Tuesday, March 23, 2021

Stompboxes: Revisiting the Stupid FET Trick

I was ordering some components for an upcoming stompbox project, so I decided to get enough extra to do some updates to my 2008 project, the Stupid FET Trick:
  • Add a 9V DC jack for adapter power.
  • Add a power light adjacent to the input jack to remind the user to unplug the input cord.
  • Add an "active" light next to the footswitch.
  • Add a graphic label.
I ended up using Valvewizard's MOSFET source-follower variant of the Millenium Bypass circuit for the "active" light.

Since I was adding the 9V jack, I also added a reverse-polarity diode, though it wasn't necessary to protect the effect circuit: The 2N5457 JFETs that make up the circuit have interchangeable source and drain, so reversing the polarity just turns every common-source gain stage into a common-drain buffer stage and vice-versa.  It won't actually work with power polarity reversed, but the JFETs won't be damaged.  The power LED can be damaged by reverse polarity so there is some purpose there.

Here's the schematic as it stands.  If you didn't read the original article, much of it was based on taking a typical tube amp approach, with component values adjusted for the lower voltages of a JFET circuit. A couple things were just outright mistakes, but I left them.

The graphic label (for the top surface only) was inkjet-printed onto vinyl label material, and features stylized versions of the original pictograms (☠💖🎶) for the knobs:

Aside from the mostly cosmetic and convenience changes, it's still the same guitar effect it was when I built it more than a decade ago:
  • It's not quite like any other distortion device I have.
  • It's still really noisy (mostly due to large-value resistors in the wrong places).
  • There's still DC current through the "Skull" and "Quaver" potentiometers, so they "scratch" when adjusted.
It also turns out that the 9V jack is not too useful: Since it's all Class-A gain stages, the power supply noise from a switching-mode "wall wart" supply that makes its way into the audio signal is just too much for even practice use.

I'll probably update this post with an audio recording soon, since I had never bothered to include one with the original design.

Here's what the simulated output is, with 1 kHz sine input and increasing "Skull" values:
Here's the effect of the "Heart" knob on frequency response:



Thursday, August 29, 2019

A Cosine-derived Well Temperament

Ever since I read about Brad Lehman's "Bach" tuning (some time in 2006), I have been interested in experimenting with unequal "well" temperaments for keyboard instruments.  But, it was only recently that I got a hardware synthesizer with reasonable capabilities for custom tunings.

While there seems to be insufficient proof that Lehman's tuning truly recreates anything used by J. S. Bach, it does adhere to the general goal that's implied in other historical well temperaments I've since read about:  A gradual change in "coloration" of the tuning as you go around the circle of fifths, with the most "just" major third (i.e. flat of equal-tempered) occurring in the key of C and the most "Pyhagorean" major third (i.e. sharp of ET) at the other side of the circle at F♯.  The width of the fifths in these temperaments appear to have been derived by trial and error to close the circle while achieving the desired thirds.  Precision was of course limited by the mechanical precision of the piano and the capabilities of the human tuner.

I decided to develop a well temperament mathematically, using a cosine function to adjust the width of the fifths gradually.  I used a spreadsheet to derive the thirds, and below are links (to Google Sheets) and a copy of the output, with two variations. The amount of stretch was chosen so that the widest fifths would be just (i.e. 3:2, about 701.955¢). For each of version, the flattest third (C-E) ends up about 7¢ sharp of just, and the sharpest third is about 20¢ sharp of just (but not quite Pythagorean).

Centered on C, with just fifths at C♯-G♯ and G♯-D♯:

-->
NoteIndexCosineFifthDegreeM3∆ Just
C00.7071698.5688400.000000C-E:393.2276786.913964
G10.9659698.044999698.568840G-B:394.1349977.821284
D20.9659698.044999196.613839D-F#:396.61383910.300125
A30.7071698.568840894.658838A-C#:400.00000013.686286
E40.2588699.476159393.227678E-G#:403.38616117.072447
B5-0.2588700.5238411092.703837B-D#:405.86500319.551289
F#6-0.7071701.431160593.227678F#-A#:406.77232220.458608
C#7-0.9659701.95500194.658838C#-F:405.86500319.551289
G#8-0.9659701.955001796.613839G#-C:403.38616117.072447
D#9-0.7071701.431160298.568840D#-G:400.00000013.686286
A#10-0.2588700.5238411000.000000A#-D:396.61383910.300125
F110.2588699.476159500.523841F-A:394.1349977.821284

