piezo preamp

[artantaaa]

Hello everyone. I have an upright bass with a piezo element under the bridge. It sounds awful plugged directly into my amp. There's almost no low end at all. I did some reading and discovered that the problem is the impedance mismatch between the pickup output and the amp input. I need a preamp. My current plan is to build this FET buffer circuit.

Discrete FET Guitar Preamp

My electronics knowledge is limited, so I'm not sure if this circuit is the best way to solve my problem. I would greatly appreciate any advice that you might have for me.

[52 Bill]

Yes that circuit will fix your impedance mismatch.

If you search, you may find other versions with some tone control or other features that you may or may not find useful.

[artantaaa]

Thanks for the reply. It's good to hear I'm on the right track. I did see some similar schematics with more features, but I chose this one because of its simplicity. The only thing I want it to do is change the impedance of my pickup signal to the impedance that my amp wants to see. The trouble is that I don't know the output impedance of my pickup or the input impedance of my amp. My proposed preamp has an input impedance of 3 Mohms but this value can be changed by swapping in a different resistor. I have read that piezo pickups have an impedance closer to the 10 Mohm range, so I might experiment with this value. The output impedance of the preamp is 6800 ohms and can not be changed because it would somehow affect the FET. Would I need to use a different FET in order to change this value? I assume 6800 ohms is in the proper range for an amp, but I was just reading that the value of the amps input impedance should be 7-10 times that of the source impedance. Is this always the case? Does this mean that my amps input impedance is closer to 68000 ohms?
Hopefully this preamp will sound great just as it is in the schematic, but I want to learn as much as I can about the way it works before my parts arrive in a few days. If you can offer any info at all, please do.

[booj]

It may not work. All it's doing is feeding the signal from thje piezo pickup into a high impedance preamp, which won't do anything but amplify the sound you allready are getting.
I had good luck plugging my Roland ready Strat with piezo pickups into a box with nothing but a .1uf cap across the input. (That is the piezo pickups only of course)
The cap shunted the unwanted highs and shifted the freq. response so it sounded to me "normal"
Take it for whnat it's worth for free info, but it worked for me

[artantaaa]

That's a good idea to use a capacitor to filter out the unwanted frequencies, but I am far from the fine tuning stage. I have been running my bass through a boss GE-7 equalizer pedal, but I can't get the low end to come through, even with the highs cut and the bass maxed out. It does add gain to the signal though. Without it, I really had to crank up the amp to hear anything at all. I would think that the input impedance of the pedal is similar to the input impedance of an amp. From what I've read, the impedance mismatch puts such a load on the piezo element that it cuts off frequencies below a certain point. Maybe someone here could provide a better explanation of how this happens.

I was able to find a manual for my amp that lists the input impedance as:
0db 220K ohms -6db 44K ohms
My amp does have high and low input jacks, and I'm guessing that those are the values for each of the jacks. To me, that range seems suitable for the output impedance of the preamp, 6800 ohms. Should I apply this same 10:1 ratio to the pickup side of the preamp? For example, if the output impedance of the piezo element is 1Mohm, should the input impedance of the preamp be 10Mohm?

[52 Bill]

Don't over think this problem before you get your preamp built.
Large mismatches in impedance will cause losses of signal frequency and strength, but it is not critical that everything meet an arbitrary standard.

The basic premise is that you should always drive an input with a source that is lower in impedance. Currently you have a super high impedance pickup trying to drive a high impedance input. Adding the buffer will convert the super high impedance down to a medium impedance, that the amp will be able to handle with no problem.

I say build the preamp and see how it sounds. Once you've tested it, then you can try altering the input resistance to see if it makes an audible difference in the sound.

If you're concerned about whether or not the impedance match is your problem, you could try and find a basic high impedance booster pedal and try it first. Something like a DOD fet preamp with a high impedance input.

Hope this helps.

[booj]

I didn't calculate anything or try to figure anything out. I just wanted to cut the high end a bit. I tried different values of caps and found that as the cap got bigger the highs went down and the bass went up. Go figure. Oops pardon the pun.

