Hi All, Just wanted to share some issues we are seeing with low power deployments of BR1 Mini units.
We build up bespoke solar powered systems for remote monitoring, all generally connected to a hosted FusionHub. Mainly using 4G but some bonded using Starlink.
What we are finding is that when a system goes flat, from extended lack of sun, and the Lithium Battery goes into BMS Protection at 10.5v, the BR1 Mini shuts off, as expected.
But from time to time, the system recovers, battery comes back online, and the voltage increases, but the BR1 Mini doesn’t recover.
The unit powers up, but shows a red status light, and no connection to InControl.
Although we are seeing zero kb sessions on Cisco Jasper, the unit just doesnt come back online until someone physically goes to the site and cycles the power on the Mini.
Event logging in the mini just shows Device Offline, and then when it comes back online it just connects up and shows no errors.
We are currently bench testing a Mini at the moment, to see if we can replicate it before we contact Peplink Support, but I thought it would be worthwhile to post to the forum and see what others think.
Cheers
Danny Ainsworth
Thanks for sharing the details.
I’m wondering if this could be related to the power condition when the battery gets too low. For example, if the power is not completely cut off and the BR1 Mini is still receiving a very low voltage, it may leave the device in an unstable state. When the battery starts charging again, the voltage may slowly come back up, but the device may not get a clean power cycle to recover properly.
Since you are able to reproduce this on the bench, could you help check:
- What voltage is actually reaching the BR1 Mini when the battery gets low and during recovery?
- Is the power completely cut off, or is there still some voltage supplied to the device?
- If you wait until the voltage is stable before powering the BR1 Mini, does the same issue still happen?
It would also be helpful if you could share the brand/model and specifications of the solar system, battery/BMS and charge controller , and how the BR1 Mini is connected to them.
You can also create a support ticket with us. Our Support team can work with you on the testing and take a closer look if the issue can be reproduced.
1 Like
Hi Danny,
Can I just confirm you’re using a 12 V LiFePO4 battery bank? I noticed some of these installations also have Starlink Minis connected to the same battery system.
I’d be inclined to investigate the battery/BMS and power recovery behaviour before assuming it’s a BR1 Mini issue.
If the LiFePO4 BMS is actually reaching its low-voltage protection threshold and disconnecting at around 10.5–10.8 V, the subsequent recovery may not necessarily present the BR1 with a nice clean power-on event.
The battery voltage can recover enough for the BMS to reconnect, the BR1 starts booting, loads come back on, and the voltage potentially sags again. Add a Starlink Mini starting at the same time and you’ve got another significant load appearing during that recovery period. The BR1 could therefore be seeing a brownout/restart condition rather than a clean off/on power cycle.
That would also potentially explain why physically cycling power restores it.
Personally, I wouldn’t want the LiFePO4 BMS low-voltage protection to be the normal mechanism shutting the system down. I’d use the solar/load controller to disconnect the load at a higher voltage, leaving the battery BMS as the final protection.
For example, if you have to stay with a 12 V Lifepo4 system, something like a Victron MPPT/load-control arrangement could disconnect the loads well above the battery’s BMS cutoff and then reconnect them at a sensible recovery voltage.
I’d also be inclined towards a 24 V battery system for installations involving Starlink. Apart from the additional voltage headroom, you’re roughly halving the DC current for the same power, which makes voltage drop and startup behaviour considerably easier to manage.
Your bench test should be interesting. I’d specifically try reproducing the BMS disconnect/reconnect rather than simply switching the BR1’s DC supply off and back on. I suspect those two tests may give very different results.
1 Like