Mesa Rectifier Design Concepts: Introduction, Input Stage, and Clean Channel

Posted: September 14, 2015 in Mesa Boogie Dual Rectifier
Tags: , , , , , , , , , , , , , , , , , , , , , ,
  1. Input Stage
  2. Stage 2: Go With The Flow
  3. Stage 3: Clean It Up
  4. Skip The Nerd Talk: Putting It All Together

I’ve been thinking a lot of about modifications recently and I’ve been going over the schematic for the 3 Channel Dual Rectifier on a regular basis for quite awhile.  I was recently thinking back to when I was starting out and how it was difficult to understand why or what a part of the amp was doing.  I don’t consider myself an expert.  I was taught solid state theories in school and had to buy books and read websites regarding tubes and how they apply to guitar amps.  I wished there was more information about the design concepts.  So, I thought I’d break down the preamp for the Dual Rectifier into the stage circuits and just talk about the general, overall concept of what each does on its own and together with its neighbors.  This might help the users with tones and circuit designers and hobbyists with ideas and insight.

As we go through, understand that too much of a good thing is not always good.  If the lows are too emphasized, blocking distortion can occur.  If the highs are too emphasized, Miller capacitance will make it oscillate.  If the gain between stages is too great, the end result can be both or either of those, plus noise and a lack of dynamics (put a distortion pedal in front of your distortion channel and see what I mean).  Taking a full, harmonically rich signal and amplifying it again takes some finesse or the result will be a jagged, flurry of shrillness.  To combat blocking distortion and Miller capacitance, RC filter networks are present throughout the amp to bleed off bass and treble, and to dump signal to ground between stages.

Input Stage

This stage has a ferrite bead following the input jack.  Ferrite has a large impedance and a low series resistance.  So the benefit of a 1M blocking resistor is met or exceeded, but the grid resistance is quite low (ferrite is usually only up to about 1 kohm).  There’s very little attenuation at the grid when this series resistance meets the internal capacitance of the 12AX7.

The high anode resistance, the cathode bypassing, and the high load (2.2M) offer a large signal gain at fairly high fidelity for a guitar amp.  As the first stage, its job is to turn that little guitar signal into something large enough for the other stages to shape.  It chops off frequencies below around 95 Hz (cathode bypass cap) and further reduces frequencies below 32 Hz (anode coupling).  The actual high frequency attenuation isn’t known to me, because the value of the ferrite isn’t specified.  At 1 K, it allows frequencies over 1 Megahertz to be amplified.  This doesn’t seem practical and I’m guessing Mesa had a ferrite bead made to some better specification.

Go With The Flow

The Channel 1 path has a tone stack immediately after the coupling cap to stage 1.  This is going to massively reduce the gain (insertion loss), as it presents a large load impedance to the signal.  Having the tone stack this far ahead in the amplifying chain will make its purpose less about a precise tone shaping tool and more about shaping the gain pot response.

The job of stage 2 (which is shared by the dirty channels) in regards to the Clean tone is to regain lost amplitude following the tone stack.  By comparison to the dirty signal path, the clean signal path has a large gain pot value and a higher frequency treble boost.  This does two things: it is brighter, allowing more air-y frequencies to come through, and there is less attenuation (dump) from the 2.2 M load resistor and 1 Meg pot (by comparison to the 510 k grid + 250 k pots in the dirty path).

The 475 k grid resistor adds with the wiper from the pot and has a maximum response of 19.7 kHz when the gain is maxed.  At normal levels of gain, the response will be lower.  The parallel 20 pf cap and 100 pf 12ax7 capacitance add, and in combination with the 475 k resistor, begins cutting frequencies over 2.8 kHz.

The anode resistor is 100 k and the cathode 1.8 k.  The 15 uf bypass cap in parallel to the 1 uf cathode bypass cap creates a more stepped attenuation of frequencies than the 1 uf cap on its own.  Some low-mid frequencies are less boosted and the sound is focused more in the upper mids, because the 15 uf cap is allowing those upper frequencies to pass much more easily.

