Practical Wiring

[alder]

Signal can't really "muscle through", but the way the control functions would definitely be affected noticeably by the impedance of your input.

What I meant is, if you hook up even the 1 Meg pot purely as a fixed inline resistor, plenty of signal gets through, such that if you didn't know it was there, you wouldn't think anything was amiss. The 500k pot has even less effect. If that much unfiltered signal can push its way right through the pot, then the filtered signal going through the cap is going to get lost. I suppose the test for that would be to place the volume upstream of the bass control, and back it down significantly.

This suggests a possible fix of placing a fixed volume control directly on the pickup output to drop it down to something reasonable. I could work out the value using a pot, and then install a fixed resistor to ground on the output of the volume control. The output of this particular instrument is very high, bordering on excessive, anyway. Now that I think about it, I think this is the answer. I'll let you know.

It's not the headphone amp, it has behaved perfectly for years on every instrument, and continues to.

 

[sunbeast]

What I meant is, if you hook up even the 1 Meg pot purely as a fixed inline resistor, plenty of signal gets through, such that if you didn't know it was there, you wouldn't think anything was amiss. The 500k pot has even less effect. If that much unfiltered signal can push its way right through the pot, then the filtered signal going through the cap is going to get lost. I suppose the test for that would be to place the volume upstream of the bass control, and back it down significantly.

This suggests a possible fix of placing a fixed volume control directly on the pickup output to drop it down to something reasonable. I could work out the value using a pot, and then install a fixed resistor to ground on the output of the volume control. The output of this particular instrument is very high, bordering on excessive, anyway. Now that I think about it, I think this is the answer. I'll let you know.

It's not the headphone amp, it has behaved perfectly for years on every instrument, and continues to.

That definitely hasn't been my experience! Possibly a limitation of your headphone response vs being in a room with a speaker and "feeling" the lost sub frequencies? With my Ric I sometimes solo the bridge with the bass-cut fully engaged (with a 500K pot) and it sounds like a guitar!

 

[alder]

Test Rig

This is a little test stand I built, along with some alligator clip jumpers from Radio Shack. It is a piece of aluminum angle attached to a piece of wood for stability. Different sized holes will take pots, switches and jacks, including even a blade switch. A 1/4" jack is installed permanently at each end. The aluminum angle automatically grounds everything.

In the picture, the input is at the right and the output is at the left, but this is entirely reversible. The red jumper is just connecting the hot leads. You can use anything for a signal source, I'm using a bass at the moment. Using the jumpers, it is possible to quickly assemble all sorts of configurations for testing, without soldering. You can also solder permanent assemblies and then remove them. One drawback is that any noise sources in the vicinity, such as fluorescent lights, are picked up very strongly.

 

[alder]

The problem with the bass control on the pico is exactly what I suspected - leakage across the pot. This is confirmed by removing the pot and using the cap by itself, the old Ric bridge pickup circuit. This works fine for a range of capacitor values. The pot is essentially a resistor in parallel with the capacitor, and even a shut-down 1 Meg pot is just not enough resistance to make the circuit effective at bass frequencies with high-impedance high-output bass pickups. The obvious solution is to increase the resistance of the pot, say, to infinity.

So I went into the parts box, and got out a 4-way 3-pole rotary switch and a pair of capacitors. These can be wired up much like pickups in series, single, or parallel. This gives the following selections ( using 'X' uf caps) :

1. no capacitor - straight pass-through
2. parallel capacitors - 2 X - greatest capacitance, least effect
3. single capacitor - X
4. series capacitors - 1/2 X - least capacitance, greatest effect

Capacitors obey the opposite rules of combination as resistors: ie parallel caps are additive. Of course, capacitors like this are cheap, you could easily get the same effect with three different capacitors, and the wiring would be simpler. I just like the elegance of getting three settings from 2 pieces.

Position 2 gives a very subtle effect. Position 3 is more pronounced, and position 4 is the old Ric 'strangled pickup' sound - very pronounced. I did a good deal of experimenting using my test rig above, and determined that X = 0.010uf caps gave the best range of results. Smaller caps are to strong, and bigger caps are too weak. So there you have it, a new kind of tone control that may very well be an original invention of mine.

So now I have a very effective rotary control with a range of settings that fits right under a Strat knob. Almost. The switch is bulkier than anything that was ever intended to be installed in a Strat, and the cavity had to be deepened and widened to fit.

