Build Report:Bass Balls

[thumposaurus]

Just finished up with this from Mark Hammers site
http://ampage.org/hammer/files/BASEBALLS.ZIP
I used the original schematic and PCB/parts layout, and following the advice given in a recent thread here, and because I didn't have any trim pots I used regular pots for the trimmer controls.
With the fuzz off it is really subtle, but with the fuzz it really comes alive, still trying to dial in where the trimmers should be set to get the best sound, but that is the good part about makeing them into regular controls, they can be adjusted as needed.
It isn't boxed up at the moment still working on the box, but even with-out it it isn't noisey.
I also used a DPDT switch and wired it up as true by-bass instead of the SPST (I think) switch the original schematic calls for.
All in all an interesting project, and a nice change from the fuzzes/boosters I've been building.

[Mark Hammer]

The fuzz sound can be improved by replacing the 47k resistor that goes between the 100k resistor and ground with a pair of diodes.  This does a few things.  First, it will get you even more distortion which will always make a filter sound more interesting.  Second, since the "fuzz" signal is coming off the high gain stage in the envelope follower, and is determined by the sensitivity setting, the amount of sensitivity you set changes the effect volume, and there is no output volume control to offset that.  I don't know why E-H never thought to fix it, but they didn't and this mod improves things, yielding a fairly constant volume across fuzz vs nonfuzz settings, and across different sensitivity settings.  You can decide for yourself whether you like silicon or germanium diodes, whether you like symmetrical or nonsymmetrical clipping, hard or soft clipping, and whether you like a little bit more volume in the fuzz setting or not.

Replacing the 100R fixed resistor in the envelope follower with a 47R resistor and a 500R or 1k pot really expands the range of sounds too.  Stock, the BB has a very fast attack, fast enough that you tend not to hear much of it and mostly hear decay, so it *seems* to go "ow" rather than "wow".  Increasing the attack time even a little makes it go wah more, which is a different feel and sound.

Finally, I modded mine to be stereo by using a toggle to redirect the 2k7 resistor on the upper filter (the one with the 4700pf caps) to its own 1uf cap and 47k terminating resistor.  This does two things.  First, it lets you use the main output jack and have either a one-filter or two-filter sound (one filter sounds more like a Dr. Q).  Second, although you'd need to dedicate a 3PDT switch to bypassing it, you can use the alternate jack as its own filter out.  Finally, you can feed two distinct filter sounds to their own amps or mixer channels for a true stereo sound.  Normal mono functioning is restored by redirecting the upper filter back to the mixing junction.  There will be a slight pop, but its not the sort of thing you'd use midsong anyways, so stereo switch when bypassed and no one has to hear the pop.

As the Fuzzyballs document in the zipfile demonstrates, there is really no end to the mods you can do if you want to invest the time, but I found the three listed here provided a whole lot more performance control and sonic options for very little investment: two diodes, a SPDT toggle, a cap, 2 resistors, a pot and knob, and a jack.

[thumposaurus]

I changed the 47k resistor to a pair of 1n4148 diodes I had laying around and it is a little more fuzzy, I think I am going to stick a socket in there and play around with it more. Other than that I'm am kinda happy with how it is right now, I am more than likely going to build another one for my brother so I may try the other mods on that one and leave this one as it is.
Gotta go work on getting the box ready for finishing now.
Thanks for the ideas.

[loki]

hi, i've read your two interesting posts, and since i've the intention of building this pedal too i have a couple of questions to ask.

the first is very stupid (consider that i don't know much about electronics): the two diodes that substitute the resistor must be connected in series (one after the other) or in parallel?

the second is stupid too... did you use 10k linear pots instead of 10k trimmers?

and for Mark, i didn't quite undderstood the second mod you discussed, about replacing the 100ohms resistor with a 47ohm one and a pot. could you explain to me exactly how to connect the new resistor and the pot?

thank you

[thumposaurus]

For the trim pots, I just used what I had laying around one is linear, and one is log, I don't know if that makes a difference or not. I have a bunch of old CB radios I've been getting the pots and wire out of, I need to get a new heat gun so I can get the resistors and caps out of the boards.
For the diodes, I hooked them up in parallel, but I'm not sure if that really matters much either, does it Mark? Anyways I'm going to be putting a socket in place of the diodes so I can experement with some different combos.
finally the other question, I'm not so sure of myself but I think it goes you'd hook one end of the resistor to the board where it normally goes, the other end of the res. to one leg of the pot, and than a piece of wire to where the other end of the res. would go.  Correct me if I'm wrong.

