There's very little written about Jackson Spectra basses that isn't simply repeated from the ad copy, and since I've committed to make one my main instrument, here's what I've learnt so far from playing it, analyzing its electronics, and fixing their deficiencies.
I have the mid-range 5-string model (X Series) in electric blue (which is darker than the apparent baby-blue of the online pictures). I got it in 2021 for about $700 CAD plus taxes (they retail for $1000 CAD now). It's a 2019 build, and the serial number starts with "ISJ", which I assume means "Indonesia, Samick, Jackson", given the similarity to Ibanez basses also built there. I suspect it's a complete sub-contract for FMIC, as writing to Jackson requesting technical info got me a reply from Fender, who had no service manuals or other information at all.
The specs are as-advertised, with a few missing bits:
The build quality is a mixed bag, and varies from individual bass to bass, given the few I've tried in-store. Generally, the "furniture" (neck, body, frets, finish) is very good, and the hardware (tuners, bridge, pickups, electronics) is very cheap.
I bought the Jackson Spectra because it seemed like a versatile instrument with some extra features usually only found on higher-end models: a 35" scale to help the B string, compound radius to enable a lower action, and a graphite-reinforced neck. I also bet on the neck-through construction requiring a higher quality control (else the whole instrument is junk), and while I don't know if that's generally true, in my case it worked out: the neck and fretwork are as perfect as imaginable. See the Setup section at the end to see how.
The stock electronics are mostly well built: good solder joints, shielded wire as much as possible, everything well-grounded. Except for the bridge ground, which has a 12 Ohms resistance, which is odd but still works. The pickups are grounded internally, but unevenly: I measure anywhere from 60 to 200 Ohms across any given pair of poles, or to the bare shield wire. The control cavity has a thin, poor coat of carbon shielding paint (300 Ohms/sq) which does not make contact with the metallized cavity cover.
The pickups are conventional neck and bridge humbuckers, with different pole spacing to follow the strings, and a bit more windings on the bridge pickup. Coils wire pairs are green/red and white/black, with red/white as the coil tap, and black usually tied to the bare shield wire, all run through a shielded 4-conductor cable. Each pickup has its coils in reverse order of the other, so when you short the coil taps to ground, the remaining active coils are the outermost, forming a humbucking J-bass configuration. The Spectra sounds very clear in this mode.
Each pickup is wired as a series humbucker and has its coil tap go to one half of a DPDT switch to short out one coil to convert the pickup to single-coil, which sounds much better (not muddy), but isn't humbucking except when blended 50/50 with the other. The switched pickups then go to a blend pot (set of MN250K) with center detent.
The blended signal then goes into the pre-amp, which is completely potted with only the gain trimpot showing. There are no markings. Bass/Mid/Treble EQ pots are B100K with center detent. The EQ is flat at noon (no difference with passive tone) and does conventional bass/treble shelving, and a mid boost/cut, at an unknown frequency point (I estimated around 1kHz, confirmed below).
Both the pre-amp output and input then go to one half of a DPDT push-pull switch attached to the volume pot (B500K). By default, the volume pot sees the pre-amp output. Pulling the switch sends the pre-amp input instead to the volume pot, which is the direct passive signal out of the blend pot. The pre-amp is not switched off in passive mode and there's a reason for that: switching on the pre-amp would bring the bias voltage at the output capacitor from 0 to 4.5V or so, and that charging of the output capacitor would make quite a bang into your amp. Normally you never see that charging since you plug into the bass, which turns on the pre-amp, before you turn on the amp. The active/passive switch still pops loudly when used, unless you turn the volume down first.
Note the complete absence of any tone cap in the signal path. This will become very important later.
The bridge will not work as-is with large tapered B strings (like a Kalium 148, or a D'Addario 145T): The saddle height screws are too short to raise the action of a tapered B string, so the B string saddle need 3mm shims under both screws. There is just enough adjustment range for a tapered E string.
The shims make the saddle intonation screw stick out so much it will touch the tapered B string when plucked. The screw must be set back with wire (or a bushing) around the thread, at the head. (This tucks into the bridge, so it's invisible.)
The slight bulge at the end of a non-tapered B string larger than about .130 will not fit through the bridge. I had to drill it out to 11/64". The bridge is NOT made of brass (only plated), but of some unknown very soft silver metal (zinc?), so drilling was very easy and neat.
The stock Spectra has a boomy, muddy sound unless both pickups are in single-coil mode and blended 50/50. Single-coil mode hums too much in any other blend configuration.
To reduce electrical noise, I re-shielded the control cavity. Based on Bruce Johnson's notes on TalkBass, I chose Solo Music Gear's carbon conductive paint as it's very cheap, well-reviewed, water-based, nearly odorless, and has good spread and adhesion. I brushed on one coat and let cure overnight. That brought the sheet resistance from 300 Ohms/sq to 30 Ohms/sq. Once the electronics were re-installed, their additional grounding wires brought that down to 15 Ohms/sq.
