ROSS / DYNA compressor (AURORA Compressor) Attack pot change sound

[Pringles1]

Hi Guys!
I have finished a new Project. It's a Ross / Dyna compressor: Aurora compressor by Aion. This is the layout https://www.dropbox.com/s/d7m5f9rbe6d68f2/aion-aurora-ross-compressor-documentation.pdf?dl=1. I have two problems: The Attack pot and the Treble switch . I tested the Attack pot before mounting it and it worked properly and it's an alpha pt 150kC, but when I turn up-down this pot at PBC, the sound is the same when I play the guitar, does not vary. The level and Sustain pots work perfect. The other Little problema is the Treble switch. I mounted in the PCB a SPDT center off and CX2 360pF; C9 180pF and CX3 821pF capacitors. When I turn down the switch the sound is little dark/normal, in the middle position bright sound and in the upper position darkest sound, is it correct?? I dobble check the PCB components and I don´t found any problem. Could you help me. Thank you for your time!

1.What does it do, not do, and sound like? – The attack pot does not modify sound when turned up-down.
2.Name of the circuit = Aurora dyna comp by Aion
3.Source of the circuit (URL of schematic or project) = https://aionelectronics.com/project/aurora-ross-dyna-compressor/

4.Any modifications to the circuit? Dyna comp version as documented. Trimmer 250k,  Attack pot 150kC, Treble switch

5.Any parts substitutions? No.

6.Positive ground to negative ground conversion? No.

8. Voltage at the circuit board end of the red battery lead = 9.06
    Voltage at the circuit board end of the black battery lead = 0

See confirmed/remeasured values below

Q1
C = 7.28
B = 2.21
E = 1.78

Q2
C= 6.84
B= 2.78
E= 22.2

Q3
C = 8.89
B = 0
E = 0

Q4
C= 8.89
B= 0
E= 0

Q5
C= 9.06
B= 8.89
E= 8.31

IC1
P1= 0
P2= 4.41
P3= 4.41
P4= 0
P5= 0.71
P6= 2.74
P7= 9.06
P8= 0

D1
A=0
K=9.06

D2
A=0
K=0

D3
A=0
K=0

[Mark Hammer]

So-called Attack controls on compressors rarely, if ever, produce an audible change in sound, since the impact will always depend on how fast one is picking.  The control alters how quickly gain gets brought back up to maximum.  If gain-recovery is slow, any note picked shortly after the first one will not be as loud, because the gain is still on its way to coming back up.  f course, the benefit of a slow gain-recovery is that notes seem to sustain longer, in a manner analogous to slowly bringing up the mixer volume control as a song starts to fade out.

The control gets called "Attack", because when the gain is recovered quickly, the next note, and ones after that, are heard at normal volume when you pick them, such that the initial pick attack is more audible.

So, unless you have done something particularly wrong, your Attack control is behaving normally.  You're just expecting too much.

[thermionix]

Nice looking build.  And good to see a Comp thread that isn't about a fake 3080.  Inspired by Dr. Hammer above, I added an "Attack" (Release) 3-way switch to my Dynacomp.  I do notice a slight boost in treble and volume when I switch to one of the faster-than-stock settings.  I guess that's just the nature of a signal that is, on average, less attenuated.  But that's with a switch, where the change is immediate, I might not notice it when turning a pot.  But if you can sweep your pot from one extreme to the other quickly, or have someone do that for you while you play, you might be able to detect that subtle difference.

[Mark Hammer]

That's why I thought EHX took the sensible approach by including a simple 3-position switch in their Soul Preacher.  My sense is that most folks who own a compressor with a continuous rotary control will generally set it to either the 7:00 or 5:00 positions, simply because they have a hard time hearing any difference between those two extremes.

[Pringles1]

Thank you for yours comments, time and help. I have tested again the pedal with fast changes of the Attack pot between 7 and 5 and I feel a beautiful discrete change in the definition of the notes, subtle, but it exists. I'm going to change the title of the topic to guide other colleagues.

