Over the years I've analized / modded / built a number of onboard bass preamps, for some of which there isn't much info out there, and I thought it would be nice to share all the info I have in a single thread. As I go along I'll update this with an index to the relevant posts. I'll also discuss some general topics such as opamp choice (post #2, post #27 and some suggestions for SMD opamps by @Passinwind in post #23) and pickup loading (post #3). These are some of the preamps I'll cover, in no particular order:
- A DIY 2-band Baxandall (post #4) I designed and made for a friend's Warmoth Jazz Bass
- Mighty Mite MM114 (post #6, no longer produced, see page 3 of this catalogue), it came in a Tobias Toby Deluxe 5 I bought back in 2001. An opamp-based, 2-band Baxandall eq with a couple of unusual twists
- Seymour Duncan STC-2P (post #8) and STC-2ASB (the Steve Bailey version, post #9), used in two Warmoth builds. Both are 2-band eq's that use opamps and transistor-based gyrators that lend themselves to many interesting mods (post #12 ff)
Edit 4-Nov-24: There was a small mistake in the previous posts, corrected schematics and sim files in post #10, plus PCB component layout and wiring diagram.
- An analysis of the 2-band Stingray preamp (post #16) and a DIY version based on it (post #17), significantly modded to get a more uniform response and to be able to use standard, easy to find pots
- Info about the 3-band Stingray preamp (post #20) for completeness (I have no experience with it) and possible mods (post #28)
Comments and additions always welcome.
Edit: the attached onboardbass.txt file contains the opamp and transistor models required for some of the simulation files posted below.
- A DIY 2-band Baxandall (post #4) I designed and made for a friend's Warmoth Jazz Bass
- Mighty Mite MM114 (post #6, no longer produced, see page 3 of this catalogue), it came in a Tobias Toby Deluxe 5 I bought back in 2001. An opamp-based, 2-band Baxandall eq with a couple of unusual twists
- Seymour Duncan STC-2P (post #8) and STC-2ASB (the Steve Bailey version, post #9), used in two Warmoth builds. Both are 2-band eq's that use opamps and transistor-based gyrators that lend themselves to many interesting mods (post #12 ff)
Edit 4-Nov-24: There was a small mistake in the previous posts, corrected schematics and sim files in post #10, plus PCB component layout and wiring diagram.
- An analysis of the 2-band Stingray preamp (post #16) and a DIY version based on it (post #17), significantly modded to get a more uniform response and to be able to use standard, easy to find pots
- Info about the 3-band Stingray preamp (post #20) for completeness (I have no experience with it) and possible mods (post #28)
Comments and additions always welcome.
Edit: the attached onboardbass.txt file contains the opamp and transistor models required for some of the simulation files posted below.
Last edited:
Opamp choice
Many opamp-based commercial products use the venerable TL06x (datasheet). It has the required high impedance FET input and a very low supply current of 0.2 mA / amp (so 0.4 mA for a dual TL062) providing very good battery life. Noise is high and THD not even specified, but these aren't critical in this application and I find it perfectly adequate most of the time.
The main caveat is that, with a 9 V supply, the maximum peak output voltage is around +/-3 V, so around 2 VRMS, which may lead to clipping if you use high output pickups and/or play hard and/or boost the bass significantly. This isn't a problem if you use a two-battery 18 V supply, but if you want to make the most of the headroom available with a 9 V supply, these are some alternatives I've used, all still available in a DIP package at the time of writing:
- TLC226x (datasheet): this is the "natural" replacement for the TL06x for supplies below 16 V total. It has the same 0.2 mA / amp supply current and a Rail-To-Rail output, plus a much lower noise (Vn @ 1 kHz = 12 nV/√Hz vs 42 for the TL06x) if that's a concern.
- AD822 (datasheet): an excellent but expensive opamp, also Rail-To-Rail output, lower noise (16 nV/√Hz) and good capacitive load drive. Higher supply current though at 0.8 mA / amp. Maximum supply is 36 V total, so it could be used in a 18 V system for more headroom that you'll ever need (probably!).
- TLV237x (datasheet): this is Rail-To-Rail both input and output, useful if you use high output pickups, such as a Stingray wired in series. Noise is comparable to the TL06x (39 nV/√Hz), supply current is 0.5 mA / amp, max. supply 16 V.
For convenience, later I'll put together and post a .lib file with all the spice models required for the simulations I'll be posting, including all of these opamps.
Many opamp-based commercial products use the venerable TL06x (datasheet). It has the required high impedance FET input and a very low supply current of 0.2 mA / amp (so 0.4 mA for a dual TL062) providing very good battery life. Noise is high and THD not even specified, but these aren't critical in this application and I find it perfectly adequate most of the time.
The main caveat is that, with a 9 V supply, the maximum peak output voltage is around +/-3 V, so around 2 VRMS, which may lead to clipping if you use high output pickups and/or play hard and/or boost the bass significantly. This isn't a problem if you use a two-battery 18 V supply, but if you want to make the most of the headroom available with a 9 V supply, these are some alternatives I've used, all still available in a DIP package at the time of writing:
- TLC226x (datasheet): this is the "natural" replacement for the TL06x for supplies below 16 V total. It has the same 0.2 mA / amp supply current and a Rail-To-Rail output, plus a much lower noise (Vn @ 1 kHz = 12 nV/√Hz vs 42 for the TL06x) if that's a concern.
- AD822 (datasheet): an excellent but expensive opamp, also Rail-To-Rail output, lower noise (16 nV/√Hz) and good capacitive load drive. Higher supply current though at 0.8 mA / amp. Maximum supply is 36 V total, so it could be used in a 18 V system for more headroom that you'll ever need (probably!).
- TLV237x (datasheet): this is Rail-To-Rail both input and output, useful if you use high output pickups, such as a Stingray wired in series. Noise is comparable to the TL06x (39 nV/√Hz), supply current is 0.5 mA / amp, max. supply 16 V.