To eliminate magnetic hum and reduce the muddy tone, I hardwired the neck and bridge pickups as parallel humbuckers. I tried having a series/parallel switch for the neck pickup, mimicking the same option on my 1992 Fender Precision Plus, but it is not useful here. When in series the neck pickup is still boomy, lessens the natural 400 Hz mid-scoop when placed in parallel with the bridge pickup, and has a high enough impedance that it lets through a bit of noise.
To reduce the muddy tone further, I replaced the blend knob, which was likely loading down the pickups too much, with a mini toggle DP3T on-on-on switch wired up to give the common neck/parallel/bridge pickup selection, also an idea copied from my Precision Plus.
To improve headroom and clarity even more, I raised the pickups as much as they would go, then fixed the warbling by playing high up the neck on both the treble and bass sides and then lowering the same side of the neck pickup very slightly as needed. No adjustment was needed to the bridge pickup.
After these modifications, I have the following configurations available:
Having these pickups as parallel humbuckers puts their resonant frequency peak very high: They sounded beautifully clear, but I would also hear (and see on a spectrum analyzer) nasty metallic fret and string noise all over 5 kHz to 10 kHz.
I analyzed the pickups and pre-amp with a Bode Analyzer on a Red Pitaya scope. It's not the best tool for the job, as it's meant for RF, but it's enough to find the resonant frequency and any boost/cut. The test signal was fed to an unshielded 100 mH inductor simply stuck to one of the pickup poles by its own magnetism.
Right away, the cause of the metallic noise was evident: when wired as parallel humbuckers, these pickups resonate, quite sharply, at around 6 kHz! There is no tonal signal in that region, only clang and other impulse noises. There is no loading capacitor anywhere, even inside the potted pre-amp, as determined by the resonnant frequency not appreciably changing when the pickups were solo or in parallel.
Some quick experimentation showed that placing a 4.4nF load capacitor in parallel with each pickup would bring their resonance back down to around 2.4 to 2.8 kHz (depending which pickup, and if they were both in parallel like a J-Bass), which is much more in line with conventional pickup+tone cap configurations, and happens to place the resonance peak in the area of highest sensitivity to the human ear (re: Fletcher-Munson Curve), and attenuates everything past that point at -12dB/8ve, where little to no tonal signal exists.
Some simple tests on the pre-amp bass/mid/treble EQ shows it to be a plain-vanilla Baxandall-like bass/treble shelving centered at close to 1KHz, and the mid boost/cut is also centered at about 1KHz. Max boost/cut in all cases is +/- 12dB.
Some more tests shows that I'm better off lowering the gain trimpot from the stock halfway mark to nearly fully CCW, which is near unity gain:
At near unity gain, the base level output of the pickups, and also of pre-amp (set flat), is about 100mVpp. A perfectly normal instrument-level signal.
The pre-amp shelving isn't restrained, so the Bass shelf affect everything below 1 kHz, including sub-sonic noise, and the Treble shelf affects everything above 1kHz, including all the fret and string noise all the way up to 10 kHz, and maybe more. Thus, boosting the Bass causes clipping and muddy sound, and boosting the Treble amplifies all the non-tonal metallic noise.
To reduce sub-sonics, I placed a 3.3nF capacitor in series between the pickup selector switch and the pre-amp input. With the (assumed) 1MOhm input impedance of the pre-amp, this creates a passive 1st order HPF with the -3dB point at about 48 Hz. This filter attenuates sub-sonic junk from the strings and the fundamentals of the B and E strings slightly, which very few speakers can even reproduce, and *might* balance the output of the B and E strings, which would be quite a bit louder due to their heavy gauges from using a balanced tension set. It seems to work fine.
I would love to have a 2nd order HPF, but it can't be built passively for 48 Hz without large, heavy, and expensive inductors. Conversely, a small inductor and a huge capacitance also won't work, as the Q factor will be so low the filter will be useless. Ditto for 2nd order passive RC networks. So this is a place where adding a unity buffer (I like the OPA145/141 parts) would enable a 2nd order RC HPF at least. Maybe another day.
Common bass mixing practice, and reports from users of HPF/LPF pedals in live settings, seems to be to EQ to remove everything above 5KHz, and I confirmed this by listening to solo bass performances through a 31-band EQ: even with bright slap/pop technique, everything above 5 kHz can be cut and not alter the sound significantly. The cut can even go down to about 3 kHz and not change the basic tone (i.e.: it will sound the same in a mix). Though, the cut must have a smooth roll-off (e.g.: -12dB/octave) because a sharp brick-wall at 3KHz gives a bass a telephone-like hollow sound. This might be an EQ artifact.
