In our previous EB832 repair, we covered the most common issue with EB832: power bar not controlling outlets due to very low-quality parts used within the unit.
This time, I am going to address yet another common EB832 problem. It gets stuck at “The EnergyBar 832 is in bootloader mode and is attempting to communicate with an Apex base unit” while flashing orange rapidly.
The symptoms of the problem are very simple. When plugged into a power outlet and Apex over aquabus cable, the power bar flashes the orange Neptune logo very rapidly, but never attaches to Apex and does not show up in the list of Apex modules. In some instances, when not plugged into Apex, the EB832 may go straight into the fallback mode with the Neptune logo flashing slower and some outlets showing as ON on the indicator panel.
To understand the reason for this behavior, we first need to understand what the EB832 is trying to do on power on. What Neptune calls the “bootloader mode” for EB832 is, in this specific case, the waiting to hear from Apex head unit so that the EB832 can be attached as a module. It will continue flashing orange rapidly until it either hears back from the head unit or times out and goes into the “Fallback Mode”. There are several reasons the EB832 cannot communicate with the head unit, for example: the AquaBus cable is bad, corrosion on USB ports, etc. But those issues are usually very easy to verify by swapping cables and ports. And if nothing helps, the problem is somewhere deeper.
How deep? Let’s find out.
After some basic troubleshooting we determined that:
- Apex head unit works correctly.
- Aquabus cable is good.
- EB832 port is in good condition. Still no communication when plugged into other aquabus ports on EB832.
From the above, we conclude that the issue is almost certainly with the EB832. Since EB832 communicates with the head unit via a USB-like physical interface, it would be useful to trace the connectivity on board of the EB832. It takes us to this small component.
This is a CAN transceiver chip that aggregates the signal from all aquabus ports and sends it off to the main controller for processing. This transceiver is our first gate from the aquabus port and into the EB832. It is not uncommon for this component to fail, especially if there was any corrosion on the port which might have shorted the wires. After all, unlike USB, AquaBus carries 12V on the pins next to the data pins.
At this point we could just try replacing this transceiver, but for the purpose of this article, we will take a step further and verify that the problem is indeed with this component without removing it from the board. This can be easily done with an oscilloscope and also with a logic analyzer. We know that the input signal is going to be a CAN encoded serial and the output is straight serial going into the UART on the controller. Thus, we could look at the input on the CAN side from the head unit and expect a decoded serial output on the Serial side of the transceiver. For this, we will build a very basic circuit where we connect jumper wires from the transceiver over to the logic analyzer. A small breadboard is not at all necessary, but it is used just in case the problem is even deeper and some additional testing is expected.
By now, we know that EB832 is waiting for a command from the head unit. It is reasonable to expect some signal coming in on the CAN side first.
Indeed, we observe some activity at an interval of less than a second. This is our head unit sending the probe to EB832 and expecting it to return an ACK.
But the ACK never comes back…
Now that we know that the input signal makes it to the transceiver, let us see if it gets decoded on the serial side.
And the answer is no. The transceiver receives the data but never decodes it on the other side. The transceiver is indeed bad and needs to be replaced. We previously verified that the transceiver gets adequate power to do its operation.
Time to remove the faulty component.
We will clean up the pads and prepare it for a replacement.
The original transceiver had fewer features. What I had in stock has additional features controlled by additional signal over the pins and that is OK because the additional features can be disabled by keeping some pins disconnected.
Let us install the new chip and test with the logic analyzer on the serial side.
Now the signal on the serial side has a similar pattern as the CAN side.
From this point, the signal from the head unit is making its way into the main controller on the EB832. After assembling and testing, we verified that the EB832 can now communicate with the head unit successfully.
We can make your Apex work again. Head over to the Mail-In Repair section to find out how to send us your device for repair.