[Mark Hammer]

My pleasure.  When more complex rackmount compressors include attack and release controls, the assumption underneath that is that they will be necessary for processing, and adjusting to, sound sources which are VERY different in their envelope characteristics (e.g., flute vs snare drum, crash cymbal vs. piano).  That does not mean that a guitar cannot be different enough at times to justify an Attack (gain recovery) control.  But I think one has to understand what the control is doing to be able to know what potential impact such adjustments may have.  The expectations have to be realistic.  Unfortunately, most of our experience with control knobs on pedals has to do with VERY obvious changes: speed, volume, distortion, tone, etc.  Some controls have an effect, but one so subtle and situationally-dependent, that players will often ask "Is this thing on?  Is it broken". 

[rankot]

Most noticeable attack control is when playing bass with different techniques - slap, pick or fingered. Not so obvious with guitar.

[rosssurf]

The "attack" pot is really only a recovery time adjustment. Set up your rig so there is a little background noise and adjust the "attack" pot fully either way. Now strike a hard and quickly mute the note. Then listen how long it takes for the hiss or background noise to rise back up after being compressed.  Now adjust the "attack" pot the opposite way and notice how the rise time is either slower or faster. I think attack pot is a misnomer.

[PRR]

> I think attack pot is a misnomer.

If anybody reads the linked documentation, they will find:

"Attack is technically a "Release" control, but it's called Attack in both the Boss CS-2 and Keeley Compressor so the name stuck. It sets the amount of time after the input signal falls below the threshold before the compressor "resets" and is ready to compress again."

[rankot]

Maybe off topic, but I have a build of Dyna/Ross compressor, where I tried to add Release mod with a B220k pot in series with 10k resistor, instead of stock 150k resistor. Most people suggest some values from 10k-150k with a switch at that spot, but I noticed that any value below 82k produces distortion on my unit. There are some mods (or schematics of similar pedals) using 47k+C250k pot in that place.

I can leave it with 82k as a lower boundary (with 100k pot in series, for example), however, I'm curious if something's wrong with my build, or not when it distorts on lower values?

Also, what is recommended pot for Ratio (or Sustain on some schematics)? It is always 500k, but shall it be C500k or B500k, as I seen on Emma TransMORGrifier, for example?

[Rob Strand]

Maybe off topic, but I have a build of Dyna/Ross compressor, where I tried to add Release mod with a B220k pot in series with 10k resistor, instead of stock 150k resistor. Most people suggest some values from 10k-150k with a switch at that spot, but I noticed that any value below 82k produces distortion on my unit. There are some mods (or schematics of similar pedals) using 47k+C250k pot in that place.

Because it's really a *release pot*.   Reducing the release time-constant ends up producing distortion due to ripple on the control signal.   Totally normal and unavoidable.     The release time constant needs to be greater than some minimum  in order to keep the distortion low.

I can leave it with 82k as a lower boundary (with 100k pot in series, for example), however, I'm curious if something's wrong with my build, or not when it distorts on lower values?

IMHO, putting it right near the boarder is probably not the best solution.   It doesn't fix the attack anyway.

Also, what is recommended pot for Ratio (or Sustain on some schematics)? It is always 500k, but shall it be C500k or B500k, as I seen on Emma TransMORGrifier, for example?

It's highly likely to be C-taper.  What happens on the internet is many schematics are drawn without tapers then later on someone adds it but it might not be what was originally used.   Often that turns out to be linear because they are easy to get.  The original Boss and Ibanez schematics are good references for the correct pot tapers.

Here's the Boss CS-2 compressor.  Despite the marking the ATTACK pot is also actually a Release pot.
https://www.hobby-hour.com/electronics/s/schematics/cs2-compression-sustainer-schematic.gif

As far as increasing the *actual* attack time.   It's not straight forward and not really easy to make adjustable without affecting how the unit works - although some schemes might be acceptable.