At the output of the pre-amp, I placed a passive 2nd order LPF with a -3dB point at about 3.6KHz and a Q of about 1.58, so a +4dB peak at 2.2 KHz (but there is a small mistake in these numbers, see later). A resonance peak of +4dB matches many existing commercial pickups, so should sound "normal".
This filter is composed of 1kOhms in series feeding a 100mH inductor, followed by a 40nF cap to ground. This filter removes even more metallic string noise, fret click, fret buzz, and some EMI sources (LED lamps) I noticed at 6KHz (worsened by the initial untuned resonance of the pickups).
The inductor is unshielded (it's what I had on-hand), in a shielded control cavity, and so it does pick up a bit of magnetic 60 Hz noise (and its harmonics), but it's tolerable even in electrically noisy rooms and much more quiet than a single-coil pickup, even near electronic equipment and electric motors. Next version will use a shielded inductor.
I went for 2.5 to 3-ish kHz filter instead of 5 kHz for three reasons:
This filter makes a headphone amp sound more like a bass amp with a cab simply by removing the distracting noises that don't get reproduced as loudly on a bass cab, but will be mercilessly reproduced by full-range studio headphones. It should have little effect on a real bass amp since their speakers usually reach up to 4-5 kHz (tweeters notwithstanding).
There is one downside: the HPF and LPF filters don't play well with eachother when connected directly by switching to passive mode. The total output drops to almost nothing and isn't usable. Having the pre-amp in the middle to isolate them is a must. I could have used that unused set of poles on the push/pull active/passive switch to simply bypass the HPF when in passive mode, eliminating the capacitive voltage divider with the 40nF in the LPF and restoring the passive output.
Given the total pickup+LPF attenuation of -24dB/8ve at about 2.5 kHz, I can see pretty much nothing on the spectrum analyzer past 2.5 kHz, and only some initial attack noise up to 5 kHz if I try hard, even with the treble EQ maxed-out. It's marvelous. It's the best of both worlds: simple shelving controls with the neat boundaries of filter-based pre-amp.
One final note: The output LPF can be affected by the capacitance of the cable to the amp (lowering the cut-off frequency), and by the amp's input impedance (lowering the Q), but a cable's typical capacitance is in the range of a few hundred pF to maybe 2nF (for a loooong cable), and even a low input impedance amp input sits at a few hundred kOhms, dwarfing the 1K output impedance here. I might notice an effect, but it's not very likely except in the most extreme setups where I would have already needed a clean boost or other buffer.
Adding the passive filters really fixes things. Before, any bass boost was muddy since the shelf went way past down the B and E fundamentals, and any treble boost highlighted string and fret noise. Now, boosting the bass is heavy and warm, without farting or muddiness. Boosting the treble makes the sound bright and the string attack very present, and does bring back a bit of string/fret noise, but nothing unpleasant. The interaction of the filters and of the treble and mid controls gives a new range of tones.
Here's the schematic of the modified electronics. However, when I write up things like this late at night, I miss details, like the 500K volume pot (which gives 333K in parallel with the usual 1M amp input impedance), or the fact the inductor has about 210 Ohms resistance, which is a 20% difference from 1K alone! So the Q values I wrote are lower in actuality, but again, precise values won't substantially change this circuit, so long as the ear says it sounds good. These are also all ordinary, cheap parts. No 1% tolerances here.
I re-strung with long scale D'Addario Chromes, 40-55-75-95-132 for balanced tension with the B string (about 40lbs per string), which just fit on the 35" scale (the silks reach the tuner side of the nut). The neck relief went down to a little less than 1/64" (about 0.010") at the 7th fret without any fret buzz. The truss rod is barely doing any work here. (I know from a past string set that the neck can handle 250lbs of tension well.) The action at the 17th fret goes from 3/64" at G to 6/64" at B, again without any fret buzz. There is no clank unless I dig in. Intonation adjustment was minimal and uniform, with about 1/4" total difference from the G to the B bridge saddles. With a precise strobe tuner to get the intonation just right (especially needed for the heavy B string), all strings stay in tune with eachother up to about the 15th fret, and the G/D/A strings play well further up the neck.
I can choose more good tones, both bright and dark, from the pre-amp. I'm playing more freely now since I don't have to fret like I'm walking on eggshells to reduce string/fret noise. The controlled sub-low end from the HPF means I'm not driving effects into clipping whenever I raise the bass shelf control, and the atonal high noise removed by the tuned pickups and LPF means any distortion isn't driven by that noise, and so doesn't sound shrill and buzzy, and I can control its tone via the B/M/T pre-amp controls.
The only downside to all these changes is that now I'm left with a "producer switch": the original DPDT coil-split switch is now unused and I'm not sure what to do with it, if anything.