The only thing that remains to make the Pico bass perfect is to tame the excessive output level. For that I tried several different resistors between hot and ground, essentially introducing a fixed volume control in addition to the variable one. I installed a tiny 1 Meg pot on top of the main volume control so I can tweak it all I want. About 800k did the trick - a very small adjustment. The excessive output level is clearly related to the vibrating mass of the extra-heavy strings. I suppose I could also lower the pickups, but they are already pretty far down. This bass also has a series pickups option which makes the excessive output even worse.

A final non-electronic mod, I sawed off the extra overhanging fret. That leaves the neck heel a little funny-looking, but that fret was stifling the ones behind it, and the wood under it cracked when I tried to level it. I wanted the extra playing space anyway.

 

[jcanderson]

Hi Alder,

I finally took a look at this thread again after posting on it back in November.

The bass-cut control is very dependent upon the load that it sees (i.e. amp input impedance, etc). If the bass-cut control sees a really high impedance, it will not work well. One of the important factors is the ratio of the bass-cut pot resistance to the load resistance that it sees. The larger this ratio is (pot to load), the greater will be bass-signal attenuation.

Let's take a look at the G&L circuit (reference the schematic I posted on the first page of this thread). Note that their volume pot is 250K. This is the load that the bass-cut pot sees (actually, it's this in parallel with the impedance any external devices (such as amplifier input impedance), but for purposes of discussion, let's limit it to the volume pot, itself).

Now, consider what happens when the bass signal frequencies are very low. In this case, the capacitor represents a very high impedance. Let's assume the that bass-cut pot is 1M (G&L value), and that the frequency is so low that the capacitor's impedance is much greater than 1M. The capacitor can now be thought of as being effectively "out of circuit", and this will be when we see the maximum amount of bass attenuation. (Note that as frequency goes up, capacitor impedance drops, and it will act as a shunt across the 1M pot, allowing more and more signal through as frequency increases).

So what is the maximum amount of bass attenuation that you should expect to see, best case?

Assuming that the frequency is low enough that the cap impedance is very high compared to the bass-cut pot value, then the best-case attenuation, A, can be calculated as follows:

A = Rvolume / (1Meg + Rvolume).

So, let's say that the volume pot is 250K (the value on the G&L schematic). The maximum bass attenuation should be 1/5 of the original signal, or about -14 dB.

But suppose there's no volume pot, and instead you're driving against the input impedance of your amp. If this input impedance is 1M, you'll only see max attenuation of 6 dB (i.e. a ratio of 1/2). And if the input impedance is higher, you might not see any appreciable attenuation.

So -- if you're seeing poor performance with the G&L circuit, the first thing I would wonder is, are you feeding the bass-cut circuit to a volume pot, and if so, what's the value of that pot? Or are you driving it directly into an amp?

(Also, the cap also has an effect, too. As the cap value is increased, you'll be allowing more and more low frequencies to get "past" the 1M bass-cut pot. So, as the cap value is increased, the less effect the bass-cut control will have.)

 

[alder]

Here are some pics of my latest project. This is a 3-humbucker passive setup for a guitar, but I have used a very similar design for a bass. The pickups are 4-wire humbuckers, 5-wire if you count the ground, which has to be separated on the middle pickup. The pickups connect to a rotary mode switch, which is the middle knob. The mode switch changes all three pickups simultaneously between series, parallel, and single-coil modes. In addition, the middle pickup is wired so in single-coil mode, it is humbucking in combination with either of the other two. As detailed earlier, the mode switch is actually eight switches ganged together, of which only six are used.

The three outputs of the mode switch go to a standard Strat 5-way parallel pickup selector switch. The output of the pickup selector goes to a double-acting tone control on the third knob. This is a dual 500k pot with a center detente. Rolled forward, it acts as a standard tone control - treble roll-off. Rolled back, it acts as a bass roll-off. You can't do both at once, and you wouldn't want to. Finally, the volume control is a 500k audio pot with a treble-bleed circuit.

With the 5-way pickup selector and the three-way mode selector, that makes 15 different possible tones, all but three of which are humbucking. The overall impedance varies from about 11k for a single pickup in series mode down to about 1.4k for 2 pickups in parallel in parallel mode. Impedance that low might compromise the tone control, I can't tell from my testing so far. That's a whole lot of action for what is outwardly a standard Strat control set. I could add a push-pull switch on the volume, perhaps for a passive distortion circuit, but for now, enough is enough.