Using the CS-2 diagram for part references, some schemes:
- I'd say the easiest way would be to use high-gain transistors for Q8, Q9.   Gain dependence makes it hard to be precise you can
  just say it's faster.
- Reducing C1 reduces *both* attack and release times. So halving C will half attack but then you will need to double the
  the release resistance VR2+R1 to compensate.    You can only go so far here before the sustain pot starts loading the circuit through via the base current of the transistor connecting to C1.   A high gain transistor will help.
- Reduce R2, R4, and/or C2, C3.    This part of for circuit works in a very complex manner.   You might be able to play with the values but you could also stuff something up as well. [Making the caps bigger decreases the attack time.  The complexity is that makes the time constants around that part of the circuit longer.  The top and bottom outputs of the transistor phase inverter circuit aren't symmetrical perspective.  The Boss CS3 uses an opamp and adds series resistors.]

About 15yrs ago I set-up a simulation of the dynacomp to analyse the time constants and effect of parts.  Unfortunately it's on another computer (and I'm not really motivated to enter a large circuit into LTspice to repeat the set-up - sorry  ).

[rankot]

Thank you, Rob! Very detailed explanation. I put A500k for Ratio (don't have C500k at the moment), but reversed pins, so it has the same behavior, although min/max ratio are reversed. Anyway, to me it seems more logical - when turned CCW, compression seems to be lowest and volume is bigger, and when I turn it CW, compression is bigger and volume drops. Right? Or not? 

Because it's really a *release pot*.   Reducing the release time-constant ends up producing distortion due to ripple on the control signal.   Totally normal and unavoidable.     The release time constant needs to be greater than some minimum  in order to keep the distortion low.

OK, I understand that, but to me seems that there is always some rattling when compression Ratio pot is low and regardless Release. Maybe I shall try to change R2 for the start.

I used 2N5088 transistors with hfe of around 400 for Q8/9.

Regarding ltSpice model, I happen to have made one:

Code Select Expand



Version 4
SHEET 1 3180 888
WIRE 0 -160 -64 -160
WIRE 224 -160 0 -160
WIRE 592 -160 224 -160
WIRE 928 -160 592 -160
WIRE 1424 -160 928 -160
WIRE 1520 -160 1424 -160
WIRE 1616 -160 1520 -160
WIRE 2048 -160 1616 -160
WIRE 928 -144 928 -160
WIRE 0 -128 0 -160
WIRE 224 -128 224 -160
WIRE 2048 -128 2048 -160
WIRE 1616 -112 1616 -160
WIRE 1424 -80 1424 -160
WIRE 1520 -80 1520 -160
WIRE 0 -32 0 -48
WIRE 80 -32 0 -32
WIRE 112 -32 80 -32
WIRE 224 -32 224 -48
WIRE 848 -32 832 -32
WIRE 928 -32 928 -64
WIRE 928 -32 912 -32
WIRE 2048 -32 2048 -48
WIRE 2144 -32 2048 -32
WIRE 2176 -32 2144 -32
WIRE 0 -16 0 -32
WIRE 80 -16 80 -32
WIRE 928 -16 928 -32
WIRE 2336 0 2304 0
WIRE 2448 0 2400 0
WIRE 2496 0 2448 0
WIRE 1616 16 1616 -32
WIRE 1680 16 1616 16
WIRE 1776 16 1744 16
WIRE 1840 16 1776 16
WIRE 1984 16 1840 16
WIRE 1840 32 1840 16
WIRE 592 48 592 -160
WIRE 1168 48 1040 48
WIRE 1776 48 1776 16
WIRE 1248 64 1248 16
WIRE 1616 64 1616 16
WIRE 0 80 0 64
WIRE 80 80 80 48
WIRE 1040 80 1040 48
WIRE 1040 80 800 80
WIRE 2304 80 2304 0
WIRE 2336 80 2304 80
WIRE 2448 80 2448 0
WIRE 2448 80 2416 80
WIRE 464 96 368 96
WIRE 528 96 464 96
WIRE 1312 96 1312 64
WIRE 2208 96 2208 48
WIRE 1424 112 1424 0
WIRE 1424 112 1392 112
WIRE 1520 112 1520 -16
WIRE 1520 112 1424 112
WIRE 1552 112 1520 112
WIRE 1936 112 1840 112
WIRE 2048 112 2048 64
WIRE 2048 112 2016 112
WIRE 2448 112 2448 80
WIRE 1776 128 1776 112
WIRE 1840 128 1840 112
WIRE 2048 144 2048 112
WIRE 80 160 0 160
WIRE 192 160 144 160
WIRE 240 160 192 160
WIRE 368 160 368 96
WIRE 368 160 320 160
WIRE 464 160 464 96
WIRE 1312 160 1312 128
WIRE 592 176 592 144
WIRE 688 176 592 176
WIRE 800 176 800 80
WIRE 800 176 752 176
WIRE 832 176 800 176
WIRE 928 176 928 64
WIRE 928 176 912 176
WIRE 944 176 928 176
WIRE 1040 176 1024 176
WIRE 1152 176 1040 176
WIRE 1168 176 1152 176
WIRE 2304 176 2304 80
WIRE 592 192 592 176
WIRE 800 192 800 176
WIRE 1040 192 1040 176
WIRE 1248 192 1248 160
WIRE 1616 192 1616 160
WIRE 1616 192 1520 192
WIRE 1680 192 1616 192
WIRE 1776 192 1744 192
WIRE 1840 192 1776 192
WIRE 1984 192 1840 192
WIRE 0 208 0 160
WIRE 192 208 192 160
WIRE 368 208 368 160
WIRE 1360 208 1360 128
WIRE 1616 208 1616 192
WIRE 1840 208 1840 192
WIRE 1776 224 1776 192
WIRE 2448 224 2448 192
WIRE 1152 240 1152 176
WIRE 1280 240 1152 240
WIRE 2144 240 2144 -32
WIRE 2144 240 2048 240
WIRE 1152 272 1152 240
WIRE 1280 288 1280 240
WIRE 464 304 464 240
WIRE 592 304 592 272
WIRE 800 304 800 272
WIRE 1040 304 1040 272
WIRE 1616 304 1616 288
WIRE 1776 304 1776 288
WIRE 1840 304 1840 288
WIRE 0 320 0 288
WIRE 192 320 192 288
WIRE 192 320 0 320
WIRE 368 320 368 272
WIRE 368 320 192 320
WIRE 0 336 0 320
WIRE 1360 352 1360 288
WIRE 2208 352 2208 176
WIRE 2208 352 1360 352
WIRE 1056 384 1024 384
WIRE 1152 384 1152 352
WIRE 1152 384 1120 384
WIRE 1280 384 1280 352
WIRE 1280 384 1152 384
WIRE 1520 384 1520 192
WIRE 2304 384 2304 240
WIRE 2304 384 1520 384
FLAG 0 336 0
FLAG 224 -32 0
FLAG 2496 0 OUT
IOPIN 2496 0 Out
FLAG 464 304 Vref
FLAG 592 304 0
FLAG 800 304 0
FLAG 112 -32 Vref
FLAG 0 80 0
FLAG 80 80 0
FLAG 1040 304 0
FLAG -64 -160 Vcc
FLAG 1248 192 Vcc
FLAG 1248 16 0
FLAG 1616 304 0
FLAG 1776 128 0
FLAG 1840 128 0
FLAG 1776 304 0
FLAG 1840 304 0
FLAG 2144 304 0
FLAG 1024 384 0
FLAG 832 -32 0
FLAG 2176 -32 Bin
FLAG 1312 160 Bin
FLAG 2208 48 Bout
FLAG 1312 64 Bout
FLAG 2448 224 0
SYMBOL voltage 224 -144 R0
WINDOW 123 0 0 Left 2
WINDOW 39 0 0 Left 2
SYMATTR InstName V1
SYMATTR Value 9
SYMBOL voltage 0 192 M0
WINDOW 123 24 124 Left 2
WINDOW 39 24 152 Left 2
SYMATTR Value2 AC ac 1.
SYMATTR SpiceLine Rser=11k
SYMATTR InstName V2
SYMATTR Value SINE(0 .1 440)
SYMBOL cap 144 144 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C2
SYMATTR Value 22n
SYMBOL res 480 256 R180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R6
