Acoustic on-board pre-amp mods????

[treguiers]

There seems to be tons of info about electric guitar mods, and mods on various pedals. But has anyone attempted anything interesting with acoustic pre-amps.

I have a cheap tanglewood with a under saddle transducer and a fishman prefix pre-amp. Can I get the amplified sound to sound like a top of the range Martin, or Taylor, or like a vulcan laird!(I am very aware of the silk purse and the sows ear analogy)

Has anyone messed around with acoustic pre-amps and come up with anything interesting. A resistor change here a cap change there. That sort of thing. Maybe there's tons of stuff, and i don't know where it is.

Again any nuggets of wisdom will be warmly welcomed

[newey]

Around here, we call JohnH "Mr. Nuggets of Wisdom". ;D

And his article on Adding a Piezo Pickup has a good deal of info on his own schematic for a preamp/buffer circuit. Now, this is for a Strat, and used to blend the mag pickups with the piezo, so it's not directly applicable to what you're after, but you may glean a few useful tidbits therein.

Required reading, if you haven't already done so.

As to wanting your Tanglewood to sound like a Martin, et al., well, you can't turn a sow's ear into a silk purse. But you can make a better pig.

[ashcatlt]

For the most part, all we've got available in analog technology is going to be filtering and compression (or saturation). That's not to say I don't think we can use these techniques to some advantage. I'm definitely interested to see what might be done with this.

The thing I worry about, when you talk about "modding" the on-board preamp, is that these things are usually tiny little circuits on tiny PCBs using those tiny SMDs, built generally by machines. A simple change of a resistor or capacitor here will require some at least intermediate soldering skills, and the patience of a saint. If you're willing to mostly just scrap it, or find a way to tack your circuit on before, after, or sometimes during, you've got all kinds of options.

Honestly any preamp, compression, OD/dist, filter, etc circuit you can imagine (or google) could be inserted here, but I think your in somewhat uncharted territory. The thread newey linked is pretty thorough, and the info re: the piezo preamp and it's attached filter can be extracted without much change at all. I don't think you'll find a whole lot more in the way of real experience with this kind of thing. We can definitely help if you can identify a little better the goal.

By "we", I mean the collective, not necessarily including me. I mean to encourage you to forge this new trail and report back. I will be watching this space.

[sumgai]

tre;

It's possible to muck up this discussion with a lot of electronics mumbo-jumbo, but that needn't be so. Cutting to the chase, you need only one word here: equalization.

Each guitar has a frequency response curve, which can be called its signature. Another word for it is, of course, is its tone. It's that tone, or signature, that you want to emulate.

In the analog world (we can do this more easily in digital, where it's called modelling), we can insert as many filters as we wish, until controlling all those variables becomes untenable. By adding one filter at a time, you might reach your goal in only one or two tries (meaning only one or two filter circuits), or you might need to spend a week on it and end up with a 10-band filter box. You just never know, until after it's all over but the shouting.

But if you stick to it, you'll get it sooner or later.

BTW, if you're out to emulate another guitar (in the purest sense of the word), then distortion of any sort, including compression, is not your goal. But tone-shaping circuits like these (consider chorus, reverb, compression, etc.) are valid, and may enhance your tone such that it surpasses your original target guitar. I humbly suggest that you not get the cart before the horse.

And finally, I fully endorse ash's suggestion - remove both your piezo pup and the original circuit board. (You can leave the battery box.) Build and install your own rig, along with a cheap (but good) piezo pup, and you won't have to worry about ruining anything.

HTH



sumgai

[treguiers]

Fascinating! Maybe the pre-amp from scratch is the way to go. The link to JohnH is very interesting. I'll have to brush up a lot on my electronics theory.

What started me thinking about it was the difference adding an acoustic exciter made to the sound of the guitar, but to take sumgai's point, that's just a good EQ box.

So definitely food for thought, meanwhile I had better get studying.