[artantaaa]

I do plan to try the preamp exactly as it is in the schematic first, but I think I will have to raise the input impedance from 3Mohm. I don't have any idea how to measure impedance, but my multimeter tells me that the resistance across my piezo element is 2.4Mohm. This makes me think that an input impedance for the preamp of 10Mohm or 20Mohm would be better. Do you think I'm overestimating the importance of this 10:1 ratio? The description of the preamp gives me the impression that it was designed for magnetic pickups with a lower output impedance.

It also occurred to me to look up the impedance of the boss ge-7. The boss site says that the input is 1Mohm and the output is 1Kohm. This makes sense to me because even when all the sliders are centered at 0, I get more bass and volume when I turn the pedal on. It still sounds nothing like the acoustic bass, but the sound does improve. I think I will try experimenting with capacitors, but even if the highs were gone, I would have to turn the amp and pedal up to the point of feedback and distortion to get enough volume on the low E string.

Hopefully my parts will arrive soon so I can find out if this preamp will work. Most likely I'll just get a new set of questions to ask though.

[J M Fahey]

Hi.
Build the preamp as-is, it will work fine.
You may want to replace the 6K8 resistor with a 10K trimmer pot, and set the J201 drain voltage to around 5.5 Volts.
Build it in a small box with a 9V battery, so it can be physically as close as possible to the Piezo Pickup.
You might wear it on your belt or shirt front pocket or in the floor with a 2 foot cable; the preamp to amp cable can be any length.
Hammond boxes are excellent for this.
You can also replace R4 with an output volume pot (50K or 100K, preferably audio taper but a Linear will do) and/or wire a footswitch across it, for muting whenever necessary.
Only after you build and test it properly, you may, if you wish, use a 10M input resistor instead of the 3M3 one, although I doubt you'll hear much improvement.
The pickup to preamp cable must be the highest quality available.
Consider recording studio grade, not music shop stuff, and shop accordingly.
Regular braided wire shield will not do, ask for metallised polyester extra shielding, music shop personnel will not even know what you are talking about.

[artantaaa]

I hadn't thought of using a pot to set the drain voltage, but that would make things easier. Thanks.
I was planning to build the circuit in a stomp box sized enclosure to use on the floor. I hadn't considered the cable length as a problem. Now that you mention it though, it would be just as easy to mount it on the tail piece in a smaller box. What is it about a long cable and/or low quality shielding that causes a problem? Picking up RF or 60hz noise? Capacitance? Does it just weaken the signal? Right now, I have some automotive type wire going from the piezo to the output jack that is wedged between the strings above the tailpiece. I then have a 10 ft no name black cable going to the amp. I'm not hearing any noise from this setup so far. Mounting the preamp on the bass would cut down on the length, but should I be worried about shielding the wires from the pickup to the preamp? Would wrapping them in aluminium foil and then heat shrink be sufficient?

[Steve Conner]

With piezo pickups:

Cable capacitance decreases the output level.
Shunt resistance rolls off the bass.

This is as opposed to ordinary magnetic guitar pickups, where shunt resistance decreases the level and cable capacitance rolls off the treble.

The fundamental reason for this is that the impedance of pickups varies with frequency. A piezo looks like a capacitor, so its impedance gets higher as the frequency goes down. A magnetic pickup is an inductor, so it does the opposite.

The rolloff is actually determined by the RC time constant, so that's why adding lots of capacitance seems to restore the bass. It just brings down the rest of the frequencies to match the low bass output.

So overall, I think you can use a reasonable length of cable to the preamp, and it doesn't have to be super-high quality. I've seen piezo sensors in industry hooked up to a preamp with 20ft of RG-174 coax. The only important thing is for the preamp to have a high input resistance, if you want a good bass response.

[J M Fahey]

Hi artantaaa.

mount it on the tail piece in a smaller box.