The pickup wires are a bit of a mess, because I hate to cut them. This way, I can tear the whole thing apart and re-use all the pieces. Also, if the pickup wires were different lengths, the signals would arrive at different times, which would cause phasing problems that I am sure I would be able to hear, because I am super-special, not like everyone else. File that one away with tone-woods and thin lacquer finishes.

The last picture is a closeup of the electronics. It's not hard to see why no one tries to produce these in a factory. I took me several hours to put the whole thing together, using the diagrams that I have uploaded to this board. I made no mistakes, which is nice; everything tests out good. The mode switch is a bit too fat to fit into the body cavity, so it is back to the wood-shop for a bit of routing now. On a Strat, I typically connect the bridge/body ground to the output jack, not the pickguard. That way there are only two connections to break if I ever want to remove the electronics. Also notice that the red/black twisted output wire is nice and long, allowing the pickguard to be removed and flipped over for servicing, which is not unheard-of with wiring as complex as this.

You might think that this project is geared toward the high-end of the frequency range, with 500k pots, treble bleed, bass roll-off, and thinned-out options for the humbuckers. You would be right. This is actually going to be a twelve-string, and I want that sixties jangle. As it is a guitar, though, I don't think it would be kosher to post the final result here on Talk-BASS.

The pickguard is pretty nice for a cheap import, not all misshapen like some I've seen, even most of the screw holes lined up. The pickups are my old favorite Belcat cheapies, I think I paid about six bucks each. This will be the first time I actually try them in a guitar. All of it is all out of the parts box, I didn't buy anything specifically for this project. Nice not to go out-of-pocket AT ALL for a change. A good little project for an afternoon that is just too d@#%! HOT to go outside, I half-killed myself on yard-work yesterday, and need a day of rest.

3 pots, 7 switches, 6 coils, 15 leads, 3 caps, 1 resistor, too many solder joints to count

 

[alder]

Batteries and Battery Boxes

I only use clip-less battery boxes for two reasons. First, those 9-volt battery clips may be tolerable for a smoke detector, but if you plan to use it a lot, it is a pain in the ass, and then it breaks. And second, 9-volt batteries ( all batteries, really, but especially 9-volts ) are guaranteed to leak, and should never be left inside a device unless absolutely necessary. I only put a battery in a guitar ( or anything else ) when I want to use it, and then take it out afterwards and keep it in a little dish. That's much easier if it just slips in and out and you don't have to mess with one of those stupid clips. And if when it leaks, I just throw it away and wash out the dish.

This is how I handle all alkaline batteries now - I've learned my lesson, I've seen more than enough expensive devices ruined by alkaline batteries. In fact, the problem has actually gotten worse over the years, not better. This applies to all brands, but Duracells are the worst. One day I went through everything I have and pulled all the Duracells, regardless of age or use, and threw them all away. Things that need batteries, like emergency lights, I installed lithiums, everything else I left empty. This leakage issue applies mainly to alkaline batteries, I have never seen a lithium cell damage anything. They may be expensive, but it is a good investment.

9-volts are especially prone to leaking because what looks like one square battery is actually six round batteries in a square case. That's six batteries that can leak, six times the likelihood, basically guaranteed to happen within a year. Coincidentally, that is how long most guitar electronics claim a battery will last. Don't leave the battery installed, and you'll never have to regret it.

Gotoh battery boxes are the best I've used, but are rather expensive. Equally functional boxes may be got from China on eBay for just a couple of dollars each.

Believe it !!!​

I just pulled a dead lithium rechargeable from an old iPod. It had swelled to twice its normal size, but did not leak or damage the device. I replaced the battery, and I am listening to it right now.

 

[alder]

It's been a long Spring ...

... but I finally got back around to guitars, and built this piezo buffer circuit for the Stratbory. And it works great !!! The switching pot allows for a dead battery mode, and there is a big difference in bass response for the peizos. The board is half of a 4x6 cm, so it will fit easily in any control cavity.