[newey]

JohnH's statements about blending the mag pickup sounds with the piezo on his Strat started me thinking. And a while ago, I posted that Tommy Emmanual vid in the coffee shop. In an interview with him, he was discussing his stage rig. He uses both a piezo under-saddle transducer as well as a condenser mic in the guitar body near the soundhole. He implied that these 2 signals were in some fashion blended, although the impression I got was that it was done outboard of the guitar.

In any event, I wonder if using multiple pickups on an acoustic, each run through an appropriate EQ stage and blended, might not give a fuller sound than just the piezo alone. After all, we put multiple pickups on electrics, why not on acoustics?

And there are multiple options to choose from: Bridge-mounted piezo, in-body contact mic, condenser mic, or magnetic pickup, either soundhole mounted or permanently mounted to the body. Or multiples of the same types- a piezo on the bridge and another up at the end of the fingerboard, perhaps.

Just thinking aloud here, but some nutzy experimentation may be in order.

[D2o]

FWIW, an article - "Wendler Instruments Profile Kansas City luthier Dave Wendler specializes in electric guitars with an acoustic voice", in Acoustic Guitar Magazine, may also be of interest.

Cheers,
D2o

It reads, in part:

Dual Pickups

[/b]
Electronically, one of the unique aspects of the electroCoustic is the way Wendler has designed the bridge-mounted pickup. The guitars have a one-inch-tall archtop-style bridge made from quarter-sawn maple (combined with a tailpiece to hold the strings), into which he sandwiches a piezo disc. The reasoning behind the bridge design is twofold, he says. “First, being sandwiched into the maple bridge allows the piezo to have the same resonant characteristics of the bridge itself. Second, it allows as much signal generation from the body as directly from the strings.” He adds, “Piezos have a huge dynamic range that you have to damp down a little bit to get an even response during the attack transient. When you hit the string, the piezo wants to let all the high frequencies come through before it lets the bass frequencies come through. So the magnetic pickup in combination with the piezo slows that down so you get a nice even attack transient. That gives you a natural feel and response very similar to the way an acoustic guitar responds under your hand.”[/quote]

[ashcatlt]

Nov 12, 2009 6:29:30 GMT -5 treguiers said:

What started me thinking about it was the difference adding an acoustic exciter made to the sound of the guitar, but to take sumgai's point, that's just a good EQ box.

Actually, these things are often frequency dependent distortion boxes.

“Piezos have a huge dynamic range that you have to damp down a little bit to get an even response during the attack transient. When you hit the string, the piezo wants to let all the high frequencies come through before it lets the bass frequencies come through. So the magnetic pickup in combination with the piezo slows that down so you get a nice even attack transient. That gives you a natural feel and response very similar to the way an acoustic guitar responds under your hand.”

And this sounds like a job for compression, possibly also frequency dependent.

[sumgai]

Sorry, ash, can't quite let these slide by without some fine-tuning....

Nov 12, 2009 11:44:23 GMT -5 ashcatlt said:

Nov 12, 2009 6:29:30 GMT -5 treguiers said:

What started me thinking about it was the difference adding an acoustic exciter made to the sound of the guitar, but to take sumgai's point, that's just a good EQ box.

Actually, these things are often frequency dependent distortion boxes.

In reality (as opposed to ackshully), these are time-domain comb filters. Discussing these things is definitely not a job for the weak-kneed. Suffice it to say, they work almost like an EQ setup, but there's more under the hood than you'd think. Sorta like a NOS setup in a street racer, hehehehehe.

Nov 12, 2009 11:44:23 GMT -5 ashcatlt said:

“Piezos have a huge dynamic range that you have to damp down a little bit to get an even response during the attack transient. When you hit the string, the piezo wants to let all the high frequencies come through before it lets the bass frequencies come through. So the magnetic pickup in combination with the piezo slows that down so you get a nice even attack transient. That gives you a natural feel and response very similar to the way an acoustic guitar responds under your hand.”

And this sounds like a job for compression, possibly also frequency dependent.