For convenience, later I'll put together and post a .lib file with all the spice models required for the simulations I'll be posting, including all of these opamps.
Pickup loading
If you want to know all the details, take a look at "The secrets of electric guitar pickups" by Helmuth Lemme (it's a web archive link because my antivirus complains about the certificate if I try to go directly to the page), here I'll keep it short: a passive pickup can be modeled as a voltage source in series with an inductance and resistance plus a capacitance to ground. This, together with the load comprising the volume and tone circuit, the cable capacitance and the amplifier input impedance, results in a 2nd order low-pass response, generally with a resonant peak before roll-off. The cut-off frequency, as well as the resonant peak's frequency and height is what gives each pickup + load combination a particular "voice".
In the case of a preamp, its input impedance plus any pots placed before it will constitute the load of the pickup and affect its sound significantly, that's why I thought I'd give some background on this topic.
As an example, here's the response of a typical Jazz Bass pickup (the article above has a link to a table with measured I and C values for many guitar and bass pickups) in three different configurations, to show the effect of different pot values and different cable capacitance: 1) connected to the stock 2 x 250k volume pots (assuming the other pickup is off) + 250k tone pot with 47n tone cap, a typical 500pF cable capacitance and a typical 1M amp input impedance, 2) with 500k instead of 250k pots and 3) with stock pots and a 1.5n cable capacitance (so a cable three times as long):
As you can see, 500k pots results in a higher resonant peak, and a longer cable brings down the cut-off frequency and also increases the peak height. Note that I don't show the effect of a different tone cap because, with the tone pot fully open, it's pretty much negligible. The sim file is attached.
If you want to know all the details, take a look at "The secrets of electric guitar pickups" by Helmuth Lemme (it's a web archive link because my antivirus complains about the certificate if I try to go directly to the page), here I'll keep it short: a passive pickup can be modeled as a voltage source in series with an inductance and resistance plus a capacitance to ground. This, together with the load comprising the volume and tone circuit, the cable capacitance and the amplifier input impedance, results in a 2nd order low-pass response, generally with a resonant peak before roll-off. The cut-off frequency, as well as the resonant peak's frequency and height is what gives each pickup + load combination a particular "voice".
In the case of a preamp, its input impedance plus any pots placed before it will constitute the load of the pickup and affect its sound significantly, that's why I thought I'd give some background on this topic.
As an example, here's the response of a typical Jazz Bass pickup (the article above has a link to a table with measured I and C values for many guitar and bass pickups) in three different configurations, to show the effect of different pot values and different cable capacitance: 1) connected to the stock 2 x 250k volume pots (assuming the other pickup is off) + 250k tone pot with 47n tone cap, a typical 500pF cable capacitance and a typical 1M amp input impedance, 2) with 500k instead of 250k pots and 3) with stock pots and a 1.5n cable capacitance (so a cable three times as long):
As you can see, 500k pots results in a higher resonant peak, and a longer cable brings down the cut-off frequency and also increases the peak height. Note that I don't show the effect of a different tone cap because, with the tone pot fully open, it's pretty much negligible. The sim file is attached.
DIY 2-band Baxandall
This is as simple a 2-band bass preamp as it can be: one opamp section is the input buffer (required if you want to present a constant load to the pickups) and the other is used for the Baxandall tone controls. There are a number of implementations of the latter (see e.g. Douglas Self, "Small Signal Audio Design"), this is the one with the lowest component count. My friend wanted to fit the preamp and battery inside the rather small Jazz Bass control cavity without further routing, so every cap and resistor counts.
The bass was wired with the usual 250k volume pots for each pickup before the preamp, so the 200k input resistor makes up for the parallel of the missing 250k tone control and 1M amp input impedance and the 470p simulates the cable capacitance, so that the pickup load is roughly similar to that of a passive bass (see "pickup loading" above). I used the AD822 because it was what I had in stock at the time that was suitable, but the circuit will work just fine with any of the other opamps mentioned above.
As for the tone controls, with 50k pots and the component values shown you get about +/-18 dB boost/cut @ 40 Hz and 10 kHz and the mid point is at ~700 Hz:
This is all very easy to adjust to taste by playing with the resistor and cap values:
Also if you want to use pots of a different value, simply scale R4, R7, R9 by the same amount and C4, C5 by the inverse (e.g. with 100k pots, multiply the resistances by 2 and divide the capacitances by 2).
The attached simulation will work as it is (no additional .lib file required) as the AD822 model is included in LTSpice.
This is as simple a 2-band bass preamp as it can be: one opamp section is the input buffer (required if you want to present a constant load to the pickups) and the other is used for the Baxandall tone controls. There are a number of implementations of the latter (see e.g. Douglas Self, "Small Signal Audio Design"), this is the one with the lowest component count. My friend wanted to fit the preamp and battery inside the rather small Jazz Bass control cavity without further routing, so every cap and resistor counts.
The bass was wired with the usual 250k volume pots for each pickup before the preamp, so the 200k input resistor makes up for the parallel of the missing 250k tone control and 1M amp input impedance and the 470p simulates the cable capacitance, so that the pickup load is roughly similar to that of a passive bass (see "pickup loading" above). I used the AD822 because it was what I had in stock at the time that was suitable, but the circuit will work just fine with any of the other opamps mentioned above.
As for the tone controls, with 50k pots and the component values shown you get about +/-18 dB boost/cut @ 40 Hz and 10 kHz and the mid point is at ~700 Hz:
This is all very easy to adjust to taste by playing with the resistor and cap values:
- Decrease/increase R4, R7 to increase/decrease, respectively, the boost/cut of the bass control, R9 for the treble control
- Decrease/increase C4 to increase/decrease, respectively, the turnover frequency of the bass control, C5 for the treble control
Also if you want to use pots of a different value, simply scale R4, R7, R9 by the same amount and C4, C5 by the inverse (e.g. with 100k pots, multiply the resistances by 2 and divide the capacitances by 2).