The circuit diagram may be found here: http://scotthelmke.com/Mint-box-buffer.html

And this is what it is all for:
https://www.talkbass.com/threads/cu...-factory-guitars.1211610/page-6#post-18702389

Stratbory MK II​

 

[alder]

DIY PIEZOS

In this application, I am using rather expensive GraphTech Ghost piezo saddles. These have very high output, and absolutely require high-impedance input buffering or they sound terrible. I built a working piezo impedance buffer, there in the vise jaws. It is small, inexpensive, quiet and sounds good. Piezos require a high impedance input, otherwise they sound very shrill and tinny - all finger noise and no bass. ( Actually, you can see one of these in the previous post, but it has been so long, I forgot all about it. )

This circuit has fully adjustable gain ( black pot ) and trim ( blue pot ), so that I can balance the piezo with the magnetic pickup. The output of the buffer goes to the pickup switch, where it can be treated like any other pickup, which was one of my main goals. The buffer runs on a 9v battery, the rest of this guitar is passive.

That is actually a 4x6 cm circuit board that I cut in half, about the size of a large postage stamp. I found this circuit online, I didn't design it. My contribution is the variable gain in place of the original resistor network, a diode on the power line for polarity protection, and an onboard 500k trim pot on the output - all simple stuff, I'm an ME, not an EE.

This wiring diagram is drawn to match the physical layout of the components on the perf board in my final design ( not the prototype above. ) I used antenna symbols to indicate the io pads on the perf board at the bottom of the diagram.

I reverted to using a switch for the gain rather than the expensive and hard-to-find 10M pot; the wiring for the gain pot is included as an alternative. The output trimmer could be 500k or 1meg, whatever you have, or omitted. The 2N5457 transistor is actually reversible, so as long as you get the middle "S" leg right, the other two don't matter. Pay attention to the polarity of the tantalum capacitor and the diode.

I have also included a bare minimum version of the circuit. This has fixed gain and output, and no noise filtering or power buffering, but if you are still developing your soldering skills, this circuit is far simpler, and there is probably no audible difference between this and the full circuit. You can also pick and choose pieces from both diagrams.

This buffer is very quiet when powered-down, although a tiny bit from the piezo does leak through, such a small amount that it would be completely drowned-out by a magnetic pickup. I think it should be acceptable to turn off the piezo by simply powering-down the buffer, without actually disconnecting the output. That opens a lot of switching possibilities, one of which is included on the diagram. This uses a common DPDT On-On-On switch to select between mag and piezo pickups, and also power-down the buffer when piezo is not selected, effectively shutting off the piezo pickup. ( You could also switch the piezo output directly, but that would leave the battery drain on all the time. )

Thus with the addition of one mini switch to the controls, you can add piezo capability to any guitar. Well, electrically anyway, the actual physical installation of the sensor, circuit board, and battery is a bit more complicated. Also, you'll want to adjust the trim on the buffer to match the existing pickups. This would be the absolute minimum visible modification to a guitar. You can find a place for a mini switch somewhere on any guitar. I like that when the piezo is off - a flick of a switch - the guitar reverts to normal. A flick the other way, and it is pure piezo. If you're willing to make more modifications, you can install a blend pot between the piezo and mags, and a simple on/off mini switch for the buffer.

Here I am testing serial number 000001. You can see the inexpensive disc piezo sensor and my high-tech mounting system. Contact me if you would like to purchase one of these "Special Luthier Tone Clips", exactingly made from exotic tone wood and high tensile steel in Nashville Tennessee. For $49.95, I'll send you not just one, but a second one free! ( Just pay a small extra charge for shipping and handling. ) You can also see my 1/4" jack to alligator clips breakout doohickey, which is very useful for this sort of testing.

I figured out how to trick the camera into macro mode, so here are some close-ups ( please, don't critique my soldering. The bigger the glob, the better the job, as they say. )

The insulated jumper connects the two grounds, crossing over the output. There's no avoiding something having to cross over something else, and it works out well to have two ground locations between the other connectors.

This is a detail of the connection pads. Using excess lead wire, I first solder it to the back, then bend it around and solder it to the front. This way, I don't have to worry about the internal leads coming loose when I attach the external wires - they are mechanically fixed in place and soldered on both sides. From the left, the pads are:

• input
• not used
• ground
• ground
• ground
• not used
• output
• not used
• ground
• ground
• not used
• power

The grounding areas are extra large to accept multiple ground wires anywhere, and all different connections are separated by unused pads for spacing.

This is getting closer to the final design. The red switch is the wrong kind, I replaced it with a computer jumper and had to do some rewiring on the back. Also, this circuit is incomplete - the power protection diode is missing, and the battery lead is not connected. The circuit can be divided into three sections:

• power on the left
• output in the center
• input on the right

This corresponds to the three legs of the transistor, D-S-G, which is at the bottom. A majority of the components connect to ground, so figuring out a good layout for that was really the most important part of the physical design. I gave myself ample places for grounding, both internal and external connections.

