Ultra Low Noise Jfet

Posted on by admin



Mark as Favorite 2SK2394 is N-Channel JFET, 15V, 6 to 32mA, 38mS, CP for Low-Noise HF Amplifier Applications.

Linear Integrated Systems (LIS) announces the immediate availability of its LSJ689, a 1.8 nV/Hz @ 1 kHz, low-capacitance, monolithic dual P-Channel JFET. The LSJ689 is a P-Channel complement to our monolithic dual N-Channel JFET, the LSK489. This latest addition to the LIS family of Ultra-Low-Noise JFETs provides users with many more design options than available by using only N-Channel JFETs. Additionally, the monolithic dual construction of the LSJ689 and LSK489 provides better solutions for obtaining tighter Idss (drain-source saturation current) matching and improved thermal tracking than matching individual JFETs.

Historically, P-Channel JFETs availability has declined. Complementary Single N-Channel and P-Channel JFETs have become limited to a few industry standards. Complementary monolithic dual N-Channel and P-Channel JFETs have not been offered for many years, leaving designers under supported. In response, LIS produced Single Complementary N & P Channel JFETs, the LSK170 and LSJ74 family. The LSJ689 and LSK489 provide additional options for designers. The LSJ689/LSK489 have 1/5 the capacitance of the LSK170 and LSJ74, 4 pF versus 25 pF. A unique Monolithic Dual design construction of interweaving each JFET on the same piece of silicon provide excellent matching and thermal tracking.

When used together, the LSJ689 and LSK489 are ideal for differential input stages for amplifier, phono, and preamplifier designs. Additional applications for the LSJ689 include voltage controlled resistors, thermal stable source followers, sample and hold, and current matched sources.

Ultra Low Input Current Noise Jfet

The LSJ689 is an ideal improved functional replacement for JFETs that have similar noise characteristics with reduced gate-to-drain capacitance. It is available in RoHS compliant SOT-23 6L, SOIC-A 8L and TO-71 6L package options. The LSJ689 SOT-23 6L and SOIC-A 8L packages are ideal for space-limited circuits in audio and instrumentation applications.

  1. The same Ic, so a low noise OA must be used. If lower noise than 1 nV/√Hz is necessary, JFETs can be paralleled as shown on Fig.4. If only two transistors are necessary, matched pair LSK389 from Linear Systems is a good choice. An ultra-low noise amplifier with FET.
  2. To microphone preamplifiers, the use of a low-noise, high-impedance device between the input and the op amp is needed in order to optimize performance. At first glance, one of Linear Systems’ most popular parts, the LSK389 ultra-low-noise dual JFET would appear to be a good choice for such an application. The part’s high input.

Summary of Features:

  • Complement to the Ultra Low Noise Monolithic Dual N-Channel JFET LSK489
  • Ultra Low noise (typically 1.8 nV/Hz @ 1 kHz)
  • Nearly zero popcorn noise
  • Idss (drain-source saturation current) matching to 10% max
  • Low offset/tight matching (|VGS1- VGS2| = 20 mV max)
  • Low capacitance (CISS=4 pF)
  • High input impedance
  • High breakdown voltage (BVGSS = 40 V Min)
  • Monolithic Dual (2 JFETS on one piece of silicon, better matching and thermal tracking)
  • Surface mount SOIC-A 8L versions and the smaller SOT-23 6L package
  • Lead-free/ROHS compliant

Applications:

Microphone amplifiers, phono preamplifiers, audio amplifiers and preamps, discrete low-noise operational amplifiers, battery-operated audio preamps, audio mixer consoles, acoustic sensors, sonic imaging, and instrumentation amplifiers, wideband differential amplifiers, high speed comparators, impedance converters, voltage controlled resistors, sample and hold, source followers.

NoiseNoise

Price:

US $8.66/ each (1,000 pcs TO-71 6L)

Availability:

Sampling now in SOIC-A 8L, TO-71 6L and SOT-23 6L package options

Delivery:

Factory Stock

Back to the index.
<< Go to part 2Go to part 4 >>
Ultra low noise amplifiers. (part 3)
Noise measurements of several NPN transistors. (For PNP transistors, see part5 )

On this page, the noise performance of several NPN transistors is tested.

Figure 1, this is the amplifier I use for measuring the noise of a transistor.
T1 is the transistor under test, the transistor works with 10 mA collector current.
For other collector currents, R7 and R2 must be changed.
With potentiometer R4, the output of the op-amp can be adjusted to about 6 volt.
R10 and C7 are added to prevent oscillation of the amplifier.
Diode D1 protects the transistor base-emitter junction for too high reverse voltage.
The gain of the circuit is about 1000 times (60 dB), determined by R11, R1 and the ESR (series resistance) of C5 and C6.
The output goes to another 60 dB amplifier in the test setup, and then to a spectrum analyser, as described inpart 1.
The input signal of (exact !) -140 dBV, is thus amplified about 120 dB and reaches the spectrum analyser at a level of about -20 dBV.
The level of the input noise floor is measured with regard to the 1 kHz -140 dBV test tone.
Figure 2, the amplifier for measuring the transistor noise.

