM8PZR > > > > > > > > > I've updated the repository on github with the latest design files. The > > > breakout board is now final as far as I'm concerned. > > > > > > I managed to implement all the requested features, this includes the > > > backup > > > battery and also the buffered PPS output with SMA footprint. > > > > > > Next step on my side will be to order a small batch of PCBs. There's > not > > > much > > > on them and I guess I will end up with around 25? for 5 boards fully > > > assembled, shipping not included. > > > > > > I eventually removed the I2C pull-ups again, thanks to Patrick Tudor > for > > > the > > > hint. The footprints are still there, I only removed them from the BOM. > > > > > > Schematic and a screenshot of the top and bottom sides attached. > > > > > > BR, > > > Matthias > > > _______________________________________________ > > > time-nuts mailing list -- time-nuts@lists.febo.com > > > To unsubscribe, go to > > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > > > and follow the instructions there. > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts@lists.febo.com > > To unsubscribe, go to > > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com and > follow > > the instructions there. > > > > > > _______________________________________________ > time-nuts mailing list -- time-nuts@lists.febo.com > To unsubscribe, go to > http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com > and follow the instructions there. >