Nope, no compression needed - the guy's exhibiting strong signs of having never been near a spectrum analyizer. It's a known fact that sounds move through a medium (air, wood, what have you) more quickly as the frequency increases. The percentage of difference between the upper and lower limits of our hearing range is very small, on the order of 1 to 1.5%, but for the alleged purists (I should probably coin the word "purerista" for them), that makes all the difference in the world. What they want to do is "slow down" the high frequencies in order to make everything come at them at the same time. Riiiiight. And how's that gonna emulate the real world, might I ask?

Not to mention, adding another pickup alongside the piezo won't do any kind of slowing down, the sum of the outputs will add algebraically, and the only way you might notice a difference in the time domain would be that the output of the "other" pickup swamps what you can/might hear from the piezo.

Yes, one can do some Fast Fourier Transforms (with either analog or digital means) and actually slow down the high frequencies, but is that what it would sound like in the real world? The easy demonstration is, does the acoustic guitar's wood (and internal air cavity) slow them down? I think not. Then why would you attempt to make the guitar sound unnatural?




sumgai

[sydsbluesky]

Nov 12, 2009 8:08:06 GMT -5 newey said:

In any event, I wonder if using multiple pickups on an acoustic, each run through an appropriate EQ stage and blended, might not give a fuller sound than just the piezo alone. After all, we put multiple pickups on electrics, why not on acoustics?

When I record with the Takamine I use the onboard undersaddle piezo type through the "Cool Tube" preamp on the guitar, run that out to a solid state amp (stays a little cleaner that my tube guys,) and into a small mixer to be EQ'd before sent out. I also run a condenser mic that I suspend in various positions - using a series of rubber bungees, because I like them more than the stands - which also goes into the mixer to be EQ'd on a separate channel. Both then go out to the power amp and the speakers.

From my understanding of recording acoustics this is very much common practice. The preamp on my Tak actually has a line in for an external source, presumably to be utilized by an added on mic or even a second piezo device of some kind, which can be blended into the undersaddle signal before it even leaves the guitar via a knob on the preamp. I've never used it.

Given, this is one heck of a guitar with one heck of a preamp, so doing it all inside the preamp itself is probably not common; usually this is done externally in a mixer, I believe. It really does add a good amount of depth to the sound when it's done well.

This is the same idea as multi-micing a guitar with no onboard electronics. There are a few commonly used ways to do it that account for blending the sound, matching volume mitigating phase cancellation and the like.

There is plenty of info out there well within the reach of google on recording tactics for acoustics.

As far as modding... eh, well... that's not really my area. Good luck though, and keep us posted!

EDIT: SSEEPPLLIINNGGG! Also not my area.

[JohnH]

This is the most discussion on acoustics tha we have ever had!

If my piezo thread is of interest, then Id like to follow up on the blended pickup notion. I find with that guitar, the best sounds are to get the piezo really bright so that it empahises the high-end stringy transient sounds, then blend it with some magnetic to add body.

I started with a magnetic system and added a piezo, which took some wrangling as you can see. Given that you already have a piezo system of good repute, how about keep that intact and add a simpler blending preamp after it to mix it some magnetic input from a sound hole pickup? Thats actually a very easy circuit because the magnetic pickup is already a good sized signal.

John

[treguiers]

Even as we speak I'm taking a rare earth magnetic pup and adding it to the guitar via a blender switch. I'm literally taking the outputs from both the piezo(after the pre-amp) and the rare earth sticking them onto a blender pot and running it to the input jack

In other words, no electronic heroics just yet.
I'll keep you posted

[JohnH]

Well give it a go! if you find that they dont mix well, just to test it. you could run the mag pu through a Boss pedal (or similar non true bypass pedal) switched off, and then blend the signals. Youll also see if the relative volumes are compatible. If they are, then an onboard mixer to blend them could consist of about 40% of the piezo circuit I used, including a buffer for the mag pu, a 500k single pot for blending then another buffer. I could draw a line around the part of my diagram that you would need

John.