The attached simulation will work as it is (no additional .lib file required) as the AD822 model is included in LTSpice.
A couple of pictures and one of the bass, which turned out really nice... In the end I had to connect the pots with wires so I could fit preamp and battery in the cavity. I'm not posting the PCB because this was many years ago and I can't find it, too many to keep track of... In any case nowadays I'd do this in Vero board, it's pretty straightforward.
Attachments
Mighty Mite MM114
First things first: this preamp is intended for active pickups, hence the 2x50k balance control (active pickups are internally buffered and don't care much about the load). However the Tobias Toby Deluxe 5 uses passive ones (JB-style single coils with a Tobias logo, but I read somewhere they were also sourced by Mighty Mite), so it's no wonder that when I first plugged it in I thought, "wow, that's dark", despite the maple body, maple neck and ebony fretboard. Your guess is as good as mine as to why Tobias decided to do this, but here's what happens when you load a passive JB pickup with this preamp, compared to the stock JB passive circuit (2 x 250k volume pots, 250k tone control, 500 pF cable capacitance, 1M amp input impedance):
Anyway, here's the schematic (sim file attached, .lib file with models in the first post so I can edit it if required) and the response of the tone control section itself, which is kind of unusual:
The bass control is pretty standard, with a boost/cut of around +14/-15 dB @ 40 Hz, but as you can see the treble is very asymmetrical in the extremes, with the typical shelf boost (around +11 dB @ 10 kHz) and a cut that's more like a low-pass response, dropping to -28 dB @ 10 kHz. Another thing of note is that the mid point is pretty low, at around 400 Hz.
Finally, the circuit includes a sort of pre-emphasis / de-emphasis arrangement, where the input buffer boosts the highs significantly and the divider at the output formed by R110, R113 and C111 restores flatness. Here's the response at the output of the first opamp section (in green) vs. the overall output:
Not quite sure what the purpose of this is... A kind of "Dolby-esque" hiss reduction? Not usually a problem in this context... Speculation welcome!
Coming up next: my mods to adapt it to passive pickups and have more "normal" tone controls.
First things first: this preamp is intended for active pickups, hence the 2x50k balance control (active pickups are internally buffered and don't care much about the load). However the Tobias Toby Deluxe 5 uses passive ones (JB-style single coils with a Tobias logo, but I read somewhere they were also sourced by Mighty Mite), so it's no wonder that when I first plugged it in I thought, "wow, that's dark", despite the maple body, maple neck and ebony fretboard. Your guess is as good as mine as to why Tobias decided to do this, but here's what happens when you load a passive JB pickup with this preamp, compared to the stock JB passive circuit (2 x 250k volume pots, 250k tone control, 500 pF cable capacitance, 1M amp input impedance):
Anyway, here's the schematic (sim file attached, .lib file with models in the first post so I can edit it if required) and the response of the tone control section itself, which is kind of unusual:
The bass control is pretty standard, with a boost/cut of around +14/-15 dB @ 40 Hz, but as you can see the treble is very asymmetrical in the extremes, with the typical shelf boost (around +11 dB @ 10 kHz) and a cut that's more like a low-pass response, dropping to -28 dB @ 10 kHz. Another thing of note is that the mid point is pretty low, at around 400 Hz.
Finally, the circuit includes a sort of pre-emphasis / de-emphasis arrangement, where the input buffer boosts the highs significantly and the divider at the output formed by R110, R113 and C111 restores flatness. Here's the response at the output of the first opamp section (in green) vs. the overall output:
Not quite sure what the purpose of this is... A kind of "Dolby-esque" hiss reduction? Not usually a problem in this context... Speculation welcome!
Coming up next: my mods to adapt it to passive pickups and have more "normal" tone controls.
Last edited:
Mighty Mite MM114 modded
Job #1 was to adapt the input impedance to passive pickups. Originally the bass had two JB-type pickups, but I routed it for a Stingray-type soapbar at the bridge and also replaced the balance pot with a rotary switch for fixed coil combos, so, load-wise, only the 500k vol. pot was left before the preamp. I replaced the 1M R101 resistor with a 100k, the 510k R102 resistor with a 1M and the 10p C102 cap with a 330p (this would simulate a shorter cable than the typical 500p, but the difference is small and I probably had more 330p caps in stock).
I also replaced the TL062 with a TLC2262 for better headroom.
As for the treble tone control, I added a stopper 6.8k resistor on the cut side to make it behave like a normal Baxandall shelf rather than the stock low-pass, removed the 33p C106 cap and replaced the 6.8n C107 cap with a 1.5n to move up the turnover frequency, which was too low for my taste. This is the result:
Sim file attached as usual.
Job #1 was to adapt the input impedance to passive pickups. Originally the bass had two JB-type pickups, but I routed it for a Stingray-type soapbar at the bridge and also replaced the balance pot with a rotary switch for fixed coil combos, so, load-wise, only the 500k vol. pot was left before the preamp. I replaced the 1M R101 resistor with a 100k, the 510k R102 resistor with a 1M and the 10p C102 cap with a 330p (this would simulate a shorter cable than the typical 500p, but the difference is small and I probably had more 330p caps in stock).
I also replaced the TL062 with a TLC2262 for better headroom.
As for the treble tone control, I added a stopper 6.8k resistor on the cut side to make it behave like a normal Baxandall shelf rather than the stock low-pass, removed the 33p C106 cap and replaced the 6.8n C107 cap with a 1.5n to move up the turnover frequency, which was too low for my taste. This is the result:
Sim file attached as usual.