The wires are ( l to r ):

• battery (+) (red)
• battery (-) (black)
• output (+) (white/yellow)
• output (-) (black)
• input (-) (red) (piezo)
• input (+) (red) (piezo)

The piezo has two interchangeable red leads.

As I build these, I am attaching battery clips, piezo sensors, and output wires for testing. Then I am leaving them as reminders which contacts are which; makes a complete kit.

I used the length of the components to span distances where I needed to, and used the excess lead length wherever I could, for example, to create the two-sided connecting pads. I saved the longer bits of cut-off leads to make random connections and jumpers, now I have a film canister full of them. Where it gets crowded, don't forget that you can stand things on end, as I did with the one resistor. I deliberately mounted the big capacitors near the transistor to protect it from getting bumped or bent, as it is pretty fragile sticking up off the board. It would be a good idea to goop the whole assembly with silicone for protection, I don't think epoxy is really necessary.

Use a small alligator clip as a heatsink when soldering the transistor, diode, and small capacitors, as they are easily ruined by heat. You can see that I left the leads on those parts uncut, and used the full length to keep the heat away from the semiconductor. If you're not an experienced electronics tech, these are things you would have to figure out the hard way.

You can get all the parts for this on eBay. Specific part numbers and values are on the diagram. Here is a list of the tools:

• 15 watt soldering iron & stand - Radio Shack
• bench-top mini vise - Harbor Freight
• needle-nose pliers
• mini wire cutter
• tweezers - drug store
• small alligator clip ( for heatsink )
• +2.00 reading glasses - Dollar Store
• hobby saw ( to cut circuit boards in half )

For the moment, I have run out of diodes and boards, so I can't assemble any more. Now I miss being able to hop on down to Radio Shack. Not that they were all that good for parts at the end. They never really stood a chance against "online," but now I have to order everything and wait. When I get the parts, I will build one more to the final design and post it here.

My ultimate goal is to develop a piezo system with the following properties:

• sounds good ( obviously )
• can be used alone or alongside regular pickups
• can be retrofitted into existing guitars with a minimum of modifications
• can be built from readily available components with simple tools
• costs under $5, including the battery box.

I think I am pretty much there. I temporarily pinned one of the button piezos under the pickguard of a six-string, and it sounded great. These button piezos seem to be very forgiving in terms of installation.

When I first started this project, I had never built anything like this before. I was basically just soldering together a circuit that I had found online, with no good idea how to lay it out. I was amazed when the first one worked, but the board was a mess - jumpers all over the place. Over the next few assemblies, I refined the design considerably, and even made my own additions.

Thanks to Scott Helmke and Walter Harvey for assistance with the buffr circuit.

 

[alder]

DIY Piezo Done Right !!! ( and cheap !!! )

see previous post

This was going to be a fairly typical octave-bass ( otherwise known as a guitar ) but since I now have all this great piezo technology, I decided to add it here. I took one of the button piezos, roughed-up the plastic back with 80-grit, slathered it in Crazy Glue Gel, and stuck it inside the top between the pickup cutouts. You can see the wires hanging out in the control cavity. This will be my first test using the cheap eBay piezo element rather than some expensive guitar piezo.

I'll build this up initially as a pure piezo and see how it sounds. Maybe the sensor should go somewhere else - I could fit one to the front of the bridge block, or carve out a spot in the neck pocket, but this was easy, and I figure it should sound the most 'acoustic'. Or maybe it sounds bad, in which case I will simply abandon the piezo and build it up as I had originally planned. Assuming the piezo is worthwhile, I will add it to the guitar wiring as outlined above, with a single switch. I'll also need to cut in a battery box.

BTW, the Jazzmaster bass in the previous post is all together and sounds good. Details here. The GraphTech saddles have a very high output. Without a buffer, the finger noise was unbearable and there was no low end, but with the buffer they sound good, with the bass restored. My $1 DIY buffer sounds no different than the $30 EMG buffer, and has the added advantage that it has not burned up ( yet? )

The piezos in the Uke bass also sound good, but that is so different from anything else that you cannot really use it for a general evaluation.

