Skip to content

bladeRF Power Consumption

Jump to bottom
Brian Glod edited this page Oct 26, 2017 · 1 revision

Table of Contents

Overview

Power consumption of the bladeRF will change depending on several factors. The largest contributors are: FPGA size, FPGA utilization, expansion boards, ambient temperature, sample rate, and the LNA gain setting. There will also be individual variations between bladeRF boards due to normal device tolerances and differences between manufacturing lots; however, these should be fairly minimal by comparison. This page catalogs the power consumption of some typical bladeRF configurations.

Background and Test Setup

Although a power supply may output 5 V nominally, not all supplies (motherboards, AC power adapters, etc.) are created equally. There are variations in the exact output voltage and current sourcing capabilities; and it is common for power supplies to experience voltage droop as the load current increases.

The bladeRF under test was configured to accept power over the DC barrel jack. The DC barrel jack was connected to a Rigol DP832 bench power supply set to 5 V. A Fluke 87‑V multimeter was used to measure voltage at J49 of the bladeRF to account for any voltage drop across the wires from the power supply. Power was then computed by multiplying the current displayed on the bench supply with the voltage displayed on the DMM. The power supply and the DMM both display 3 fractional digits.

All measurements were taken at a room temperature of 72°F (22.4°C) in still air. No enclosure was used, and no heat sinks or fans of any kind were used. The bladeRF was allowed to "warm up" for approximately 10 minutes prior to recording any power readings. Stock "hosted" FPGA images were used. Transmit testing was performed with a continuous waveform at the center frequency.

A final test was performed to determine the amount of acceptable voltage droop before the bladeRF lost its connection to the host machine. This test was repeated in the opposite direction to determine the voltage point at which the bladeRF powers up and is detected by the host.

TL;DR

  • Expect a bladeRF alone to consume up to 4 W at room temperature.
  • Expect a bladeRF + XB-200 to consume up to 4.5 W at room temperature.
  • Expect a bladeRF + XB-300 to consume up to 10 W at room temperature.
  • Minimum supply voltage before disconnect: 2.950 V
  • Minimum supply voltage to power up: 3.500 V
A standard USB 2.0 port can deliver up to 2.5 W of power. A standard USB 3.0 port can deliver up to 4.5 W of power. Deviations from these standards exist, such as high-power USB ports. The bladeRF will work on many USB 2.0 ports (with limitations). Your mileage may vary! Consider using an external AC adapter if your power requirements are expected to exceed the USB host bus power specifications.

Detailed Measurements

UUT information:

UUT Type bladeRF-cli libbladeRF Firmware FPGA
fce6..062c x40 1.5.0-git-6116d090 1.8.0-git-6116d090 2.0.0 0.6.0
fd4b..7d7f x115 1.5.0-git-6116d090 1.8.0-git-6116d090 2.0.0 0.6.0

More UUTs may be added at some point in the future, time and resources permitting. Please consider contacting us if you have data you would like to contribute.

General / RF Idle

State x40 Power (W) x115 Power (W)
Unconfigured 0.717 0.717
Unconfigured, with XB-200 0.717 0.717
Unconfigured, with XB-300 0.731 0.731
Configured, FX3 GPIF idle, RF idle 2.136 2.176
Configured, FX3 GPIF idle, RF idle, with XB-200 2.136 2.177
Configured, FX3 GPIF idle, RF idle, with XB-300 2.161 2.194
Configured, FX3 GPIF active, RF idle 2.204 2.250
Configured, FX3 GPIF active, RF idle, with XB-200 2.204 2.251
Configured, FX3 GPIF active, RF idle, with XB-300 2.223 2.276

FX3 GPIF, FPGA, and RF Active

bladeRF x40

RX TX RX / TX Freq (MHz) IBW (MHz) / SR (MSPS) RX Gains (dB)
LNA / VGA1 / VGA2 		TX Gains (dB)
VGA1 / VGA2 		Power (W)
ON OFF 3800 / 3798 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.441
ON OFF 3800 / 3798 2.5 / 5 6 / 30 / 30 (max) -35 / 0 (min) 2.488
ON OFF 3800 / 3798 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.580
ON OFF 3800 / 3798 28 / 40 6 / 30 / 30 (max) -35 / 0 (min) 2.662
ON OFF 302 / 300 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.484
ON OFF 302 / 300 2.5 / 5 6 / 30 / 30 (max) -35 / 0 (min) 2.540
ON OFF 302 / 300 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.624
ON OFF 302 / 300 28 / 40 6 / 30 / 30 (max) -35 / 0 (min) 2.739
ON ON 3800 / 3798 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.628
ON ON 3800 / 3798 2.5 / 5 0 / 5 / 0 (min) -4 / 25 (max) 3.214
ON ON 3800 / 3798 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.782
ON ON 3800 / 3798 28 / 40 0 / 5 / 0 (min) -4 / 25 (max) 3.362
ON ON 302 / 300 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.686
ON ON 302 / 300 2.5 / 5 0 / 5 / 0 (min) -4 / 25 (max) 3.386
ON ON 302 / 300 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.831
ON ON 302 / 300 28 / 40 0 / 5 / 0 (min) -4 / 25 (max) 3.529
ON ON 302 / 300 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.644

bladeRF x40 & XB-200

The XB-200 filter paths were set to auto_3db to automatically select the band based on center frequency.