[newey]

I could draw a line around the part of my diagram that you would need

Now John, that wouldn't mean that it was a modular design or anything, would it? ;D

Maybe I've gone off the G-nutz deep end, but I just envisioned a stock-appearing acoustic guitar sporting a mag neck pickup discreetly mounted at the end of the fingerboard, a condenser mic for the soundhole, and an under-saddle piezo, all running into a preamp/buffer circuit designed to accomodate each type of input. A Strat 5-way switch to give various combos of the 3 pickups. Mount it all in one of those Fender "Stratacoustic" models.

Voila! A 3-pickup acoustic Strat!

[treguiers]

Well If I'm gonna be the guinea-pig on this "acoustic strat" then you'll need to take things REALLY slowly!

Right! Blender pot is in. It seem to be acting like a switch. Dead centre the mag and piezo are on, move a hairs breath each side and the relevant pup goes. I searched the blender pot but could find an ohm rating, so I can't tell what it resistance is.(but I do have a multi-meter)

Also at the i/p jack there a terminal for the battery, I would imagine a cut off to the battery when there's no jack in the i/p. Anyway both the mag pup and the piezo have a wire serving this purpose, I connected them together onto the battery lug, now the piezo has a constant battery on signal(led light) even when there's no jack

Let's call this a tentative start. I suppose if I got the blender switch actually blending and sorted out the battery difficulty that would be a big improvement.......over to you guys!!!!!!..........

BTW my fishman prefix pre-amp has a vol pot and 4 faders: bass, contour, treble and frequency

[JohnH]

the magnetic pickup is just pssive - ie it does not use battery power?

You can measure the pot resistance with your meter. Is it a dual pot? set to mid point and measure between centre to left lug and center to right lug on one pot, Use the 2000k setting on the meter

John

[treguiers]

The mag pup is active with two 1.5v batteries(connected in series). They're those disc shaped batteries.

I put the meter to 2K and measured the resistance and got a reading of .583 and of .562 for each side. So would I be right in assuming that they're 500K pots?

Both pups have their own vol controls, the piezo uses a 9v battery and is considerably hotter than the mag pup.

As I said before although I have a little knowledge of electronic I am by no means an expert. Please keep that in mind when sharing your wisdom.
Thanks

[sumgai]

tre,

If you had the meter set to the 2KΩ range, and you read half of the pot's resistance, twice (once for each side), then added them together, you'd find yourself with a 1KΩ pot, not a 500KΩ pot.

Even for an active circuit this is a bit on the low side. I have a feeling that you're also reading other components in the circuit, which are in parallel with the pot, and that skews all readings into useless-land. To get a valid measurement, you'll have to disconnect at least one side of the pot. If I'm not mistaken, your unit is on a PCB, right? If so, then the whole pot will have to be dismantled from the board, sorry to say. Not terribly difficult, but it's a PITA, to be sure.

If you're looking at a two-ganged pot, then what I just said holds true for both sections - each section has to have one terminal lifted free, or in the worst case you'll have to remove the whole thing, before taking any readings.

HTH



sumgai

[treguiers]

Hi Sumgai,
I have a few of these pots, so I measured one that isn't connected to anything. So measuring two sets of three lugs one side was zero the other was .583 and on the other set of three lugs one side was .562 the other zero.

I get the feeling I'm going to need circuitry get this thing working the way I want it. But I'm up for that. Meanwhile I'm going out to put my superstrat thru its paces

[JohnH]

treguiers

Those pot values are very low as sumgai notes. Are you sure you were not on a 2000k range?

Anyway, I think the best way to blend two sources is with a single linear pot. Given that both pickups are active, you can do this with the part of the circuit outlined here:



The blending pot R32 has centre lug to output and outer lugs to the two pickup signals. The functions of the rest of the circuit you already have covered by the active circuitry on the pickups.

After the pot, there is a single stage buffer, Q3. The purpose of this is to take what has become a fairly high impedance signal from the pot and make a low impedance output, so that it is suitable for going to the line input of a PA, which is a basic need for electro-acoustic guitars. However, if you are testing this into a high impedance guitar amp input or a stomp box, you can take a signal straight after the pot without the buffer. In fact, you can probably also do away with the input capacitors C4 and C9, so only the pot is needed to test it.