Seymour Duncan STC-2P
The classic graphic eq topology using gyrators has been well covered elsewhere, see e.g. Rod Elliott's article (section 14). Here the gyrators are transistor-based (section 7.1 in that article) and 4 of them are used: two for the bass and treble controls and two for the push-pull activated contour curve. This is a clever way of implementing a more complex eq than a typical 2-band that still fits in a tiny 55 x 35 mm PCB (see attached picture).
Note that these are not Baxandall type shelf controls: they are bell filters, as you can see in the frequency response graphs below. The bass control is rather broad and centered around 30 Hz, so in practice it isn't that different from a typical shelf, with a boost/cut of around +14/-15 dB @ 40 Hz. The treble on the other hand is much narrower, especially towards the extremes, with a boost/cut of around +/- 17 dB centered around 5 kHz:
Some comments about the circuit:
https://www.seymourduncan.com/images/wiring-diagrams/TC_STC2.jpg
The contour curve consists of a bass boost around 90 Hz and a mid cut around 630 Hz. Here's the response for various settings of the trim pots from max (~ +9 dB @ 90 Hz / -11 dB @ 630 Hz) to min (~ +3 dB @ 75 Hz / -3 dB @ 630 Hz). Of course you can adjust the bass boost and mid cut independently if you wish:
The sim file is attached. I'll cover cloning the circuit and possible mods in another post, watch this space.
The classic graphic eq topology using gyrators has been well covered elsewhere, see e.g. Rod Elliott's article (section 14). Here the gyrators are transistor-based (section 7.1 in that article) and 4 of them are used: two for the bass and treble controls and two for the push-pull activated contour curve. This is a clever way of implementing a more complex eq than a typical 2-band that still fits in a tiny 55 x 35 mm PCB (see attached picture).
Note that these are not Baxandall type shelf controls: they are bell filters, as you can see in the frequency response graphs below. The bass control is rather broad and centered around 30 Hz, so in practice it isn't that different from a typical shelf, with a boost/cut of around +14/-15 dB @ 40 Hz. The treble on the other hand is much narrower, especially towards the extremes, with a boost/cut of around +/- 17 dB centered around 5 kHz:
Some comments about the circuit:
- The topology is conventional in the sense of using a dual opamp (TL062) with one section as the input buffer and the other one for the eq circuit.
- The supply section has an extra voltage divider (R14, R15) to provide ~6.3 V (labeled Va) for biasing the transistor bases.
- The tone pots are 100k W-taper (what Alpha calls Taper "B" here). This is required for the controls to operate smoothly across the whole range: if you use linear pots, there's very little variation over a wide range around the center, and the whole eq action is "compressed" towards the extremes.
- As you can see in the PCB picture, there are two 10k trim pots. These allow fine-tuning of the contour curve and come pre-set to around 3k. More on this below.
- The blend pot (2x250k in the passive pickups version, 2x100k in the active pickups one), not shown in the simulation, goes before the preamp, the volume pot (10k log + push-pull) is at the output. I'll have more to say about the input impedance of the circuit in a later post.
https://www.seymourduncan.com/images/wiring-diagrams/TC_STC2.jpg
The contour curve consists of a bass boost around 90 Hz and a mid cut around 630 Hz. Here's the response for various settings of the trim pots from max (~ +9 dB @ 90 Hz / -11 dB @ 630 Hz) to min (~ +3 dB @ 75 Hz / -3 dB @ 630 Hz). Of course you can adjust the bass boost and mid cut independently if you wish:
The sim file is attached. I'll cover cloning the circuit and possible mods in another post, watch this space.
Attachments
Seymour Duncan STC-2ASB
This is the Steve Bailey version, intended to be used with the AJB-2ASB active pickups, which, as far as I can tell, can only be bought as a set together with the preamp. The whole set is supposed to be voiced specifically for fretless bass, and I must say it does sound great in mine.
The differences with the STC-2P above are:
And the contour curve:
This is the Steve Bailey version, intended to be used with the AJB-2ASB active pickups, which, as far as I can tell, can only be bought as a set together with the preamp. The whole set is supposed to be voiced specifically for fretless bass, and I must say it does sound great in mine.
The differences with the STC-2P above are:
- As it is intended for active pickups, the included blend pot is 2 x 100k instead of 2 x 250k, but the PCB and circuit are the same for the active and passive versions.
- The component values for the bass and treble controls are different, the bass filter centered lower around 25 Hz, resulting in a bass boost/cut of around +11 / -12 dB @ 40 Hz, and the treble filter centered around 6.5 kHz, a bit broader and with a boost/cut of around +/-15 dB at that frequency.
- Instead of one boost and one cut, here the contour curve has two boosts, one at around 600 Hz and one at around 5.5 kHz. Again these are adjustable with the trim pots.
- It uses PN2484 transistors instead of MPSA18, which doesn't really impact the circuit operation. More on transistor choice in a later post.
And the contour curve:
Minor correction to the schematics above
Apologies but as I was double-checking the PCB, I realized that I had missed something: there are two 1M resistors effectively connected across each pole of the push/pull switch, presumably to keep the two sides at the same voltage when open to avoid pops when closing it. In the previous schematic I had included only one of the resistors and it was connected to the wrong side of the corresponding trimmer.
This has no effect at all on the frequency responses posted and any sims you may have run, but here are the corrected schematics and sim files (attached):
STC-2P
STC-2ASB
What I was double-checking was this PCB layout diagram that I made to help locate components when modding (the component value are for the STC-2P, the STC-2AB has the same PCB and component layout):
And, for completeness, here's the wiring diagram in case the link I posted before disappears:
Apologies but as I was double-checking the PCB, I realized that I had missed something: there are two 1M resistors effectively connected across each pole of the push/pull switch, presumably to keep the two sides at the same voltage when open to avoid pops when closing it. In the previous schematic I had included only one of the resistors and it was connected to the wrong side of the corresponding trimmer.