RX TX RX / TX Freq (MHz) IBW (MHz) / SR (MSPS) RX Gains (dB)
LNA / VGA1 / VGA2 		TX Gains (dB)
VGA1 / VGA2 		Power (W)
OFF OFF 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.269
ON OFF 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.589
OFF ON 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.081
ON ON 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.410
ON ON 54 / 50 2.5 / 5 6 / 30 / 30 (max) -4 / 25 (max) 4.228
ON ON 54 / 50 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 3.570
OFF OFF 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.255
ON OFF 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.575
OFF ON 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.100
ON ON 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.420
ON ON 225 / 222 2.5 / 5 6 / 30 / 30 (max) -4 / 25 (max) 4.246
ON ON 225 / 222 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 3.555

bladeRF x40 & XB-300

During these tests, the state of the LNA and PA on the XB-300 matched the state of the RX and TX activity, respectively. In other words, when receiving, the LNA was powered up; and when transmitting, the PA was powered up. Conversely, when not receiving and/or transmitting, the LNA and/or PA was powered down.

RX TX RX / TX Freq (MHz) IBW (MHz) / SR (MSPS) RX Gains (dB)
LNA / VGA1 / VGA2 		TX Gains (dB)
VGA1 / VGA2 		Power (W)
OFF OFF 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 2.314
ON OFF 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.133
OFF ON 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 8.584
ON ON 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 9.013

bladeRF x115

RX TX RX / TX Freq (MHz) IBW (MHz) / SR (MSPS) RX Gains (dB)
LNA / VGA1 / VGA2 		TX Gains (dB)
VGA1 / VGA2 		Power (W)
ON OFF 3800 / 3798 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.506
ON OFF 3800 / 3798 2.5 / 5 6 / 30 / 30 (max) -35 / 0 (min) 2.568
ON OFF 3800 / 3798 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.657
ON OFF 3800 / 3798 28 / 40 6 / 30 / 30 (max) -35 / 0 (min) 2.759
ON OFF 302 / 300 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.560
ON OFF 302 / 300 2.5 / 5 6 / 30 / 30 (max) -35 / 0 (min) 2.613
ON OFF 302 / 300 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.706
ON OFF 302 / 300 28 / 40 6 / 30 / 30 (max) -35 / 0 (min) 2.814
ON ON 3800 / 3798 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.708
ON ON 3800 / 3798 2.5 / 5 0 / 5 / 0 (min) -4 / 25 (max) 3.320
ON ON 3800 / 3798 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.864
ON ON 3800 / 3798 28 / 40 0 / 5 / 0 (min) -4 / 25 (max) 3.474
ON ON 302 / 300 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 2.765
ON ON 302 / 300 2.5 / 5 0 / 5 / 0 (min) -4 / 25 (max) 3.487
ON ON 302 / 300 28 / 40 0 / 5 / 0 (min) -35 / 0 (min) 2.923
ON ON 302 / 300 28 / 40 0 / 5 / 0 (min) -4 / 25 (max) 3.638
ON ON 302 / 300 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.782

bladeRF x115 & XB-200

The XB-200 filter paths were set to auto_3db to automatically select the band based on center frequency.

RX TX RX / TX Freq (MHz) IBW (MHz) / SR (MSPS) RX Gains (dB)
LNA / VGA1 / VGA2 		TX Gains (dB)
VGA1 / VGA2 		Power (W)
OFF OFF 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.349
ON OFF 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.672
OFF ON 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.190
ON ON 54 / 50 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.500
ON ON 54 / 50 2.5 / 5 6 / 30 / 30 (max) -4 / 25 (max) 4.331
ON ON 54 / 50 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 3.655
OFF OFF 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.347
ON OFF 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.669
OFF ON 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.234
ON ON 225 / 222 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 4.557
ON ON 225 / 222 2.5 / 5 6 / 30 / 30 (max) -4 / 25 (max) 4.365
ON ON 225 / 222 2.5 / 5 0 / 5 / 0 (min) -35 / 0 (min) 3.641

bladeRF x115 & XB-300

During these tests, the state of the LNA and PA on the XB-300 matched the state of the RX and TX activity, respectively. In other words, when receiving, the LNA was powered up; and when transmitting, the PA was powered up. Conversely, when not receiving and/or transmitting, the LNA and/or PA was powered down.

RX TX RX / TX Freq (MHz) IBW (MHz) / SR (MSPS) RX Gains (dB)
LNA / VGA1 / VGA2 		TX Gains (dB)
VGA1 / VGA2 		Power (W)
OFF OFF 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 2.424
ON OFF 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 3.227
OFF ON 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 8.614
ON ON 2402 / 2400 28 / 40 6 / 30 / 30 (max) -4 / 25 (max) 9.204

Minimum Supply Voltage

The DC barrel voltage was continuously reduced until the host lost its connection with the bladeRF device. This minimum voltage was determined to be 2.950 V, matching the datasheet of the LM2014x regulators.

This test was repeated in the opposite direction. The DC barrel voltage was continuously increased until the host was able to detect the bladeRF device. This minimum power on voltage was determined to be 3.500 V.

These experiments are provided for informational purposes only. Nuand does not recommend operating at these voltage extremes. And as always, your mileage may vary!

Clone this wiki locally