In my circuit, I was starting with a very low raw piezo signal and boosting it and EQing it to match the magnetic. Hence the piezo stages Q2 and Q4 have significant gain while the input for the mag pickup Q2, has no gain.

In your case. if you wish to equalise the levels by boosting the magnetic signal, you could add another stage after the mag pickup before the blending pot, to provide a boost. The circuitry around Q1 or Q4 would do this job, but placed between mag pu and pot, where Q2 is.

However, I suggest your next step is to just try with a simple single pot into a guitar amp or stomp box.

cheers

John

[treguiers]

Hi John,
I set the meter on 2M ohms.

If my blender pots are of too low a value, is that why they're behaving like switches?

It is my intention to give this circuit a go but i need to get up to speed. Are the "Q" sections some sort of transistor/amplifying components?(I'm sure this is like asking Pat Metheny if he knows the pentatonic scale!)

Any thoughts on reconciling the two different battery sources, I'm sure they drain when not in use, due to my connecting the mag and the piezo battery wires together. But there's one i/p jack and one lug for the battery wire on the i/p jack.

I'll get there it might just take a while, but I'm at an age now where I'm not afraid to ask dumb questions.

Thanks John

[JohnH]

yes, those are transistors, 'Junction Field Effect Transistors'(JFETs) to be exact

If you scan down the first post of this thread, you will see details of where I installed something very similar in an LP copy

guitarnuts2.proboards.com/index.cgi?board=nutzoid&action=display&thread=3773&page=1

The two battery's that you have are a bit of a hassle. Usually on a guitar with active electronics, a stereo jack socket is used, and the battery negative is taken to the ring terminal, which is not connected to anything until a mono jack plug is inserted, shorting it to ground and letting current flow. But taking both battery negatives to the single ring terminal will create an unwanted circuit that will cause the 9V to push current backwards through the 3v, with no jack inserted.

I think the best answer will be to take out the 3V battery, put a 10uF capacitor in its place, and a resistor from the old positive, to the positive of the Piezo 9V supply. The value of this resistor is not yet known, but it has to be selected so that the 9V is reduced to 3V if that is what the mag pu needs. The value depends on the current taken by the mag pu. Can you measure that with your meter?

There are other possible solutions to the two power supplies, including the use of a diode, and the use of a switched jack socket

John

[treguiers]

I think a trip to my local electronics shop is called for! As regard checking the current of the mag pup,I assume I put the meter across the battery terminals, with the batteries still in it an have the meter on around 200mA? (pure guess)

[sumgai]

tre,

Nov 14, 2009 9:41:33 GMT -5 treguiers said:

As regard checking the current of the mag pup,I assume I put the meter across the battery terminals, with the batteries still in it an have the meter on around 200mA? (pure guess)

NO! Don't do that!!

To read current, you insert the meter leads into the circuit as if it were a component wired in series. You read voltage across the circuit, or one of the components such as a battery.... that's putting the meter in parallel with the DUT (device under test).

Even though we're talking about a small amount of current here, it's still a good idea to always start your measurement on the highest range. You turn it down (on your manual range meter) until the reading makes sense. That way, you don't risk blowing up the meter because of a bad guess. (Auto-ranging meters do this, that's why they take a second or two to settle down.)

At least you had the good sense to ask before you destroyed your meter. (OTOH, it'd be a good excuse to go buy a new one that has auto-ranging. )

HTH




sumgai

[treguiers]

The mag pup is a fishman rare earth. They are popular pups would there be a current rating in the specs? What value blender pots do you think would be appropriate.

The Article on the LP is very interesting. I am the kind of guy that just followed diagrams posted by you guys, but I am determined to understand whats going on so I would ask for your patience, I will get there.

