The choice between the two P-Bus battery management shunts is not based on the size of the battery bank, but on the highest current that can flow through the measuring point. That is a different calculation — and it regularly leads to a different result.
NoteBefore making a choice, it is worth asking whether a shunt is even necessary: A BullTron battery with P-Bus authorization reports its own values. The trade-off is discussed in the article Shunt or BMS data: where the state of charge comes from.
What both can do equally well
The SHX 348 and SHX 648 sit in the battery's negative cable and report current, voltage, and capacity to the P-Bus network. They do not differ in four key points — and it is these four that separate a battery management shunt from a simple current measurement.
Measuring range from 10 milliamperes
Both start at 10 mA. That sounds like a footnote, but it is the most important practical value on the entire device. Currents that drain a bank unnoticed lie in this range: a device in standby, forgotten electronics, or a consumer that no one has on their radar.
A device with 30 milliamperes of quiescent current drains about 0.7 ampere-hours per day and five per week from the bank. A shunt that only starts evaluating at nearly an amp shows a clean zero — yet the bank is empty after four weeks without the cause being found. Only resolution in the milliampere range makes such a search possible at all.
±0.5 percent across the entire range
The second half of the statement is the decisive one: The accuracy of ±0.5 percent applies to both across the entire measuring range — for the SHX 648, therefore, across five orders of magnitude, from milliampere quiescent current up to 600 amperes.
Why this is more than just a line on a data sheet: The state of charge is not a measurement, but an ongoing calculation. Every single measured value goes into a continuous ampere-hour balance, and measurement errors accumulate over days rather than canceling each other out. A shunt that is only accurate at high currents therefore provides a display that slowly drifts during normal standby operation — not abruptly, but creeping, and thus particularly difficult to notice.
Starter battery voltage without a second shunt
Both shunts measure the starter battery voltage directly as well. A second shunt is not necessary for this — an assumption that persists stubbornly and makes planning unnecessarily expensive.
This covers the most common practical failure case: a starter battery that is no longer being recharged for weeks because the isolating relay or charging converter is no longer working. The voltage reports this condition early on — long before the start attempt where it would otherwise be noticed.
Battery temperature directly at the shunt
The temperature is also recorded directly at the measuring point: The TEMP-BT temperature sensor is connected to the shunt, and the temperature appears on the system monitor in addition to current, voltage, and capacity. A separate temperature interface is not required for this.
This is more than just a display. The usable capacity of a battery depends on its temperature, and with LiFePO4, charging below freezing is the point at which damage occurs. Anyone who does not know the temperature also only knows the state of charge approximately.
Both also have a galvanically isolated P-Bus connection. This makes it possible to monitor a battery group that is disconnected from the vehicle electrical system — for example, an emergency battery for radio equipment — without creating an unwanted connection via the measurement cable.
The difference in numbers
The SHX 348 is designed for 300 A continuous current and can tolerate 600 A for one minute and 1500 A for half a second. Connection is via an M8 bolt, the operating voltage is 8–64 V, and current consumption is 20 mA or 5 mA in sleep mode. It measures 118 x 40 x 65 mm.
The SHX 648 carries 600 A continuous current, briefly 800 A for one minute and 2500 A for half a second. Its connection bolt is M16, the operating voltage 8–60 V, the current consumption 6 mA at 12 V. At 185 x 44 x 75 mm, it is correspondingly larger.
The connection bolt is more than just a number on the data sheet: M8 and M16 require different cable lugs, and those who only choose them during installation regularly have the wrong ones on hand.
How to determine the decisive current
Three values belong in the calculation, and the largest of them is decisive:
- Continuous current during normal operation. The sum of what is usually running at the same time.
- Start-up peaks. What does the largest consumer pull when starting? Compressors, pumps, and inverters under load are significantly above their operating value.
- Short-circuit current. What must the measuring point withstand until the main fuse blows? With LiFePO4 banks, this value is significantly higher than with lead-acid.
An electrical system with cooling, lighting, pumps, and a small inverter usually remains well below 300 A and is well taken care of with the SHX 348. If bow thrusters, anchor windlasses, or a large inverter are added, peaks above that are the norm — then the measuring point should be designed to be larger accordingly.
WarningA shunt that is selected too small does not limit the current; it gets hot. And heating changes the resistance of the measuring element and thus the measurement — the device does not fail, it quietly measures incorrectly.
Connection and accessories
Both shunts draw their power supply via the existing voltage measurement cable — an additional supply cable is not necessary. The P-Bus connection is made via M12 connectors.
With the SHX 648, the additional question arises as to how the load side is distributed. For this, there is the shunt connection bar SAS 4, which provides several cable connections to match the M16 bolt. This is not only more convenient but also the clean solution for the basic rule of every shunt installation: Exactly one cable goes to the battery negative terminal, namely the one to the shunt; everything else is on the other side.
When a second shunt actually becomes necessary
A second shunt is not needed for the voltage and temperature of the second battery — the existing one does both. A second shunt is set for exactly one reason: if a full current balance with state of charge is also desired for the second bank.
This is rarely necessary for a classic starter battery. It sees very high currents for a short time, but hardly any continuous load, and the question asked of it is not "how full is it," but "is it there when I need it" — and the voltage answers that. The situation is different where the second bank does more than just start, for example as a buffer for a winch or as a second consumer bank.
To be distinguished from this is the energy shunt: If the yield of a solar system or the consumption of an individual device is to be recorded, this is not a battery measurement, but an energy measurement — for this, an energy shunt is placed next to the battery management shunt, not as a replacement.
Typical errors
- Designed by bank size instead of current. A large bank with small consumers does not need a large shunt — but a small bank with a bow thruster does.
- Planned a second shunt for the starter battery voltage. The existing shunt already records the voltage and temperature of the second battery.
- Short-circuit current not considered. With lithium, it is significantly higher than what lead-acid systems provide.
- Cable lugs only ordered during installation. M8 and M16 are different worlds.
- A ground remains at the battery terminal. The most common installation error of all — and the one noticed latest.
- Did not pay attention to the lower measurement limit. Without resolution in the milliampere range, any search for inexplicable losses remains fruitless.