SYMATTR Value 510k
SYMBOL cap 384 272 R180
WINDOW 0 24 56 Left 2
WINDOW 3 24 8 Left 2
SYMATTR InstName C6
SYMATTR Value 150p
SYMBOL res 208 304 R180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R1
SYMATTR Value 2Meg2
SYMBOL res 224 176 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 0 56 VBottom 2
SYMATTR InstName R2
SYMATTR Value 10k
SYMBOL npn 528 48 R0
SYMATTR InstName Q1
SYMATTR Value 2N5088
SYMBOL res 576 176 R0
SYMATTR InstName R4
SYMATTR Value 10k
SYMBOL cap 752 160 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C1
SYMATTR Value 10µ
SYMBOL res 784 176 R0
SYMATTR InstName R7
SYMATTR Value 1Meg
SYMBOL res 16 -32 R180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R5
SYMATTR Value 56k
SYMBOL res 16 80 R180
WINDOW 0 36 76 Left 2
WINDOW 3 36 40 Left 2
SYMATTR InstName R8
SYMATTR Value 27k
SYMBOL cap 96 48 R180
WINDOW 0 24 56 Left 2
WINDOW 3 24 8 Left 2
SYMATTR InstName C3
SYMATTR Value 10µ
SYMBOL AutoGenerated\\LM13700_NS 1280 112 M180
WINDOW 0 5 -86 Bottom 2
WINDOW 3 -49 128 Top 2
SYMATTR InstName U1
SYMBOL res 1024 176 R0
SYMATTR InstName R9
SYMATTR Value 1Meg
SYMBOL res 1440 -96 M0
SYMATTR InstName R33
SYMATTR Value 150k
SYMBOL cap 1536 -16 R180
WINDOW 0 24 56 Left 2
WINDOW 3 24 8 Left 2
SYMATTR InstName C4
SYMATTR Value 1n
SYMBOL npn 1552 64 R0
SYMATTR InstName Q2
SYMATTR Value 2N5088
SYMBOL res 1600 -128 R0
SYMATTR InstName R11
SYMATTR Value 10k
SYMBOL res 1600 192 R0
SYMATTR InstName R12
SYMATTR Value 10k
SYMBOL cap 1744 0 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C5
SYMATTR Value 10n
SYMBOL cap 1744 176 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C7
SYMATTR Value 10n
SYMBOL schottky 1792 112 R180
WINDOW 0 24 64 Left 2
WINDOW 3 24 0 Left 2
SYMATTR InstName D1
SYMATTR Value BAT85
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res 1824 16 R0
SYMATTR InstName R13
SYMATTR Value 1Meg
SYMBOL npn 1984 -32 R0
SYMATTR InstName Q3
SYMATTR Value 2N5088
SYMBOL res 2032 -144 R0
SYMATTR InstName R14
SYMATTR Value 10k
SYMBOL schottky 1792 288 R180
WINDOW 0 24 64 Left 2
WINDOW 3 24 0 Left 2
SYMATTR InstName D2
SYMATTR Value BAT85
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res 1824 192 R0
SYMATTR InstName R15
SYMATTR Value 1Meg
SYMBOL npn 1984 240 M180
SYMATTR InstName Q4
SYMATTR Value 2N5088
SYMBOL cap 2160 304 R180
WINDOW 0 24 56 Left 2
WINDOW 3 24 8 Left 2
SYMATTR InstName C8
SYMATTR Value 10µ
SYMBOL res 1344 192 R0
SYMATTR InstName R16
SYMATTR Value 27k
SYMBOL res 2192 80 R0
SYMATTR InstName Sustain
SYMATTR Value 1
SYMBOL res 928 160 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R17
SYMATTR Value 1k
SYMBOL res 1040 160 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName R18
SYMATTR Value 1k
SYMBOL res 912 -32 R0
SYMATTR InstName R19
SYMATTR Value 220k
SYMBOL res 1136 256 R0
SYMATTR InstName R10
SYMATTR Value 15k
SYMBOL cap 1296 352 R180
WINDOW 0 24 56 Left 2
WINDOW 3 24 8 Left 2
SYMATTR InstName C10
SYMATTR Value 10n
SYMBOL cap 1056 400 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName C11
SYMATTR Value 1µ
SYMBOL res 2032 96 R90
WINDOW 0 0 56 VBottom 2
WINDOW 3 32 56 VTop 2
SYMATTR InstName Attack
SYMATTR Value 1
SYMBOL res 912 -160 R0
SYMATTR InstName R23
SYMATTR Value 220k
SYMBOL cap 848 -16 R270
WINDOW 0 32 32 VTop 2
WINDOW 3 0 32 VBottom 2
SYMATTR InstName C14
SYMATTR Value 1µ
SYMBOL cap 2288 176 R0
SYMATTR InstName C17
SYMATTR Value 100n
SYMBOL res 2320 96 R270
WINDOW 0 32 56 VTop 2
WINDOW 3 0 56 VBottom 2
SYMATTR InstName R28
SYMATTR Value 10k
SYMBOL cap 2400 -16 R90
WINDOW 0 0 32 VBottom 2
WINDOW 3 32 32 VTop 2
SYMATTR InstName C18
SYMATTR Value 1n
SYMBOL res 2432 96 R0
SYMATTR InstName R29
SYMATTR Value 50k
TEXT 296 -80 Left 2 !.tran 0 .04 .02
TEXT 296 -112 Left 2 !;.ac oct 16 10 50k
TEXT 1384 304 Left 2 ;68k?
TEXT 1688 -80 Left 2 ;orig. 10k