 

[alder]

A/B Comparison

Perhaps because I am crazy, I decided to add one of the cheap piezo buttons to the Jazzmaster. So I pulled the bridge and used the Dremel to gingerly rout out a shallow cavity for it. I trimmed the flanges of the sensor down to a minimal size and added extension wires. Once I cut through the veneer, I was surprised to find that this is not actually plywood - the core has no wood grain, it must be some kind of fiberboard.

Here is the button installed, using Crazy Glue Gel. This stuff bonds wood almost as fast as it bonds fingers, but the fillet I added around the edge will have to dry overnight. For good measure, I added a puddle of polyurethane on top of the button, because, why not? This will make it polyphonic.

Note that the sensor is not 'pinned' under the bridge, it is firmly bonded to the body, and does not touch the bridge. The alternative would have been to go through the sound hole and glue the button inside the top, but this seems to be the preferred method of installation for a bass.

The new wires go down the same big hole as the rest. I changed the tone control for a push-pull -- down for the saddles, up for the button. The rat's nest inside is now that much more. Good thing it's a hollowbody.

Now I can compare the $100, and ostensibly magnificent, GraphTech piezo pickups, with a 50 cent button from China. I can actually hold a note and work the switch for some true A/B comparisons. Imagine that - scientifically testing 'tone.' None of the "chrome strap buttons sound brighter" nonsense, this is actually valid testing with measurable results. Don't tell the luthiers union - they will be casting spells and hexes against me, I know it.

I have a feeling I'm going to get pissed off -- either I wasted a hundred bucks, or fifty cents.

 

[alder]

The Rat's Nest Grows

After reassembling the bridge, I found there is a huge difference in output between the button and the saddles. So I took it apart, and added a shim between the bridge and the piezo for better coupling. I actually used a penny wrapped in tape, it was about the right size and thickness. That improved matters somewhat, but the button was still lagging behind the saddles in both overall output and bass response. It is impossible to compare the quality of two sound sources if they are not roughly the same loudness - the louder one will always sound better - that is just the nature of psycho-acoustics.

I need to dial back the saddles, while maintaining the full output of the button. The only way to do that is to install a second buffer, as you can see above. Fortunately, these cost next to nothing to build, so I don't mind, and right now I have several spares. The trim pot on the second buffer broke, so I removed it and sent the output straight out. For the moment, I am forgetting the magnetic pickup, and just dealing with the piezos. I had to rearrange the wiring on the tone pot switch, and while I was at it, I set it up so that only the buffer in use gets power. No sense powering both at once - I am not interested in what the saddles sound like in combination with the button, just one or the other.

And ... the saddles are still winning, by a good margin. Not only that, but the button still has very anemic bass response. At least it is not picking up a lot of spurious noise from the body, but it is also not picking up the strings very well. In fact, far from being the best spot for the piezo, under the bridge may be the worst. Of course, on a solid body, you don't have many other choices. But this is a hollow body, I can install piezos anywhere I want.

So I installed a second piezo under the top between the bridge and the sound hole. I figure that location won't get thumped a lot by fingers, but still have some resonance. And I was right, it sounds better than the bridge. But still losing out to the GraphTechs. Then I wired both buttons in parallel, combining their output. At first, I had them out of phase, and they cancelled each other perfectly until I reversed one. This brought the volume up closer to the saddles, and I tweaked the saddle buffer to match the buttons, which are at max.

I'm not going to install any more buttons in this guitar. More sensors would raise the volume incrementally, but I don't think it would improve the tone. Finally, I addressed the volume difference between the piezos and the humbucker by adding a trim control to the pickup mode switch and dialing it back to match the piezos.

I think I've learned a good deal:

• the $100 GraphTech Ghost saddles are clearly better than the 50 cent buttons, in every way
• the GraphTechs are brighter and lighter sounding than the humbucker, but with good bass response
• the buttons are sorely lacking in bass response, and are probably best suited for six-strings, not basses
• the buttons are also lacking in output, and I would install them in pairs to compensate
• under the bridge is clearly not the best place for a piezo, probably not even a good place

This has been my first real experiment with DIY piezos. I'd say the buffer is a raging success, while the button piezo sensors are a disappointment. You get what you pay for with the GraphTechs. I'm going to look for some other piezo sensors, maybe something else would work better.

This guitar has a nice unplugged tone, the humbucker sounds awesome in either series or parallel mode, and the Ghost saddles also sound good with the tone backed just a bit. It has become a monster under the hood, but still shows a classic Jazz Bass set of controls on the outside. Although the experimental piezos are lacking, there is no harm in leaving them.