This has no effect at all on the frequency responses posted and any sims you may have run, but here are the corrected schematics and sim files (attached):
STC-2P
STC-2ASB
What I was double-checking was this PCB layout diagram that I made to help locate components when modding (the component value are for the STC-2P, the STC-2AB has the same PCB and component layout):
And, for completeness, here's the wiring diagram in case the link I posted before disappears:
Last edited:
Cloning the SD preamps
Just a couple of comments if you want to clone these. 100k W-taper pots aren't easy to find, but you can actually use 10k linear ones without the response changing too much:
As for the transistors, the MPSA18 are currently available from the usual sources, but through-hole parts keep disappearing at an alarming rate and maybe when you read this they are gone. Fortunately the transistor isn't critical at all, you could use pretty much any small signal NPN with a high beta, such as 2N5089, BC54xC, KSC1845, etc.
Just a couple of comments if you want to clone these. 100k W-taper pots aren't easy to find, but you can actually use 10k linear ones without the response changing too much:
As for the transistors, the MPSA18 are currently available from the usual sources, but through-hole parts keep disappearing at an alarming rate and maybe when you read this they are gone. Fortunately the transistor isn't critical at all, you could use pretty much any small signal NPN with a high beta, such as 2N5089, BC54xC, KSC1845, etc.
Modding the SD preamps (part 1)
I haven't modded mine because I'm happy with them as they are, but there are lots of possibilities here, so let's explore some of them.
First of all we can play with the load (post #3) if using passive pickups. Here's a comparison of the response of a JB pickup with the standard, passive JB circuit (2 x 250k volume pots, 250k tone pot, 500p cable capacitance, 1M amp input impedance) in green, the stock SD preamps in red and the result of adding a 470p cap in parallel with the 1M input resistor (in the PCB layout above, it's the one to the left of the opamp near the left edge of the board):
A smaller cap will result in a resonant peak that's lower in height and higher in frequency, a larger cap will do the opposite. If you want to match the passive bass response exactly, add also a 270k resistor in parallel.
Another thing you may want to do is tweak the gyrators to adjust the shape of the different filters more to your taste. While looking into this, I came up with a Gyrator Calculator in LTSpice that I thought deserved its own thread here. If you use it, note that the gyrators are drawn "upside down" with respect to the ones here, this is how the components correspond to each other:
In the calculator you also enter R3, which is the value of the input and feedback resistors around the opamp, which in this case are 10k. You would also have to edit the pot parameters to match the ones used here.
You can also tweak the filters directly in the preamp simulations, it isn't that difficult to simply change the component values until you're happy with the result if you follow these rules of thumb:
I haven't modded mine because I'm happy with them as they are, but there are lots of possibilities here, so let's explore some of them.
First of all we can play with the load (post #3) if using passive pickups. Here's a comparison of the response of a JB pickup with the standard, passive JB circuit (2 x 250k volume pots, 250k tone pot, 500p cable capacitance, 1M amp input impedance) in green, the stock SD preamps in red and the result of adding a 470p cap in parallel with the 1M input resistor (in the PCB layout above, it's the one to the left of the opamp near the left edge of the board):
A smaller cap will result in a resonant peak that's lower in height and higher in frequency, a larger cap will do the opposite. If you want to match the passive bass response exactly, add also a 270k resistor in parallel.
Another thing you may want to do is tweak the gyrators to adjust the shape of the different filters more to your taste. While looking into this, I came up with a Gyrator Calculator in LTSpice that I thought deserved its own thread here. If you use it, note that the gyrators are drawn "upside down" with respect to the ones here, this is how the components correspond to each other:
In the calculator you also enter R3, which is the value of the input and feedback resistors around the opamp, which in this case are 10k. You would also have to edit the pot parameters to match the ones used here.
You can also tweak the filters directly in the preamp simulations, it isn't that difficult to simply change the component values until you're happy with the result if you follow these rules of thumb:
- The 18k emitter resistors simply set the bias of the transistors and have no effect on the filter, so you won't have to touch those.
- Generally you won't have to touch R1 either, though you can use it to fine-tune the centre frequency if it doesn't land where you want when using standard values for C1. For a given Fc, decreasing R1 increases the required C1 and vice versa.
- The feedback resistor R2 sets the maximum boost/cut gain, decrease it for more gain, increase it for less.
- If you if you increase both C1 and C2 by some factor, Fc will decrease by the same factor and vice versa.
- If you increase C1 and decrease C2 by some factor, Fc stays the same and Q increases by the same factor and vice versa.
Well, the idea of an onboard bass preamp is to be able to adjust your sound while playing and to have more control than the very basic tone pot of a passive bass. For example a contour like the one in the STC-2P above works great to get a good slap tone at the flick of a switch, and the other controls are very useful to re-balance your sound depending on what different songs call for, or when playing in different venues / through different amps / with different musicians.
Modding the SD preamps (part 2)
Adding adjustable bands to this type of EQ is as simple as adding extra pots in parallel with their wiper connected to a gyrator, so with four at your disposal, it would be pretty straightforward to turn this into a 4-band EQ. On the other hand that would mean losing the contour switch, which I personally find very useful. Also four bands sounds like overkill, but I wouldn't mind an extra mids band around 500-600 Hz, where the growl lives... 😎
Luckily, freeing up one of the gyrators is easy: if you connect the wiper of the treble control to ground via a resistor and cap in series, basically you get a Baxandall-type control. 1k8 and 12n works well, which is convenient because we're going to remove the 12n cap from the gyrator that's going to give us the mids control.
