MODIFIED FENDER TWEED DELUXE

1
MODIFIED FENDER TWEED DELUXE

• ECE 445- SENIOR RESEARCH DESIGN
• Gabe Jacome
• James Milsk
• T.A Adam Gustafson

2
HISTORY

• The 1957 Fender Tweed Deluxe (5E3) was created by
  Leo Fender, based off of old radio circuits
• One of the first commercial guitar amplifiers
  made.
• Most guitarists agree that although it was a
  simple circuit, it had a great vintage sound.
• Because it was one of the first amplifiers ever
  made, the design and construction procedure was
  flawed.

3
Problems
4

• The Fender Deluxe sounds great at low volume
  levels but the output signal quickly becomes
  distorted thereafter.
• We want the ability to drive the amplifier at
  higher volume levels and retain the signals
  integrity.

5
OTHER DEFICIENCIES

• The original Fender Tweed Deluxe 60 Hz hum was
  too loud. This was caused by ground loops and
  interference in the conducting wires.

6
Power Stage Problems

• Power tubes were driven too hard.
• The specifications for the 6V6 power tube
  specify a maximum power dissipation of 14 Watts.
  The original design drove 13 Watts of power,
  which led to a short tube life.
• The screen of the power tubes, which has its own
  power rating, drew too much current, which also
  caused the tubes to burn out quickly.
• In addition, the two screens were coupled
  together. Since most tubes do not have exactly
  the same behavioral characteristics, the tubes
  were usually not matched. So, different currents
  flowed from each screen and worked against each
  other. They tried to force each other to the
  same voltage.

7
Frequency Response Problems

• Lacked a flat frequency response at mid-tone
  levels, which is supposed to evenly amplify the
  signal at all audible frequencies.
• The original amplified more low frequency signal
  at the middle tone-level. The amplification at 2
  kHz was twice as high as the amplification at 20
  kHz.

8
Original Schematic
9
(No Transcript)
10
Solutions and Modifications
11
(No Transcript)
12
Changing the bias on the second preamp stage
triode

• By removing the capacitor, it allowed the voltage
  at the cathode to oscillate in direct proportion
  to the grid signal.
• This allows the grid to cathode voltage to remain
  constant and self-bias the tube. It lowers the
  voltage gain by 50.
• In the original amp, the 3dB point of the RC
  parallel configuration is about 4Hz. The lowest
  signal coming from a guitar is well above that.
  So, the frequency response in the range of the
  guitar is practically unchanged.

13
Changing the power tube cathode bias

• By increasing the cathode resistor, we reduced
  the cathode current from 40mA to 25mA.
• This reduced the power dissipation in the power
  tubes from 13 Watts to 7-8 Watts, ensuring
  reliable function and extending the tube life.

14
Introduced a Feedback Loop

• Added a 56K ohm resistor from the output
  transformer secondary to the cathode of the 2nd
  Pre-amp. This feeds back a portion of the output
  voltage. Since it is 180 degrees out of phase
  with the voltage at the grid, it reduces the
  voltage entering the second preamp (negative
  feedback).
• Feedback loop Flattens frequency response and
  decreases distortion produced in the stages
  following the second preamp. This is caused by
  the reduction in input voltage at the second
  preamp.

15

• Replace the 1st Pre-amp triode, a 12AY7, with a
  12AX7 triode
• The original amplifier used 12AY7 tubes but they
  are no longer made. So we replaced it with a
  12AX7 tube, which is the closest available
  replacement tube. However, it has 4 times as
  much gain as the 12AY7. This drives the amp into
  cutoff sooner, but the removal of the cathode
  capacitor at the second preamp offsets this
  problem.
• Decoupled the Screen Resistor of Power Tubes
• Reduces the screen current, ensuring proper
  function well within the maximum power rating.
  This also extends tube life
• Allows each screen to operate at its own voltage
  point instead pulling at each other. This
  eliminates previous damage to the tube
• Added a Standby Switch
• Increases the lifetime of the tubes by heating
  the filaments before the main power supply is
  applied to the tubes.

16
Fabrication Concerns

• Used coaxial cable in input stages this helps
  reduce interference in low voltage areas that are
  later amplified. When the braided shielding is
  grounded, the coax acts like an antenna that
  shunts high frequency interference to ground.
• Placed plastic washers between chassis and input
  jacks This prevents grounding the input jacks
  directly to the chassis and eliminates the ground
  loops originally present.

17
More Building Alterations

• Used a star ground method Helped to eliminate
  ground loops, which added interference in the
  input.
• Twisted the wires leading to the filaments
  Because the two wires carried equal signals 180
  degrees out of phase, any 60Hz interference is
  cancelled out through destructive interference.
  The same method was applied to the speaker
  connections.

18
(No Transcript)
19
Tests and Analysis
20
Original vs. Our Modified Amp

• From the graph it is evident that the original
  Fender circuit goes well into cutoff distortion
  at volume 4.
• Our modified amp still retains the integrity of
  the original input signal.

21

• By comparing the graphs that vary the volume
  setting, it is apparent that the original fender
  amp goes into cutoff much sooner.
• It is even more distorted at volume 4 than the
  original amp is at volume 10.

22
Harmonic Distortion Comparison

• The original amp has harmonics of higher
  amplitudes by several orders.
• Our modified amplifier drastically reduces the
  amount of harmonic distortion originally output
  by the fender amp.

23

• Higher order harmonics and overtones are
  introduced in the original amplifier at earlier
  volume levels.
• These harmonics are not introduced until much
  higher volumes in the modified amplifier. Even
  at volume 10, the higher harmonics have
  comparably lower amplitudes.

24
Noise Reduction

• As demonstrated in the graph below, there is a
  clear reduction in the amount of noise generated
  in the circuit.

25
Noise Reduction Modified amplifier compared to
the original

• Through our wiring techniques, we were able to
  reduce 60Hz hum and ground loops
• This resulted in a base noise reduction of 55 on
  average

26
Comparing Frequency Response with and without a
feedback loop

• Low frequencies are attenuated more to achieve
  the desired flat mid-tone frequency response

27
Comparing frequency response with/without a
feedback loop at different tone settings
28
Conclusion
29
Achieved Goals

• We were able to reduce the amount of distortion
  for higher volume levels.
• Significantly reduced 60Hz hum.
• Flattened the frequency response for a more
  consistent amplification.
• Created a more durable and reliable circuit by
  reducing the power dissipation on the power tubes
  and protecting the tubes with the standby switch.
• Overall, produced a cleaner sounding amplifier
  with a greater volume range.