Also tried to add switch for sparkle mode (change C5 from you schematic to 150p, 480p or 1n) as proposed on another thread, but I can't hear the difference. Maybe my strings are too old? But it is 10n in this schematic you posted, and 1n everywhere else?

[Rob Strand]

Anyway, to me it seems more logical - when turned CCW, compression seems to be lowest and volume is bigger, and when I turn it CW, compression is bigger and volume drops. Right? Or not? 

Haha, I think you get used to whatever way you wire it up.    The threshold pots on noise gates never make sense to me one way or the other until I turn the knob.

As a side issue, the weird thing about the Dynacomp is the compression pot is really a *input* threshold control.



It's a hard limit like curve #5 and the compression control moves the knee downward (to the left).   When the knee is near the perceived signal level it acts like a compression ratio because you are near the curved part.      Don't worry if you don't get it.

OK, I understand that, but to me seems that there is always some rattling when compression Ratio pot is low and regardless Release. Maybe I shall try to change R2 for the start.

R2 and R4 should match in value.   I'd expect very subtle behaviour with these parts.  Maybe you can't even hear so easy.   (Fine details like this is where the simulation gives a clearer indication about what is going on.)

I used 2N5088 transistors with hfe of around 400 for Q8/9.

Yeah, you won't get much more gain than those.  So perhaps reducing C1 might help (provided you up the VR2 etc).

Also tried to add switch for sparkle mode (change C5 from you schematic to 150p, 480p or 1n) as proposed on another thread, but I can't hear the difference. Maybe my strings are too old? But it is 10n in this schematic you posted, and 1n everywhere else?

To me some of the sparkle mods don't look like they work so well.   I think Mark Hammer has a few posts where here gives a circuit which looked OK.  At the end of the day it's just a treble boost tone control to restore the *perceived* balance.    IIRC there was a later Boss CS-2 with a tone knob I'm thinking of the CS-1 with a treble switch!.   

https://www.diystompboxes.com/smfforum/index.php?topic=118742.0

I found this Boss CS3 circuit, you can see it's just a treble control.



As for the 1n vs 10n.   The 1n is normal for the LM3080 etc circuits but the 10n is correct for the Boss.  The difference is the LM3080 circuits have a 150k resistor whereas the CS2 has a different VCA and the resistor is 15k so you need 10 times the cap size to get the same result.     The 10n + 15k at the input boosts the treble into the VCA and the 1n + 150k (or 10n + 15k) cuts it at the output.  It's a way to reduce noise.   The input boost and output cut cancel out to give a flat response.    If you mess with this matching the net result can be a treble boost which acts as a sparkle mod.