Transistor typeManufacturerView datasheet
of transistor		hfe
at 25°C
Ic=10 mA		Measured input noise
of test amplifier
at 1 kHz, Ic =10 mA.
nV/√Hz		Calculated transistor
base resistance
Rbb (Ω)

Calculated Input noise
of only the transistor.
(R_circuit = 0Ω)
at 1 kHz, Ic=10 mA.
nV/√Hz

View spectrum
of measurement.
Ic=10 mA.
BC337-40PhilipsBC3374001.406116.561.397 Spectrum_BC337-40_10mA.jpg
BC550CPhilipsBC5505003.148595.613.144 Spectrum_BC550C_10mA.jpg
BD139 (old one
from '70 or '80 s)		Telefunken BD139 (datasheet
from Fairchild)		851.183? (spectrum not flat) Spectrum_BD139_10mA.jpg
ZTX618Diodes inc. / ZetexZTX6184000.47110.580.443 Spectrum_ZTX618_10mA.jpg
ZTX690BDiodes inc. / ZetexZTX690B8600.51613.260.491 Spectrum_ZTX690B_10mA.jpg
ZTX851Diodes inc. / Zetex ZTX8511850.2741.720.223 Spectrum_ZTX851_10mA.jpg
ZTX853Diodes inc. / ZetexZTX8531830.2711.620.220 Spectrum_ZTX853_10mA.jpg
ZTX857Diodes inc. / ZetexZTX8571850.2741.720.223 Spectrum_ZTX857_10mA.jpg
ZTX1048ADiodes inc. / Zetex ZTX1048A4500.3966.650.363 Spectrum_ZTX1048A_10mA.jpg
ZTX1051A Diodes inc. / Zetex ZTX1051A4500.4207.840.389 Spectrum_ZTX1051A_10mA.jpg
ZXTN25012EFLDiodes inc. / Zetex ZXTN25012EFL7800.74630.790.729 Spectrum_ZXTN25012EFL_10mA.jpg
ZXTN25050DFHDiodes inc. / ZetexZXTN25050DFH4450.48811.570.461 Spectrum_ZXTN25050DFH_10mA.jpg
2SCR533PFRARohm2SCR533PFRA2700.4358.610.405 Spectrum_2SCR533PFRA_10mA.jpg
2SCR542PFRARohm2SCR542PFRA3100.4308.350.400 Spectrum_2SCR542PFRA_10mA.jpg
2SCR552PFRARohm2SCR552PFRA3100.47110.580.443 Spectrum_2SCR552PFRA_10mA.jpg
2SD2662Rohm2SD26623600.56016.120.537 Spectrum_2SD2662_10mA.jpg

Ultra Low Noise Jfet Amplifier

This table gives an overview of the transistors tested, with the measured input voltage noise of the test amplifier, and the calculated base resistance Rbb of the transistors.
Also the voltage noise of only the transistor is calculated at 10 mA collector current.
For an explanation of the Rbb value, seepart 4 .
For more info about the testing method, seepart 1 andpart 2 .
The collector current of the transistors is in all cases 10 mA.
For the BD139, the Rbb could not be determined, because the spectrum is not flat, so over the entire spectrum we have influence of 1/f noise, and / or current noise building up a non-flat noise across input capacitor C1.
For all other transistors, the Rbb value was determined from the noise at 1 kHz.
In the next figure, the input noise voltages of all measured transistors are plotted in one diagram.
Figure 3, input noise voltage of the test setup, with different input transistors.
Also the AD797A low noise op-amp is added for comparison.
For this diagram, I compensated the values for the frequency response of the test setup.
Figure 4 shows the test setup response, and the compensation I applied for making figure 3.


Figure 4, frequency response of test setup, and the compensation curve.

Ultra Low Noise Amplifiers - JanasCard



Figure 5, noise spectrum of AD797A amplifier version 4, as described inpart 2 .
The AD797A is one of the lowest (voltage) noise op-amps available.
Figure 6, noise spectrum of the test amplifier with a ZTX851 transistor at 10 mA collector current.
Compared to the AD797A, this transistor amplifier has much lower noise.
For the test setup with the ZTX851 transistor at 10 mA, two sound files are recorded.
The first is a 1000 Hz test tone at -140 dBV, the first 10 seconds the tone is on, the last 5 seconds it is off.
ZTX851_10mA_1000Hz_-140dBV.mp3

The second recording contains music with an average level of -143 dBV (0.07μV) at the amplifier input.
ZTX851_10mA_music_-143dBV.mp3
For a good comparison, the recordings are made with the same input levels and content as with the AD797A amplifier, which you can find at the bottom ofpart 2.
As you see in the picture of the noise spectrum of the ZTX851, there is some hum visible above the noise floor.
I applied a 80 Hz high pass filter on the audio files, to remove at least the strongest component (50 Hz) which was very annoying in my speakers.

2N3460


Figure 7, here the ZTX851 is again tested at 10 mA, just like in figure 6, but the input capacitor of the test amplifier (C1) is changed from 100μF electrolytic to 100 μF plastic film capacitor.
The electrolytic capacitor has an ESR (series resistance) of about 0.5 Ω, while the plastic film capacitor has an ESR of probably below 0.01 Ω
This causes the noise floor to be about 0.5 dB lower.
But the 100 μF film capacitor is so large, that I can't place the metal lid on the test setup anymore.
This causes the level of 50 Hz (+ harmonics) interference to increase considerably, the situation got better when I wrapped some aluminium foil around the test setup, which shielded most of the interference.
But anyhow, if you ignore the 50 Hz harmonics, you see the noise floor at 1 kHz is about 0.5 dB lower then in figure 6.
By the way, the interference near 20 kHz comes from my computer and / or monitor.


Figure 8, a large 100 μF film capacitor at the input of the test amplifier.

2N3459

Once I read on the internet, that you can ruin the noise performance of a transistor, by putting a reverse current trough the base-emitter junction.
I have tested this with the ZTX851 transistor which is tested above.
A 12 volt supply was connected with the minus to the base, and the plus via a 1 kΩ resistor to the emitter.
That gives about 4 mA reverse current, this means the b-e junction comes in reverse conduction (like a zenerdiode) at -8 volt.
I let the current flow for 10 seconds, then put the transistor back in the test setup, and tested the transistor noise once again.
But the noise was exactly the same as before.
<< Go to part 2Go to part 4 >>