I did not suggest it to you just to make "mechanical" things easier.
I have built small preamps attached with Velcro to the "heavy" part of the bridge, where it does not hurt the sound but the jack must allow the exit cable perpendicular to the double bass, because if it touches the body you can hear the scratching noise; so I guess the box in the floor will be fine, plus you will find the mute footswitch very useful.

Picking up RF or 60hz noise? Capacitance? Does it just weaken the signal?

All of the above.

Right now, I have some automotive type wire going from the piezo to the output jack that is wedged between the strings above the tailpiece. I then have a 10 ft no name black cable going to the amp. I'm not hearing any noise from this setup so far.

You're right, so far.
Fact is , you are running into a relatively low impedance (for a Piezo) guitar preamp; when you run into a 3 Meg one things will start to get hairy, .... just imagine into 10 Megohms.
Don't know what automotive wire is, although I imagine is some kind of shielded braid wire.
What I suggest isn't even expensive, only Guitar players don't use it, but it's very common Recording Studio fodder.
Let me explain it:
imagine we want to shield *light* (which is, after all, an electromagnetic wave)
Imagine you put a lamp inside a regular shielded cable: most of the light will be covered, but in a dark room you'll see a zillion minuscule dots of light between crisscrossing shield wires, they are not a *solid* layer of metal after all but some kind of "cloth".
That's a regular shielded wire.
Now you wrap it in a sheet of kitchen-quality aluminum paper (your idea): no light will be visible.
The idea is electrically excellent, but you twist that wire a couple of times and the aluminum cracks.
Enter Belden Corporation, the best wire maker in the World.
They wrap a wire with a foil of metallised polyester and fold back the joint to avoid it becoming an "entrance point"
See:Reliable Performance in Broadcasting Environments
and: Hum Rejection in Unbalanced Audio Cables -- Blue Jeans Cable
Maybe all this is just useless talk, because I did not ask you an important question, which should heve been the first:
Does the microphone already have its own wire? What length, connectors? Is it homemade or commercial?

Hi Steve:

I think you can use a reasonable length of cable to the preamp, and it doesn't have to be super-high quality. I've seen piezo sensors in industry hooked up to a preamp with 20ft of RG-174 coax

Respectfully, I'd like to emphasize two point (not exactly disagreeing)
1) RG-174 coax *IS* high quality cable.
Fact that it's cheap and easily available does not mean it's not so.
Personally I don't buy the theory that good audio equipment needs to be made by Tibetan Monks in top of the Himalayas shops "to be closer to the Cosmic Energy" or whatever marketing cr*p like that. I guess you also do.
2) Piezo sensors (and other industrial types) can be perfectly happy with a 10 dB S/N ratio (and even less) because signal is "cleaned" via Schmitt triggers, comparators, monostables, etc., before being further processed.
I often work with Symphonic Orchestras , String Quartets, large Tango Orchestras, etc. supplying them with custom made pickups and preamps.
There even the *slightest* hum/hiss/crackle is unacceptable.
As a side note:
A favorite client of mine are "Les Luthiers" Les Luthiers - Wikipedia, the free encyclopedia who use my stuff. The "marimba de cocos" , a sort of xylophone made out of tuned coconut shells and the "musical typewriter", a modified typewriter where the letter-hammers hit tuned aluminum bars. proved specially difficult, needing a combination of Piezo and Electret mics.
I got the job in the late 70's when their Tascam live/recording console proved useless against heavy RF interference.
In the 90's, the artificially cheap Dollar let them replace all my stuff with different models of Neumann and Sennheiser mics. Oh well !!
You NEED to see their "Informal Instruments": Les Luthiers-------------------- Sitio Oficial
These are the latest ones, please check also all posted by category on the left , under "instrumentologia". Have fun.