[Mark Hammer]

The diodes are run in parallel, just like they would be in the MXR Distortion+, Boss DS-1 or the DOD 250 Overdrive.  Although these respective distortions have caps to ground in parallel with the diodes to "round off" the distorted sound, such a cap is not needed here because the filter sections do that for you.  Clipping of this sort *ahead* of the pedal tends to compress the dynamics of the signal, making for less sweep.  In the E-H Q-Tron and other more complex pedals like the Funk-A-Duck and Meatball, there will be patch-points for sending the signal out for additional processing (which can include fuzz but doesn't have to) *after* the envelope signal is derived.  In the BB, that is all done internally; the distortion signal is simply a by-product of what has to be done to the signal to condition it for envelope detection.

The 100R resistor and the 4.7uf capacitor set the time constants for the envelope follower; i.e., how fast it adapts to the incoming signal and how fast it drains off.  As the resistor just ahead of the cap gets larger it takes longer for the cap to charge up, and this is heard in the upward sweep taking a little longer.  So, if you replace it with a 47R fixed resistor in series with a 1k pot where only two adjacent lugs are used (as variable resistor), then you have a resistor that can vary from a minimum of 47R (very fast attack) to a maximum of 1047R, providing a wide range of attack times.  Note that this series resistance also reduces the current flowing through so lengthening the attack time also slightly reduces the apparent sensitivity.  You will need to increase the sensitivity control a bit when turning up the attack time, and turn the sensitivity down when moving for shorter attack times.  

The easiest way to implement this is to solder the 47R resistor to one of the lugs of the pot, and put some heat shrink over it and the lug to keep it from bending too much and fracturing.  You can run wires from the two pads where the 100R resistor used to go to the 47R resistor and the other lug of the pot.  

Another thing you can do is to install a 1k resistor in place of the 100R one, but solder it in so that it stands up a little from the board.  Now wrap the ends of two pretinned leads around the exposed lead of the resistor on the component side.  Those leads can now run to a toggle or slide switch that you can use to insert another resistor or combination to change between preset attack times.  For instance a 3-position on-off-of SPDT toggle would allow for you to switch in a 56R, no resistor or a 270R resistor in parallel for effective resistances of 53R, 212R or 1000R for a fast, medium and slow attack time.  These three settings may accomplish your needs more cheaply, reliably, and with less panel space than a pot+knob.

As for the filter pots, they can be either linear or log, but you may have the part of the sweep range of most interest to you crammed into only a few degrees of rotation if you use the "wrong" taper, so use your ears.

It is worth noting that the filter trimpots also act as sort of secondary sensitivity controls and not just purely as starting-frequency controls, so playing with them also allows you to change how much each formant "moves" during a sweep.  This is, I think, part of what makes the BB a more vocal-sounding envelope controlled filter than its close cousin the Doctor Q.  Its not just the fact of having two filters for a thicker sound, but the different dynamics of each filter section.

[loki]

thanks for your wonderful explanation mark, now it's all clear!

[thumposaurus]

Got the box finished, sort of still no lables, and the knobs are just crappy one's I had laying around, but it works and fit (just barely) in the box, I don't think I'd try to put one of these in this style box again I had to trim alot off of the pcb in order to squeeze it in.
Anyways, here are pics:


The box started out as the white one in the pics found here:
http://www.diystompboxes.com/sboxforum/viewtopic.php?t=19322&highlight=

[downweverything]

hey did you use the revised layout the one with bass balls in the etch resist or the ball buster one...arent there errors in the ball buster one (missing a jumper or something).  think im going to build one myself

[thumposaurus]

I used the one that says BASS BALLS on the PCB, worked perfectly the first time I plugged it in.