 

[alder]

I've been doing some reading, and gotten some ideas, in particular, something I would call "reaction mass". Piezo elements generate electric current based on the strain they are subjected to. Take, for example, the piezo under the top of the Jazzmaster. The strain on that element is basically caused by its own inertia as it is moved by the vibrations of the underlying wood, and whatever bending stresses may be occurring. As the element itself is very light, the strain is going to be small. If a mass is attached to the piezo, that will increase the strain, as the piezo is pushed and pulled between the underlying wood and the mass it is carrying.

So, after further misadventures with Crazy Glue and fingers, I managed to attach two quarters to that piezo, and sure enough, there is a slight but noticeable increase in output and low end response. Then I re-removed the bridge, and dug out the piezo there that I previously installed. I took another piezo and glued two quarters to the top of it. ( This is getting expensive! I will make some lead discs for the future. ) That is going to take a while to set. I will then enlarge the cavity under the bridge, and glue the piezo to the bottom of the bridge plate, so the weight is suspended from the piezo inside the cavity, and the whole assembly is free to vibrate.

The result of this will hopefully be much better low end response and overall loudness from that sensor, at the expense of some of the acoustic nature of the piezo.

 

[Coolhandjjl]

Well, there's about 3 hours I won't get back....

 

[alder]

Here is round two. You can see, I've cut right through the top, and made a generous cavity for the piezo assembly. You can see the new piezo glued to the bottom of the bridge plate, and two quarters glued to the bottom of the piezo. This Musicman-style bridge has an extra-large bridge plate - perfect for this kind of experimenting. When I reassembled it, nothing showed.

And the result is: some improvement, not much. The body-mounted unit sounds better than the bridge-mounted one. I wired them together for greater output. I think these piezos are just not suited for a bass.

 

[alder]

Here's another go at it. I have taken a button piezo, cut away the plastic ring, and crimped the edges of the brass backing plate into a bottlecap shape. I worked the edge on a piece of sandpaper until it was flat and even. A quick test showed that I did not kill it.

Here is the bottlecap glued by the edges only to the bridge plate, after removing the previous experiment. The ceramic element is free to flex. In the body, I added a small piece of wood to the bridge block, and set a flat head screw into it. The screw impinges on the piezo roughly at the center. A piece of tape protects the ceramic from the metal. The screw should drive the piezo like a piston.

The immediate result is much higher output from the piezo, but no great change in tone. Bass response is improved, but still lacking compared to the saddles. This setup is sensitive to the amount of force of the screw on the piezo. If the screw is backed away from the piezo, it becomes very tinny, with increased finger noise.

At this point, my only conclusion is that these piezo buttons are not suitable for bass use, although they sound good in a guitar. My next experiment will be to build some kind of simple equalization into the buffer to compensate for the poor tone of the buttons.

 

[Matt Liebenau]

View attachment 2800448

Here's another go at it. I have taken a button piezo, cut away the plastic ring, and crimped the edges of the brass backing plate into a bottlecap shape. I worked the edge on a piece of sandpaper until it was flat and even. A quick test showed that I did not kill it.

View attachment 2800449

Here is the bottlecap glued by the edges only to the bridge plate, after removing the previous experiment. The ceramic element is free to flex. In the body, I added a small piece of wood to the bridge block, and set a flat head screw into it. The screw impinges on the piezo roughly at the center. A piece of tape protects the ceramic from the metal. The screw should drive the piezo like a piston.

The immediate result is much higher output from the piezo, but no change in tone. Bass response is improved, but still lacking compared to the saddles. This setup is sensitive to the amount of force of the screw on the piezo. If the screw is backed away from the piezo, it becomes very tinny, with increased finger noise.

At this point, my only conclusion is that these piezo buttons are not suitable for bass use, although they sound good in a guitar. My next experiment will be to build some kind of simple equalization into the buffer to compensate for the poor tone of the buttons.

I did take apart some Radio Shack piezo buzzers a long time ago and put the elements under the bridge feet on an electric upright bass. I think I wired them in series but it's been a while ago. Like you said, there was a lot of output but also a lot of string noise and not as much low end as I wanted. At the time, I figured it needed some type of preamp or buffer but never got around to building one because I had magnetic pickups on the bass also.

Good luck on the experiments. It's been interesting reading.