Note that what follows is all based on the STC-2P. Let's have a look first at the response we get before we add the extra band:
I mentioned above that easier to find 10k linear pots can also be used in this preamp. Not only that: you can mix and match these with the 100k W-taper ones and it works just fine. So, we add the 10k pot, connect to it the first gyrator and simply replace the 680p cap with a 3n3, the 12n a 150n and the 1k3 resistor with a 2k. This is the final response and schematic (sim file attached):
The .func Rpot(x) at the top is a little trick to be able to step simultaneously the two types of pot. Here's also the modified wiring diagram and PCB component layout:
As I said before, I may actually do this... I'll report back if I do.
Adding adjustable bands to this type of EQ is as simple as adding extra pots in parallel with their wiper connected to a gyrator, so with four at your disposal, it would be pretty straightforward to turn this into a 4-band EQ. On the other hand that would mean losing the contour switch, which I personally find very useful. Also four bands sounds like overkill, but I wouldn't mind an extra mids band around 500-600 Hz, where the growl lives... 😎
Luckily, freeing up one of the gyrators is easy: if you connect the wiper of the treble control to ground via a resistor and cap in series, basically you get a Baxandall-type control. 1k8 and 12n works well, which is convenient because we're going to remove the 12n cap from the gyrator that's going to give us the mids control.
Note that what follows is all based on the STC-2P. Let's have a look first at the response we get before we add the extra band:
I mentioned above that easier to find 10k linear pots can also be used in this preamp. Not only that: you can mix and match these with the 100k W-taper ones and it works just fine. So, we add the 10k pot, connect to it the first gyrator and simply replace the 680p cap with a 3n3, the 12n a 150n and the 1k3 resistor with a 2k. This is the final response and schematic (sim file attached):
The .func Rpot(x) at the top is a little trick to be able to step simultaneously the two types of pot. Here's also the modified wiring diagram and PCB component layout:
As I said before, I may actually do this... I'll report back if I do.
Last edited:
MusicMan Stingray 2-band preamp
Although there's some contradicting information out there, the consensus seems to be that this is the correct schematic of the original circuit (source):
The LM4250 opamp from National Semiconductor (datasheet), which hasn't been available for a while, has a programable bias current via that 2M2 resistor from pin 8 to ground. As stated in the schematic, this sets the current drain to just ~44 uA, so I speculate that it was chosen mainly to provide very long battery life. As you can see in the datasheet, this setting affects other parameters too: in particular, for the given ~4 uA set current, the slew rate will be ~0.08 V/us and the GBP ~150 kHz. These are less than stellar figures to put it mildly, but consider how often a bass pickup signal is going to require the preamp to produce a perfectly clean +/-3.5 V signal at 20 kHz. In any case, some people claim that using that particular opamp does have a significant impact on the sound of the preamp.
As I don't have an LM4250 handy, all I can say is: you might very well think that; I couldn't possibly comment... I will mention though that MusicMan used the TL062 in their 3-band preamp (which I'll cover later in the thread) and, apart from the obvious differences in the controls, some of those who have tried both do mention differences in the overall sound, but these rarely go beyond "subtle" and this despite the fact that the 3-band is a completely different circuit altogether, not just the opamp. Make of it what you will, YMMV, etc.
That out of the way, let's have a look at the circuit. A unique feature is that its input impedance varies significantly with frequency within the band of interest, which will definitely have an impact on frequency response for different pickups (see Pickup loading above). I have a Seymour Duncan SMB-5A (a clone of a'70s Stingray pickup) that, with the coils in parallel (as it's normally wired) I measured to have 2.2k resistance and 1.3 H inductance. Seymour Duncan provides a chart (here) with the resonant frequency of each of their pickups, in this case 8.35 kHz. This means that the parasitic capacitance must be ~280 pF.
The bass uses a 100k log pot and the treble a 1M reverse log pot. I'll asume that the taper is 10%, which is typical, and I'll use the TL062 opamp model. With both controls in the centre position, we get a sort of broad ~+2 dB shelf centered around ~3 kHz instead of the usual resonant peak before roll-off, as well as a broad, slight ~1 dB dip around ~260 Hz:
Also worth noting that, with the 220k feedback resistor, the circuit has a gain of ~6 dB. Other schematics out there show a 100k that makes it roughly unity gain. It seems that both versions exist and it isn't clear whether it was originally 100k and was later changed to 220k or the other way around. In any case, aside from the gain difference, this has little impact on the overall response.
Let's turn now to the tone controls. To understand better how they work, it's worth looking at each of them as if the other one wasn't there:
Finally, let's have a look at the response of the tone controls (sim file attached). It's all very asymmetrical and hard to describe, a great example of "a picture is worth a thousand words", so here it is (the pots are stepped in opposite directions, i.e. when bass goes up, treble goes down and vice versa, this makes it easier to see their response):
You'll find a lot of information about cloning and modding it if you google something like "stingray preamp clone", so I won't cover what others have done here. You can also experiment with the attached simulation and see if you come up with something you think would work better. If you do, please share!
Although there's some contradicting information out there, the consensus seems to be that this is the correct schematic of the original circuit (source):
The LM4250 opamp from National Semiconductor (datasheet), which hasn't been available for a while, has a programable bias current via that 2M2 resistor from pin 8 to ground. As stated in the schematic, this sets the current drain to just ~44 uA, so I speculate that it was chosen mainly to provide very long battery life. As you can see in the datasheet, this setting affects other parameters too: in particular, for the given ~4 uA set current, the slew rate will be ~0.08 V/us and the GBP ~150 kHz. These are less than stellar figures to put it mildly, but consider how often a bass pickup signal is going to require the preamp to produce a perfectly clean +/-3.5 V signal at 20 kHz. In any case, some people claim that using that particular opamp does have a significant impact on the sound of the preamp.