[JohnH]

what im sayin' is - use a 100k or 500k single linear pot as your blender, with a buffer circuit as shown on my outlined diagram.. Outer lugs go to the pickup outputs, centre lug to your amp. This will give smooth continous blending across the full turn of the pot.

can you find an online link to your mag pickup?

John

[treguiers]

Specifications(rare earth pups)
All Models:
Output Impedance 1k Ohm
Recommended Minimum Load Impedance 10k Ohm
Supply Voltage 3 .0 – 3 .6V

Rare earth Humbucking:
Maximum Output Level -3dBu RMS
Output Noise Level <-98dBu RMS, A-weighted
Current Consumption 0.5mA
Battery Life >240 hours

www.fishman.com/products/details.asp?id=48

So with the single linear pot could i have it so that the mag or the piezo could be on by it self, or would there be one pup constantly on and you bleed the other pup into it via the single linear pot?

Also on your diagram the JFETS have three terminals, anything I looked at from my local electronic shop catalog had at least 6 legs(terminals) .They looked like integrated circuits. Is "J2N5457 NJF njfet.mod" a product number? With specific parameters and ratings?

Questions! questions! questions!

(BTW-The rare earth humbucker is the version I'm using)

Thanks John

[JohnH]

OK, thats good info. The JFET type is 2N5457, thats all you need to specify.

Heres a data sheet showing what it looks like:
web.mit.edu/6.301/www/2N5459.pdf

And for the buffer circuit shown, other types such as J201 and MPF102 will also work well, but with different resistor values.

For the battery voltage fix, it would be better if you can measure the current. Just remove the battery, first measure its voltage seperately (using 20V dc voltage range to see if the two1.5V cells are actually that), then wire up as follows:

battery(+) to meter(+), meter(-) to pickup(+)power terminal

pickup(-)power terminal to battery negative.

So al the current goes through the meter to get to the pickup. set meter to 20mA Amps range. It should measure and confirm the 0.5mA speced on the data.

If the current is 0.5mA, then a suitable resistor to drop the voltage from a 9V battery would be 12k = 6V (the voltage reduction)/0.0005A =12000 ohms = 12k

(question for sumgai and others - do you think this warrants a zener diode for a more stable supply? Im thinking not, but to have a 10uF cap across the pu power input to keep it steady durring transients)

On the linear pot as a blender, yes it should give 100% of one pup at one end, 100% of the other at the other, an a smooth transistion in between ie, exactly what you would want!

cheers

John

[treguiers]

WOW! You're good!

[sumgai]

tre,

Nov 15, 2009 7:16:22 GMT -5 treguiers said:

WOW! You're good!

So Exalt him! ;D


~!~!~!~!~!~
John,

No, I don't think one need worry about transients and protection therefrom.

First, we're talking about a large oversupply here, so there's no percentage to having an additional storage device (such as a cap) to hold a charge "for when the power supply goes low due to a high demand". If the battery is very weak, the problem might occur, true, but at that point, it's time to replace the battery anyways.

Second, a zener diode is meant, in these cases, to clamp the voltage at some maximum level, disallowing any increase to get through for any reason. This is actually the preferred way to regulate a supply voltage, versus a resistor, but it is more expensive. However, if a maximum current is the goal, then a zener is not useful, the resistor, operating as a current limiter, is indeed more appropriate. Which explains why you sometimes see both in a power supply circuit.

Not to mention that as the battery starts losing charge, the resistor maintains an equilibrium (think Ohm's Law) that says "less current available" to the circuit, which in turn means that the device is no longer operating at design spec. Yet another reason to consider a zener as the voltage limiter, and use the resistor as a current limiter.

At a "mere" half milliamp, I doubt you're going to see any transients that affect the supply by more than a percent or two. That shouldn't be enough to affect things (in a negative way), but if the active device is operating near its maximum ratings, then it might be a good idea to take steps to prevent transients, sure. I do have to say though, that I'd design to use the device at no more than 50% of its maximum ratings, and not worry about such things. But that's just me, YMMV.

HTH


sumgai