Centered between C and G
, with a just fifth at G♯-D♯:

-->
NoteIndexCosineFifthDegreeM3∆ Just
C00.5000699.0225000.000000C-E:393.6813387.367624
G10.8660698.306920699.022500G-B:393.6813387.367624
D21.0000698.044999197.329419D-F#:395.3744189.060704
A30.8660698.306920895.374418A-C#:398.30692011.993206
E40.5000699.022500393.681338E-G#:401.69308015.379367
B50.0000700.0000001092.703837B-D#:404.62558218.311868
F#6-0.5000700.977500592.703837F#-A#:406.31866220.004948
C#7-0.8660701.69308093.681338C#-F:406.31866220.004948
G#8-1.0000701.955001795.374418G#-C:404.62558218.311868
D#9-0.8660701.693080297.329419D#-G:401.69308015.379367
A#10-0.5000700.977500999.022500A#-D:398.30692011.993206
F110.0000700.000000500.000000F-A:395.3744189.060704

Here are the tunings in Scala format if you would like to use them:

! Cosine C Well.scl
!
C-Centered Cosine Well
 12
!
 94.65884
 196.61384
 298.56884
 393.22768
 500.52384
 593.22768
 698.56884
 796.61384
 894.65884
 1000.00000
 1092.70384
 2/1

! Cosine CG Well.scl
!
C/G-Centered Cosine Well
 12
!
 93.68134
 197.32942
 297.32942
 393.68134
 500.00000
 592.70384
 699.02250
 795.37442
 895.37442
 999.02250
 1092.70384
 2/1

Incidentally, these are rather nice-looking when viewed through Scala's temperament radar:

C-Centered Cosine Well Temperament - Radar

C/G-Centered Cosine Well Temperament - Radar






Monday, September 3, 2018

Stopping the Squeaks on a Burghardt Scissor-Jack Piano Bench

With a good piano, decent microphones, a [Really] nice preamp, and a not-bad A/D converter, you'd hope that playing ability would be the only thing to prevent a good piano recording in a home studio…  Then you play back to find that the squeaking piano bench is louder than the all but the loudest passages. Judging from the numerous discussion threads on piano-related online forums, I presume it's a common problem.  I discovered a relatively simple solution that might be helpful to others.

My bench is an adjustable Burghardt with a "scissor jack" style lifting mechanism, but my squeak solution should be applicable to some other types of benches.



I searched online for some solutions:

  • "Frequently tightening bolted joints." – Considering the bench came with its own wrench, the manufacturer must have been expecting this approach, but eventually it stops working.  It also only works for squeaks between the legs and apron board.
  • "Ensuring the bench is upright, on a level surface before tightening leg bolts." – I loosened the leg bolts and tried, but there was no difference.
  • "Putting Vaseline (or other lubricants) in wood-to-wood joints between apron and legs." – It seemed like this would be bad for the wood finish.
  • "WD40 at various locations." – Also bad for the wood finish, and not a good long term solution.

As I was reading these aloud, my wife suggested inserting a piece of fabric into the joints, and offered a scrap of nylon. That's when I remembered I had a partial roll of Rockler "Nylo-Tape" from a furniture project.  Nylo-tape is adhesive-backed nylon tape. I had 3/4" wide × 10 mil thick on hand, which worked just right.

First, I removed the legs (keeping track of which goes where!) and applied short pieces of Nylo-tape to the ends of each apron board, where they bear against the legs.

Then I applied long strips of Nylo-tape between the the lifting mechanism and the two wooden frames.  I just cut them a little longer than needed, marked the backing with the position of each hole, and then used a standard handheld office paper punch to make holes for the screws in each strip of tape.  (I actually removed each frame from the mechanism in turn, rather than all at once, to make it easier to put back in the right orientation.)



I did not apply any Nylo-tape between the corner brackets and the apron, partly because the squeaking seemed to stop after the first round of fixes, and partly because I ran out of Nylo-tape.  If new squeaks develop there, I may have to apply some later: