Difference between pages "RF-Amp" and "A Termination Insensitive Amplifier for Bidirectional Transceivers"

From Land Boards Wiki
(Difference between pages)
Jump to navigation Jump to search
 
 
Line 1: Line 1:
[[File:RF-Amp_P1943-720px.jpg]]
+
[[file:W7ZOI_TIA_3D.png]]
  
== RF Amplifier Features ==
+
== W7ZOI Design ==
  
* From [https://zl2ctm.blogspot.com/2020/11/go-qrp-portable-ssb-rig.html Charlie Morris' (ZL2CTM) Go QRP Portable SSB Rig]
+
Paper - [http://w7zoi.net/bidirectional_matched_amplifier.pdf A Termination Insensitive Amplifier for Bidirectional Transceivers] - W7ZOI (Wes Hayward) design
** Charlie references Solid State Design for the Radio Amateur (pp 19-20)
 
* Single 2N3904 NPN transistor
 
** Ft = 300 MHz (Gain Bandwidth Product)
 
*** Theoretical gain
 
**** +20 dB at 30 MHz
 
**** +29.5 dB at 10 MHz
 
* +22 dB gain @12V, +25dB gain @14V
 
* Input connectors: SMA or BNC
 
* +12V nominal power
 
* 49x49mm card
 
* 4x 4-40 mounting holes
 
  
== RF Amplifier Design ==
+
* Termination Insensitive - input impedance that does not depend upon the output load
 +
* Bi-directional - amplifiers in both directions
 +
* 50 ohms in/out
 +
* BNC or SMA Connectors
 +
* Transformer-less
 +
* Power applied to one half at a time - determines signal flow direction
 +
* Use
 +
** Between RF/LO mixer (IF output/input) and crystal filter
 +
** Between Crystal Filter and Product Detector / Balanced Modulator
  
=== Schematic ===
+
[[FILE:W7ZOI_SCHEMATIC.PNG]]
  
[[file:RF_Amp_Schematic-4.PNG]]
+
== Schematic in KiCAD ==
  
== LT Spice Simulation ==
+
* Single channel
 +
* With SPICE directives
  
* [https://github.com/land-boards/lb-boards/blob/master/HamRadio/RF-Amp/LTSpice/2n3904%20amp.asc LTspice Simulation] - GitHub source file
+
[[file:W7ZOI_SCHEMATIC_KiCAD.PNG]]
** +28.4 dB at 9 MHz
 
  
[[File:RF-AMP-LTSPICE_XFMRS.PNG]]
+
* Both channels
 +
* With SPICE directives
  
== Charlie Morris Design ==
+
[[file:W7ZOI_SCHEMATIC_KiCAD-2.PNG]]
  
* From Charlie's notes with mods for my use
+
== NGSPICE Simulation ==
** [https://zl2ctm.blogspot.com/2020/11/go-qrp-portable-ssb-rig.html Charlie Morris' (ZL2CTM) Go QRP Portable SSB Rig]
 
** Charlie describes the design in detail in his video [https://www.youtube.com/watch?v=CHdtoupH2Vg Simple SSB Rig: Part 6 - IF Amplifiers] (Feb 2021)
 
** Based on the Class A RF Amplifier in [https://www.amazon.com/Solid-State-Design-Radio-Amateur/dp/0872590402 Solid State Design for the Radio Amateur] pp 19-20
 
* [https://www.mouser.com/datasheet/2/308/1/2N3903_D-2310199.pdf 2N3904 data sheet]
 
* [https://www.electronics-tutorials.ws/amplifier/emitter-resistance.html Emitter Resistance] - helpful paper
 
  
=== Beta DC ===
+
* NGSPICE run from KiCAD
 +
** [http://ngspice.sourceforge.net/ngspice-eeschema.html Tutorial: ngspice simulation in KiCad/Eeschema]
 +
* +24 dB @12V
 +
* 2N3904 Alternate Node Sequence (MMBT3903 - SOT23)
 +
** KiCAD order 1 2 3 is the SPICE normal order
 +
** NGSPICE order: 3 1 2
  
* Geometric mean min/max beta at operating current
+
[[file:W7ZOI_Simulation_KiCAD-3.PNG]]
** =sqrt(100*300) = 173
 
