Hi,
Many of Quectels devices have a 3.3 - 4.4V supply voltage range, which is obviously geared towards Li-Ion batteries. It is not clear, however, if a “normal” 3.3V supply rail is acceptable. In practice, such a rail may have a +/- 3% accuracy per example and this would in theory voilate the minimum 3.3V requirement.
Can you provide some insight? Would this impact the power consumption (in case some power on the module is derived from an LDO)?
Looking forward to your reply.
Just checking in to see if there was anyone who might help?
Hi,
Yes, you are right. The reason why Quectel defines a 3.135–4.4 V supply range (3.3 V Typ.) is so that it can be connected directly to a single-cell Li-Ion/Li-Po battery (nominal 3.7 V, 4.2 V on charge).
3.135 V is the absolute minimum allowed.
However, in practice, due to tolerances (±3% → 3.20–3.40 V), the module may experience a “brown-out” if the supply voltage dips close to 3.1 V (for example, when the TX burst draws up to 2 A, the instantaneous voltage may drop).
Therefore, it is not safe to operate the module strictly at 3.3 V; although the datasheet specifies this as typical, some margin is required in real applications (around 3.5 V or higher is more reliable).
The recommended supply design target is 3.8–4.0 V. This ensures both Li-Ion compatibility and enough regulation margin.
If your system only provides a 3.3 V rail, the power supply must hold tight tolerance (±2%, i.e. 3.23–3.37 V) and you must place sufficient low-ESR bulk capacitance close to VBAT.
If possible, regulate slightly higher (3.4–3.5 V) instead of exactly 3.3 V.
During RF transmission, current consumption can reach 2 A peaks. If you observe a 200–300 mV droop at these peaks, the module may reset.
Hi Emre,
Thanks for letting me know. This certainly helps. Is the 3.8V fed into a switching regulator or an LDO? In other words: does lowering the voltage reduce power consumption (LDO) or not (switching)?
Thanks!
Kris
Hi Kris,
Quectel modules contain a high-current-drawing RF PA (Power Amplifier) and a baseband section. The VBAT pin goes directly to these blocks. This means there’s no additional step-down (buck) internally, only local LDO/analog regulators. Therefore, you determine the module input voltage (3.135–4.4 V). If your input is 5 V, for example, the LDO will directly heat the input-output difference (5–3.8 = 1.2 V). This means efficiency is low, and power consumption is high.
Using switching (buck) will increase efficiency to 85–95%. At the same load, it significantly reduces power consumption. The module’s current consumption doesn’t change linearly with voltage. To generate the same TX power, the PA requires a certain wattage.
For example: 3.8 V × 1.5 A ≈ 5.7 W
3.3 V × 1.73 A ≈ 5.7 W
In other words, the module balances its own power draw. You only affect dropout and regulator efficiency.
Therefore, you can use a high-efficiency buck regulator (≈3.8–4.0 V target) as the main supply for VBAT. You can use an LDO only if the Vin–Vout difference is small and the average current is low (≈≤0.3–0.4 A), or as an additional noise filter on the buck output. Place ≥470 µF low-ESR +100 nF close to the VBAT pin and keep droop <400 mV on conduction peaks.
Much appreciated. This helps cooking up a suitable power architecture.