As I don't have an LM4250 handy, all I can say is: you might very well think that; I couldn't possibly comment... I will mention though that MusicMan used the TL062 in their 3-band preamp (which I'll cover later in the thread) and, apart from the obvious differences in the controls, some of those who have tried both do mention differences in the overall sound, but these rarely go beyond "subtle" and this despite the fact that the 3-band is a completely different circuit altogether, not just the opamp. Make of it what you will, YMMV, etc.
That out of the way, let's have a look at the circuit. A unique feature is that its input impedance varies significantly with frequency within the band of interest, which will definitely have an impact on frequency response for different pickups (see Pickup loading above). I have a Seymour Duncan SMB-5A (a clone of a'70s Stingray pickup) that, with the coils in parallel (as it's normally wired) I measured to have 2.2k resistance and 1.3 H inductance. Seymour Duncan provides a chart (here) with the resonant frequency of each of their pickups, in this case 8.35 kHz. This means that the parasitic capacitance must be ~280 pF.
The bass uses a 100k log pot and the treble a 1M reverse log pot. I'll asume that the taper is 10%, which is typical, and I'll use the TL062 opamp model. With both controls in the centre position, we get a sort of broad ~+2 dB shelf centered around ~3 kHz instead of the usual resonant peak before roll-off, as well as a broad, slight ~1 dB dip around ~260 Hz:
Also worth noting that, with the 220k feedback resistor, the circuit has a gain of ~6 dB. Other schematics out there show a 100k that makes it roughly unity gain. It seems that both versions exist and it isn't clear whether it was originally 100k and was later changed to 220k or the other way around. In any case, aside from the gain difference, this has little impact on the overall response.
Let's turn now to the tone controls. To understand better how they work, it's worth looking at each of them as if the other one wasn't there:
- Without the treble control, the bass control is simply a frequency-dependent feedback in a standard inverting opamp configuration
- Without the bass control, the treble control is a sort of asymmetrical Baxandall arrangement, with a reverse log pot and caps of different value at each side
- With the bass pot at 10k, at high frequencies, where the 47n cap approaches a short, we have a standard inverting opamp with equal 220k input and feedback resistors, so unity gain, and at low frequencies, where the 47n cap approaches an open circuit, we have a 10k-10k voltage divider (the 10u cap is just for DC blocking) providing half the output voltage to the 220k feedback resistor, so with half the feedback, we have a ~6 dB bass boost.
- With the treble pot in the centre position, the resistance to the left is 100k and to the right 900k, which combined with the different cap values, results in a ~8 dB treble boost.
- In both cases, the roll-off at high frequencies is due to the 120p cap in parallel with the feedback network (I don't think it's worth showing it but if you remove that cap, both frequency responses continue flat up to 20 kHz).
Finally, let's have a look at the response of the tone controls (sim file attached). It's all very asymmetrical and hard to describe, a great example of "a picture is worth a thousand words", so here it is (the pots are stepped in opposite directions, i.e. when bass goes up, treble goes down and vice versa, this makes it easier to see their response):
You'll find a lot of information about cloning and modding it if you google something like "stingray preamp clone", so I won't cover what others have done here. You can also experiment with the attached simulation and see if you come up with something you think would work better. If you do, please share!
The Cabipre - A Stingray-based DIY 2-band preamp
I've only played an actual MusicMan Stingray with the 2-band preamp once, years ago at a music store. I've always loved the Stingray sound and I know the preamp is a big part of it, so I took my time to investigate it. I found the treble control very responsive and "colourful"; in contrast the bass control felt limited in range and "uninspiring". Also I hated the fact that the pots didn't have centre detents: all my active basses have them and I love that reassuring click when you're adjusting things on the fly in a poorly lit stage.
A few years after that I decided to make me a Warmoth Stingray clone, so it was time to look into the preamp circuit. Since I'm not aware of any log or reverse log pots with a centre detent, job #1 was to try to adapt the circuit to linear pots. There was a lot of trial and error and in the end what I came up with is a two stage circuit using a dual opamp, the first stage for the treble control with a similar topology to the original but adapted to a linear pot, and the second stage a standard Baxandall type bass control, with more uniform action and much less interaction with the treble control.
First here's the frequency response with both controls flat to show the effect of pickup loading compared to the original. I have a bypass, active/passive push-pull, so I show also the response in passive mode. For variety, I wanted to have a different sound when switching to passive, with the typical resonant peak before roll-off:
As you can see the response in active mode isn't exactly the same as the original, but it's close enough for my taste. Here's the tone control action (sim file attached):
The treble control is, again, not exactly the same but close enough for my taste. Some comments:
I've only played an actual MusicMan Stingray with the 2-band preamp once, years ago at a music store. I've always loved the Stingray sound and I know the preamp is a big part of it, so I took my time to investigate it. I found the treble control very responsive and "colourful"; in contrast the bass control felt limited in range and "uninspiring". Also I hated the fact that the pots didn't have centre detents: all my active basses have them and I love that reassuring click when you're adjusting things on the fly in a poorly lit stage.
A few years after that I decided to make me a Warmoth Stingray clone, so it was time to look into the preamp circuit. Since I'm not aware of any log or reverse log pots with a centre detent, job #1 was to try to adapt the circuit to linear pots. There was a lot of trial and error and in the end what I came up with is a two stage circuit using a dual opamp, the first stage for the treble control with a similar topology to the original but adapted to a linear pot, and the second stage a standard Baxandall type bass control, with more uniform action and much less interaction with the treble control.
First here's the frequency response with both controls flat to show the effect of pickup loading compared to the original. I have a bypass, active/passive push-pull, so I show also the response in passive mode. For variety, I wanted to have a different sound when switching to passive, with the typical resonant peak before roll-off:
As you can see the response in active mode isn't exactly the same as the original, but it's close enough for my taste. Here's the tone control action (sim file attached):
The treble control is, again, not exactly the same but close enough for my taste. Some comments:
- The opamp is a rail-to-rail input and output TLV2372 (see opamp choice above).