  
=== Beta AC ===
+
* Closely matches
  
* Gain bandwidth product divided by operating frequency
+
[[file:W7ZOI_Gainz.PNG]]
** Assume operating frequency of 9 MHz (IF frequency)
 
** = 300/9 = 33.3
 
  
=== DC Operating Point ===
+
* Across Band
 +
[[file:W7ZOI_Simulation_KiCAD_S2.PNG]]
  
* Max HFE RF gain at CE current of 10 mA
+
== CAD ==
** If Vce = 6V, this is 60 mW power dissipation
 
* Assume Ve (voltage across emitter resistor) = 1/10 Vcc = 12V/10 = 1.2V
 
** R3 is Re (emitter resistor) = 1.2V/0.01A = 120 ohms
 
* VCE = 0.7V (typical from data sheet)
 
* V(emitter) at 10% of Vcc rule of thumb = 1.2V
 
* V(base) = V(emitter) + VCE = 1.9V
 
* Base current is collector current divided by Beta DC
 
** Biasing resistors = 10x current needed by base current
 
** 10 mA in C-E, beta DC less = 10 mA/173 = 58 uA
 
** 10x the current in the biasing resistors = 580 uA (calculated)
 
* R2 is 1.9V at 580 uA = 3.29K use 3.3K
 
* R1 sources current to R2 and transistor base
 
** Voltage = Vcc (12V) - 1.9V = 10.1V
 
** Current = 577 uA + 58 uA = 635 uA
 
** R1 = 10.1 / .635 mA = 15.9K, use 15K
 
  
==== Measured DC operating point ====
+
[[FILE:W7ZOI_CAD.PNG]]
  
* Measured with no input
+
== SMT Build ==
* Vcc = 11.96V
 
* Current draw = 12 mA
 
** Expected 11 mA - close enough
 
* +BUFF = 11.84V
 
** 0.12V which is 12 mA through R4 10 ohms - expected
 
* V emitter = 1.41V
 
** 12 mA through 120 ohm = 1.44V - close
 
* V on input divider = 2.06V
 
** Vbase + 0.7V - close
 
  
=== Input resistance ===
+
[[FILE:W7ZOI_TIA_SMD_3D.png]]
  
* Xc for 0.1uF cap from emitter to ground
+
[[FILE:W7ZOI_TIA_SMT_CAD.PNG]]
** C=0.1uF
 
** F=10MHz
 
** 1/2*pi*F*C = 0.16 ohms
 
* Parallel resistors R1, R2 paralleled with transistor input impedance
 
** R1=15K, R2=3.3K
 
** Transistor resistance = Beta AC (33.3) times re
 
*** re = 26 / Ie (10 mA in mA) = 26/10 = 2.6
 
*** SSDRA uses 25 as constant - close enough
 
*** Beta AC * re = 33.3*2.6 = 83.3 ohms - predominates
 
** All in parallel are 80.8 ohms
 
  
=== Input/Output Transformers ===
+
== Videos ==
  
* [http://toroids.info/FT37-43.php FT37-43 Toroid]
+
<video type="youtube">JjcxEYk9atg</video>
  
[[file:FT37-43_10_Turns.PNG]]
+
<video type="youtube">T8lq8Qtjqe0</video>
 
 
==== Tracks ====
 
 
 
[[file:RF-Amp-tracks.PNG]]
 
 
 
==== Input Transformer ====
 
 
 
* Input Transformer (T1 on Charlie's - T2 on this board)
 
* Need to calculate turns ratio
 
* 50:80.8 Ohms
 
* n = sqrt(Zout/Zin)sqrt(80.8/50) = 1.27 turns ratio
 
* Turns choices
 
* Minimum number of turns
 
* Rule of thumb - want Xl (coil impedance smallest value) to be least 4-5X the load
 
** Load = 80.8 ohms
 
** 5 * 80.8 ohms = 404.2 ohms minimum
 
*** More turns = larger capacitance and drops bandwidth
 
** Toroid is FT37-43
 
** From [http://toroids.info/FT37-43.php Toroid page]
 
***  Xl = 404.4 at 9 MHz is 4.5 turns, round up to 5
 
** Try nearest integer numbers turns ratios
 
*** 5:6 = 6% error
 
*** 6:8 = -4.6%
 
*** 7:9 = -1.1% << good choice
 
*** 8:10 = +1.7%
 
*** 9:11 = +4.0%
 
*** 10:13 = -2.19%
 
* Use 7:9 turns ratio for optimal input transformer
 
 
 