[Rob Strand]

Also tried to add switch for sparkle mode (change C5 from you schematic to 150p, 480p or 1n) as proposed on another thread, but I can't hear the difference. Maybe my strings are too old? But it is 10n in this schematic you posted, and 1n everywhere else?

For this, and continuing on from the my last paragraph.    When you reduce C5, the treble boost starts to rise at about 1kHz.  The amount if rises and at what frequency it levels off is determined by C5.



1n = level, no boost
480p    flattens off at about 2.2kHz, boost about 6.8dB
150pF  flattens off at about 7kHz, boost about 17dB.

The 1kHz start point cannot be shifted with C5.    The way to shift that point is with the 10nF cap on the input side (C8 on Boss CS2).

Even though the start point is at 1kHz, I would expect to hear some brightening with either 480pF or 150pF.  If you doubled C8 then that should definitely produce a noticeable change but IMHO the way it's intended to be used it's probably too low.

On your sim, try analyzing from 0 to 150ms but put a delay of say 10ms on the sine source.    To start maybe set the input frequency to 10kHz so you can see a clean envelope.     When have an attack of say 2mS and a 1kHz signal the attack isn't distinct.

[rankot]

To me some of the sparkle mods don't look like they work so well.   I think Mark Hammer has a few posts where here gives a circuit which looked OK.  At the end of the day it's just a treble boost tone control to restore the *perceived* balance.    IIRC there was a later Boss CS-2 with a tone knob I'm thinking of the CS-1 with a treble switch!.   

https://www.diystompboxes.com/smfforum/index.php?topic=118742.0

I found this Boss CS3 circuit, you can see it's just a treble control.

I've found two ways of implementing "sparkle":

• Replace R8 (15k on Boss CS2 sch.) with a A100k pot in series with 7k5 resistor (Walrus Audio Deep Six Compressor) - I will try this control, too, and
• Replace C5 with a 180p in parallel with a SPDT On/Off/On switch, connected with 360p and 820p on poles, as proposed by Aion (see schematic). I tried this one.

As for the 1n vs 10n.   The 1n is normal for the LM3080 etc circuits but the 10n is correct for the Boss.  The difference is the LM3080 circuits have a 150k resistor whereas the CS2 has a different VCA and the resistor is 15k so you need 10 times the cap size to get the same result.     The 10n + 15k at the input boosts the treble into the VCA and the 1n + 150k (or 10n + 15k) cuts it at the output.  It's a way to reduce noise.   The input boost and output cut cancel out to give a flat response.    If you mess with this matching the net result can be a treble boost which acts as a sparkle mod.

You're right, I simply didn't notice that resistor is reduced 10 times. 

[Rob Strand]

Replace R8 (15k on Boss CS2 sch.) with a A100k pot in series with 7k5 resistor (Walrus Audio Deep Six Compressor) - I will try this control, too, and
    Replace C5 with a 180p in parallel with a SPDT On/Off/On switch, connected with 360p and 820p on poles, as proposed by Aion (see schematic). I tried this one.

Both will work.   Because the first one changes the resistor it drops the level of the lows going to the VGA and in theory you have to up the compression pot to compensate.      The treble boost (actually "bass" cut) tends  to act around the 1kHz mark and below.   It will sound different.

The Aion is more along the lines of the intent of the control.

Another way is to bridge R3 in the CS2 diagram with a cap.  If the pot is 50k and R3 is 10k you wont get much treble boost but cap is can be chosen to be any value so to have complete freedom in choosing the frequency.   The other methods kind of tack on to the existing pre-emphasis/de-emphasis circuits so the frequencies get tied to the existing frequencies,

The CS3 just says screw it an adds a tone pot, again you can put the frequencies wherever you want.   The Boss LM-2B and LMB-3 also use separate tone controls - they tend to act on the higher frequencies.

Another way is to mix a small amount of dry signal at the output (you have make sure the phase is correct).   Subtle but in a good way.