(Aside:) Throughout the preamp, it seems that the target frequencies generally lie between 100 Hz and below 2 kHz.  Anything on either side is shunted, by small or large degrees, several times, and nearly every frequency is attenuated at least once to keep it manageable using multiple, parallel RC networks and non-cathode bypassed stages.  While  Clean Stage 2 has less attenuation in the highs than anywhere else in the amplifier, the next stage is its opposite.

Clean It Up

The last stage of the Clean preamp begins with a treble boost circuit on the large grid resistor of 3.3 M.  Frequencies above 1.2 kHz are allowed to pass more easily, but are then mostly dumped to ground by the 330 k load resistor.  Only 10% of the signal continues and is attenuated by the parallel capacitor and triode capacitance to 400 Hz low-pass filtering.  The lack of a cathode bypass cap keeps the frequencies attenuated at 400 Hz and reduces gain.  The coupling cap size of .047 uf cuts more inaudible bass than the standard .022 uf cap would.  It rolls off at 7 Hz.  This kind of filtering is used to counteract the wide response of stage 1 and the focus on the high frequency response of stage 2.

The RC filter combination after stage 3 is to add a permanent bright boost to the Clean channel and remove it for Pushed.  When clean is selected, the 220 k resistor sums with the 330 k resistor and the 1.5 M separately.  The 1.7 M resistance impedes all frequencies and bleeds to the Presence circuit.  The combined 550 k resistance has a 47 pf cap.   This prevents the 330 k resistor from reducing with the 1.5 M resistor in parallel.  The 550 k resistance also “sees” the 47 pf and 2 nf caps reduce from being in series, for a treble boost of 6.4 kHz.

The Presence circuit controls down to 318 Hz when fully counter-clockwise, which is quite low.  In addition, the frequencies blocked by the 1.72 M resistance are shunted down to 56 Hz, effectively blocking them, since the other signal will be stronger.  In combination with the 550 k resistance, this gives the Presence pot a range from 318 Hz to 6.4 kHz.  Smaller residual effects range from 56 Hz to 14 kHz.  This is wider than any other ‘Presence’ circuit in the amp

The last bit of filtering is a 250 pf cap that goes across the volume pot.  This throws high frequencies to ground as a low pass filter at a steady frequency of 636 Hz.  Since it is on the outside lugs, with lug 1 to ground, changes on the wiper from varying the volume affects the frequency to a lesser degree than attaching it to the wiper (as on the gain controls and the Channel 3 MV).  It “sees” the full resistance of the pot at all times.

Putting It Together

The flow is thus:

  • Stage 1 is a wide band amplifier.
  • Stage 2 has gain pot shaping EQ.
  • Stage 2 brightens the signal and brings the signal back up after insertion loss.
  • Stage 3 attenuates most of the signal.  Before the triode, it cuts above 400 Hz; after the triode, it cuts above 6.4 kHz and, to a lesser extent, 56 Hz.
  • Stage 3 has further attenuation: The Presence operates from 318 Hz to 6.4 kHz.  The Volume cuts above 636 Hz.

Due to the dirty channels sharing the first two stages of the Clean, the third stage makes the necessary adjustments to clean it up.  Having so much attenuation in stage 3 can present problems in achieving (subjectively) good tones.  Since the volume has permanent treble cut and the presence is working from a low frequency, getting a bright response will involve producing more treble in stage 2 to overcome the cutting and dumping.  From experimentation, I can recommend the following:

  • Treat the EQ as gain shaping tools and adjust them to add frequencies until before distortion.
  • Turn up the Gain control to get a more even and bright response.  Set overall volume with it.
  • Adjust the Presence control for brightness/treble response post-preamp.
  • Keep the volume low enough to prevent clipping the phase inverter/power amp.
  • The Volume setting has a lesser impact on the tone shaping compared to Gain; this is different from Channels 2 or 3.

I hope this helps.  The following installments will focus on Channels 2 and 3.  The power amp is set by the factory and can’t really be adjusted, so it may not be covered.  Thanks for reading and good luck.

  1. […] already covered the input stage and clean channel.  I’m not going to repeat the information about the input, except for applications to the […]


  2. […] Channels 2 and 3 Par… on Mesa Rectifier Design Concepts… […]


  3. […] Channels 2 and 3 Par… on Mesa Rectifier Design Concepts… […]


  4. […] Channels 2 and 3 Par… on Mesa Rectifier Design Concepts… […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s