[[file:RF-Amp-T2.PNG]]
 
 
 
==== Output Transformer ====
 
 
 
* Output transformer (T2 on Charlie's - T1 on this board)
 
* T2 - different than Charlie's design since my Crystal filters are all 50 ohms in/out
 
* SSDRA suggest presenting 200 ohm load to the collector
 
** Can't find reference in SSDRA
 
** Reflecting back 50 ohms load to 200 ohm collector...
 
* 200:50 ohms
 
* n = sqrt(200/50) = 2.0:1 turns ratio
 
* 10:5 turns
 
** 10 turns primary (on transistor collector)
 
*** 10 turns = 35 uH
 
** 5 turns secondary (towards output)
 
*** 5 turns = 8.75 uH
 
** 15 turns = 9.5 in
 
 
 
[[file:RF-Amp-T1.PNG]]
 
 
 
=== Charlie's Notes ===
 
 
 
[[FILE:IF Amp_0046A.jpg]]
 
 
 
[[FILE:IF Amp_0046B.jpg]]
 
 
 
[[FILE:IF Amp_0046C.jpg]]
 
 
 
[[FILE:IF Amp_0047A.jpg]]
 
 
 
[[FILE:IF Amp_0047B.jpg]]
 
 
 
[[FILE:IF Amp_0047C.jpg]]
 
 
 
== NanoVNA Measurements ==
 
 
 
* '''Goal''': Measure RF-Amp performance using a [[NanoVNA]] running [https://nanovna.com/?page_id=90 NanoSaver software on PC]
 
* S21 (gain) needs to be measured with a [[RF_Attenuators#40_dB_Attenuator|40 dB attenuator]] on input to RF-Amp to avoid compression on the output
 
* S11 (reflection) input impedance can't be measured with input [[RF_Attenuators#40_dB_Attenuator|40 dB attenuator]] because S11 just ends up measuring the attenuator
 
** Output should be terminated to 50 ohms for S11 measurement
 
* DC current = 12 mA
 
 
 
=== Measure S21 ===
 
 
 
* Put [[RF_Attenuators#40_dB_Attenuator|40 dB attenuator]] on RF-Amp input, measure S21 at output
 
** [[NanoVNA]] provides 50 ohm load to RF-Amp to properly terminate output
 
* Measure S21 with 9:11 input transformer
 
** S21 @ 100 KHz = -8 dB dB
 
** S21 @ 1.45 MHz = 35.4 dB (peak gain)
 
** S21 @ 9.1 MHz = 24.3 dB
 
** S21 @ 16 MHz = 20.1 dB
 
** S21 @ 30 MHz = 12.7 dB
 
* Peak gain justifies use of 40 dB attenuator to protect [[NanoVNA]]
 
 
 
[[file:RF-Amp_S21_40dBAttenInput_1-30MHz.png]]
 
 
 
==== LTspice vs NanoVNA ====
 
 
 
* LTspice simulation was pretty similar to [[NanoVNA]] results
 
** -10 dB at 100 KHz
 
** +32 dB at peak
 
** Lower output at higher frequencies
 
 
 
[[file:RF-Amp_S21_LTspice-vs-NanoVNA_1-30MHz.png]]
 
 
 
=== Measure Input Compression ===
 
 
 
* Is there compression if the [[NanoVNA]] drives the input directly?
 