[artantaaa]

Well, my parts arrived yesterday and I've got it working. The low end is nice and loud. I still have a few things to figure out before I call it done though.
I need to put it in a box and mount it on the bass somewhere. I was thinking that tying it to the back of the tail piece with some kind of cushion under it would work well. I'm going to wire the pickup directly to the preamp and get rid of the jack between the strings. I thought I had stopped the jack from rattling, but I can hear it again now that I'm running through the preamp. The wires are now supported by some foam weather strip stuffed in the wing of the bridge. They were making noise too. I'll also use some shielded wire. What I meant by automotive wire was insulated copper wire that came out of a car. No shielding at all.
I did use the 10M input resistor. It gave me noticeably more bass than the 3M, but there wasn't a huge difference, just as everyone said. I also tried a 20M and got no improvement at all. I need to get that 10K pot for the drain voltage. It was a little low, and I got it close to 6V with a smaller resistor. I couldn't hear any difference though. What does this adjustment affect?
I am still using the eq pedal to boost the signal +15 db so I don't have to turn up the amp so much. This isn't a big problem but I would rather not have to use a pedal. Is there an easy way to add more gain to this circuit, or would I have to use a different schematic with multiple transistors for that?
Currently I have a radio shack 273-073 under the bass foot of the bridge. Would adding another to the other side give me a stronger signal? Would I get a more natural sound with them in a different place, like the side of the bridge? Are these piezo elements better or worse than others?
I also tried my guitar through the preamp into my ampeg ss-150, and it sounds like a tube amp. Now I need to build another one.

[J M Fahey]

Hi artantaaa.
To begin with: CONGRATULATIONS !!
You did real well, and a lot of it on your own, uncoached.
You ask many things, I'll try to answer what I can:

I was thinking that tying it to the back of the tail piece with some kind of cushion under it would work well. I'm going to wire the pickup directly to the preamp and get rid of the jack between the strings.

As I told you, I mounted a few to the (front of the) tailpiece with Velcro, but if you can fit it *behind* (I found that space too tight) go ahead.
Avoid bumps and scratches again the Bass body, they will be picked better than the strings, believe me.
Short wiring is excellent, 4" of regular wire are better than 3' of a premium one.

I need to get that 10K pot for the drain voltage. It was a little low, and I got it close to 6V with a smaller resistor. I couldn't hear any difference though. What does this adjustment affect?

Now you don't, you already corrected it.
Signal can get no higher than 9V and no lower than around 2V, depending on the FET.
For maximum clean signal you need ([9-2]/2)V=3.5V below 9V, which translates into 5.5V from ground.
In theory you should apply signal to the gate until light clipping and adjust the preset for symmetry while watching a scope, but this calculated value seats you within millivolts of that.
In short: between 5 and 6V on the drain are good.

Is there an easy way to add more gain to this circuit,

Yes, but you risk clipping, not worth the risk.
In fact if you run your amp volume too low and hit it with a boosted output, you risk that too.
I suggest you send the preamp straight to the amp and use a sensible volume setting.
Of course, if you have to run it on 9 or 10 all day, there's some mismatch there, but I doubt so.

Would adding another to the other side give me a stronger signal?

Maybe some, but the point is, it would sound more balanced.
Buy 10 piezos, many times I found them out phase, I think that's not checked because it's unimportant in buzzers.
I have used up to 5 or 6 covering a classical piano soundboard and in the beginning got crazy with unexpected random dead keys, which *acoustically* did sound perfect, until I suspected phase and ... Bingo!!
Now I test them one by one and they get marked with a drop of red or black.
I just tap them with my NASA approved , NBS calibrated , multipurpose Bic pen, and watch the pulse on the scope.

I also tried my guitar through the preamp into my ampeg ss-150, and it sounds like a tube amp.

Fact is, your guitar sound now *reaches* the SS140; and the 3x boost certainly helps.
And ..... Fets *are* good.

[Albert Kreuzer]

Is there an easy way to add more gain to this circuit, or would I have to use a different schematic with multiple transistors for that?

You'll get more gain if you bypass the source resistor with a 100µ cap. If you still need more, just add another stage, but you'll risk clipping as JM said.