- The 1M and 100k resistors at the input and output are bleed resistors for the input and output caps. These prevent pops when switching between active and passive. I don't show the switch in the schematic for clarity.
- The tone pots are 100k linear with a centre detent, widely available from the usual suspects (Mouser, Farnell, Digikey, etc.). Actually there's no reason why you couldn't use a different pot value for the bass if you scale the resistors and cap accordingly, but watch out for the load on the first opamp stage which, worst case (at high frequencies and with the treble at minimum) is around the parallel of R7 and R9. As it is that's ~11k, which any suitable opamp should handle without problem.
- The volume pot is 100k log and it's what sets the response in passive mode (the bypass switch is before it). Its value adjusts the resonant peak, if you want it flatter (~1.5 dB) you can use a 50k.
MusicMan Stingray 3-band preamp
And now for something completely different:
Here we have a dual TL062 opamp: the first stage is a second-order, Sallen-Key high-pass filter with an F3 of ~57 Hz and the second stage is a standard 3-band Baxandall arrangement. Also at the input there's that 2n2 cap to ground as the pickup load. This is the resulting frequency response with all controls flat:
The sharp ~16 dB (!) peak at around 2.8 kHz is due to the 2n2 cap and the fact that the resistance to ground at high frequencies is just ~560k from the filter resistor. If you have / plan to build this preamp and want to adjust that peak, it's easy: add a resistor in parallel with the 2n2 cap to lower the peak height (e.g. a 68k will bring it down to ~6 dB) and lower/increase the capacitor to increase/lower its frequency, respectively (e.g. a 1n cap will move the peak to ~3.9 kHz). If you remove it altogether, the peak moves to ~8.3 kHz, the intrinsic resonant frequency of the pickup due to its inductance and parasitic capacitance.
As for the high-pass filter, attenuation at 41 Hz (fundamental of the low E string in a 4-string bass) is ~8 dB, which sounds like a lot to me... Your guess is as good as mine as to why they thought this was necessary and/or desireable, maybe to prevent excessive cone excursion and get a tighter sound with smaller combos? Anyway, you can easily lower the F3 by increasing the 8n2 caps, e.g. 12n will give you F3 = ~37 Hz and a more reasonable attenuation of ~1.5 dB at 41 Hz. You could also simply remove the filter altogether and leave the first stage as a buffer, with pickup loading adjusted to taste.
The 3-band Baxandall implemented with a single opamp stage is well known to have an intrinsic problem: there's a lot of interaction between the bands. If we plot the response stepping the three controls in the same direction, this is what we get (sim file attached):
Not a lot to see here. It helps to fix the mids control (make Rc=25k and Rd=25k) to see better what the bass and treble do:
And then the opposite, fix bass and treble and step mids (swap Ra for Rc and Rb for Rd in the parameters list):
For comparison with previous preamps, the boost/cut at 40 Hz is around +/-14 dB, at 10 kHz around +/-18 dB and the mids, centered at ~440 Hz with very low Q, around +/-7 dB.
As you can see, even with the mids fixed, there's significant interaction between the bass and treble, and the mids looks a bit limited in range. So, all in all it's a "quirky" preamp... Of course I've never played with it and maybe if I did I'd find it perfectly fine and even love its character, but, since playing with simulations is fun (and free!), shortly I'll look into possible mods to iron out some of those quirks.
And now for something completely different:
Here we have a dual TL062 opamp: the first stage is a second-order, Sallen-Key high-pass filter with an F3 of ~57 Hz and the second stage is a standard 3-band Baxandall arrangement. Also at the input there's that 2n2 cap to ground as the pickup load. This is the resulting frequency response with all controls flat:
The sharp ~16 dB (!) peak at around 2.8 kHz is due to the 2n2 cap and the fact that the resistance to ground at high frequencies is just ~560k from the filter resistor. If you have / plan to build this preamp and want to adjust that peak, it's easy: add a resistor in parallel with the 2n2 cap to lower the peak height (e.g. a 68k will bring it down to ~6 dB) and lower/increase the capacitor to increase/lower its frequency, respectively (e.g. a 1n cap will move the peak to ~3.9 kHz). If you remove it altogether, the peak moves to ~8.3 kHz, the intrinsic resonant frequency of the pickup due to its inductance and parasitic capacitance.
As for the high-pass filter, attenuation at 41 Hz (fundamental of the low E string in a 4-string bass) is ~8 dB, which sounds like a lot to me... Your guess is as good as mine as to why they thought this was necessary and/or desireable, maybe to prevent excessive cone excursion and get a tighter sound with smaller combos? Anyway, you can easily lower the F3 by increasing the 8n2 caps, e.g. 12n will give you F3 = ~37 Hz and a more reasonable attenuation of ~1.5 dB at 41 Hz. You could also simply remove the filter altogether and leave the first stage as a buffer, with pickup loading adjusted to taste.
The 3-band Baxandall implemented with a single opamp stage is well known to have an intrinsic problem: there's a lot of interaction between the bands. If we plot the response stepping the three controls in the same direction, this is what we get (sim file attached):
Not a lot to see here. It helps to fix the mids control (make Rc=25k and Rd=25k) to see better what the bass and treble do:
And then the opposite, fix bass and treble and step mids (swap Ra for Rc and Rb for Rd in the parameters list):
For comparison with previous preamps, the boost/cut at 40 Hz is around +/-14 dB, at 10 kHz around +/-18 dB and the mids, centered at ~440 Hz with very low Q, around +/-7 dB.
As you can see, even with the mids fixed, there's significant interaction between the bass and treble, and the mids looks a bit limited in range. So, all in all it's a "quirky" preamp... Of course I've never played with it and maybe if I did I'd find it perfectly fine and even love its character, but, since playing with simulations is fun (and free!), shortly I'll look into possible mods to iron out some of those quirks.