** Test by driving directly from NanoVNA set to CW = 9 MHz
 
** Measured output with scope - not clipped at 9 MHz
 
*** Approx. 1Vpp input = +22.1 dBm gain which matches the S21 with the attenuator on the input
 
*** Vpp = 12.4V with 50 Ohm load resistor
 
** Starts clipping at 7 Mhz and down
 
* Therefore, can measure input impedance at 9 MHz
 
* Other evidence of compression
 
** Compare S21 gain with no input attenuator, put external [[RF Attenuators|40 dB RF Attenuators]] on output of RF-Amp to protect [[NanoVNA]] input
 
** S21 shows lower gain in lower frequencies so clipping/compression is happening
 
** Was: 35 dB at 1.4 MHz
 
** Is: 23.1 dB at 1.5 MHz
 
* Due to compression can't accurately measure lower frequencies with attenuator at output
 
* Compression below 7 MHz matches what was on scope
 
 
 
[[file:RF-Amp_S21_40dBAttenOutput_1-30MHz.png]]
 
 
 
=== Measure Input Impedance ===
 
 
 
* Shows VSWR at 14.4 MHz = 1.56:1
 
* At 9 MHz
 
** VSWR = 1.7:1
 
** Impedance = 81-j10
 
 
 
[[file:RF-Amp_AttenOutput_VSWR_2021_1-30MHz.png]]
 
 
 
=== Change Input Transformer turns ratio ===
 
 
 
* Above had 9:11 turns ratio
 
* Change to 7:9 turns ratio
 
* Slightly better gain at higher frequencies
 
* Was: S21 @ 30 MHz = 12.7 dB
 
* After:  S21 @ 30 MHz = 15.3 dB
 
*  Small additional gain at 8 MHz
 
** Was: S21 @ 9.1 MHz = 24.3 dB
 
** After: S21 @ 9.1 MHz = 24.8 dB
 
 
 
[[file:RF-Amp_S21_40dBAttenInput_Turns7to9_1-30MHz.png]]
 
 
 
* New turns improved the input VSWR slightly
 
* Was: At 9 MHz, VSWR = 1.7:1, Impedance = 81-j10
 
* After: At 9 MHz, VSWR = 1.6:1, Impedance = 76.7-j12
 
 
 
[[file:RF-Amp_vswr_40dBAttenInput_Turns7to9_1-30MHz.png]]
 
 
 
==== Tune input transformer====
 
 
 
* Isolate output by replacing output transformer with 200 resistor
 
* Add one more output winding to input transformer T2 (7:10)
 
* VSWR nearly 1.04:1 at 11.1 MHz
 
* -19 dB return loss at 9 MHz VSWR = 1.249:1
 
 
 
[[file:RF-Amp_VSWR_1-30MHz_7to10Turns.png]]
 
 
 
* With output transformer
 
* Slightly better with 1 extra winding
 
 
 
[[file:RF-Amp_VSWR_1-30MHz_7to10Turns-2.png]]
 
 
 
== Video ==
 
 
 
<video type="youtube">CHdtoupH2Vg</video>
 
 
 
<video type="youtube">YJTsWV2kzFY</video>
 
 
 
<video type="youtube">xPFzFhM0ojE</video>
 
  
 
== Assembly Sheet ==
 
== Assembly Sheet ==
  
* [[RF Amplifier Assembly Sheet]]
+
* [[W7ZOI TIA Assembly Sheet - Rev 1]]

Revision as of 10:30, 12 November 2021

W7ZOI TIA 3D.png

W7ZOI Design

Paper - A Termination Insensitive Amplifier for Bidirectional Transceivers - W7ZOI (Wes Hayward) design

  • Termination Insensitive - input impedance that does not depend upon the output load
  • Bi-directional - amplifiers in both directions
  • 50 ohms in/out
  • BNC or SMA Connectors
  • Transformer-less
  • Power applied to one half at a time - determines signal flow direction
  • Use
    • Between RF/LO mixer (IF output/input) and crystal filter
    • Between Crystal Filter and Product Detector / Balanced Modulator

W7ZOI SCHEMATIC.PNG

Schematic in KiCAD

  • Single channel
  • With SPICE directives

W7ZOI SCHEMATIC KiCAD.PNG

  • Both channels
  • With SPICE directives

W7ZOI SCHEMATIC KiCAD-2.PNG

NGSPICE Simulation

W7ZOI Simulation KiCAD-3.PNG

  • Closely matches

W7ZOI Gainz.PNG

  • Across Band

W7ZOI Simulation KiCAD S2.PNG

CAD

W7ZOI CAD.PNG

SMT Build

W7ZOI TIA SMD 3D.png

W7ZOI TIA SMT CAD.PNG

Videos

Assembly Sheet