Double Bass Piezo pickups - how to care for

[jj.833]

I have Realist copperhead (10 years), Realist Lifeline (4 years), both work nicely. I put them on the bass and keep them there.
I also have Ehrlund which I take off very often as I'm switching it between two basses. I keep it in the original metal box. So far it doesn't look like it suffers, but it's been only a few months.

 

[DoubleMIDI]

I measured some capacitances of piezo elements/pickups. But I’m not sure that the values an be put into wasnex high pass filter, because I think the schematic doesn’t fit how the piezo works.

The voltage is produced inside the piezo material (distributed and depending on deformation/pressure), so the capacitance is in parallel to the voltage source. But the voltage source can only deliver a limited current, so this could be modeled with a capacitor of unknown value in series to the voltage source, and the known and measurable capacitance of the electrodes (brass disc, silver plating) will connect to the outer connectors of the series of voltage source and inner capacitor, like the connected cable capacitance.

So my measured capacitance is connected between drawn capacitance and resistor on one side and ground/shield on the other side.

That gives us a capacitative voltage divider between inner capacitance and electrode/cable capacitance, where the impedance defining resistor and (pre-)amp input is connected.
And it also gives us a high pass filter via the inner capacitance and resistor.

That does not answer the question of the high pass filter corner frequency, since the inner resistance is unknown.

Measuring by voltage drop with the same vibration and different external (parallel to the piezo connectors/cable/electrodes) might be possible but is not simple to do, so we still don’t know the inner capacitance.

A lot of words for telling you we are (almost) no step further, but maybe some of you with a bit of electrotechnical knowledge can understand and verify or contradict my ideas.

Now the not very useful measurements of piezo elements and (passive) piezo pickups:

Elements

Piezo disc (ceramic piezo material)
Diameter of silver plating electrode: 23 mm
Thickness: not measured, rather thin (buzzer)
Capacitance at brass and silver plating 70-76 nF

Piezo disc (ceramic piezo material)
Diameter of the silver plated electrode 14 mm
Thickness: not measured, rather thin (buzzer)
Capacitance at brass and silver plating 10 nF

Piezo rod (ceramic piezo material)
Silver plating electrodes at both ends
Length 15 mm
Diameter 6.8 mm
Capacitance at the electrodes 27 pF (!)

Pickups

Shadow SH-950 (double element bridge wing pickup, Underwood clone)
Thick ceramic piezo disc elements
Electrode distance (thickness): (less than) 4.8 mm
Electrode diameter: (less than) 10 mm
Capacitance at connector 1.3 nF
(Don’t forget that both elements are connected in parallel, so the capacitance of one element is half of the value!)

Yamahiko (single element, probably ceramic piezo rod inside adjuster axis)
Capacitance at connector 540 pF = 0.54 nF

Ehrlund (probably synthetic piezo film pickup)
Capacitance at connector 13 nF

MSP (probably ceramic piezo disk)
Capacitance at connector: 11 nF

Harley Benton (Thomann brand) piezo pickup for tuners
Capacitance at connector 6.9 nF

Keep in mind that this is the outer capacitance from hot to shield, not the inner one which is important for the frequency response under resistive load (input impedance).

I would be happy to get a comment to my ideas from an electronics guy here like @fdeck or @drurb or anyone else with a professional knowledge of electronics.

p.s.: I forgot an inner resistor (in parallel to the inner capacitance) for self discharge of the piezo. There might be more that I have overseen…

 

[Ric Vice]

Durb, is no longer on Talkbass, he left several years ago.

 

[DoubleMIDI]

I forgot an inner resistor in series with the voltage (or current?) source and the inner capacitance. But I think that makes it even more complicated to understand. It is for the inner resistance that limits the current flow, I think.

And yes, I haven’t heard from drurb for years too, but he still has a profile here and was named as an example mostly.

 

[Wasnex]

I measured some capacitances of piezo elements/pickups. But I’m not sure that the values an be put into wasnex high pass filter, because I think the schematic doesn’t fit how the piezo works.

The voltage is produced inside the piezo material (distributed and depending on deformation/pressure), so the capacitance is in parallel to the voltage source. But the voltage source can only deliver a limited current, so this could be modeled with a capacitor of unknown value in series to the voltage source, and the known and measurable capacitance of the electrodes (brass disc, silver plating) will connect to the outer connectors of the series of voltage source and inner capacitor, like the connected cable capacitance.

So my measured capacitance is connected between drawn capacitance and resistor on one side and ground/shield on the other side.

That gives us a capacitative voltage divider between inner capacitance and electrode/cable capacitance, where the impedance defining resistor and (pre-)amp input is connected.
And it also gives us a high pass filter via the inner capacitance and resistor.

That does not answer the question of the high pass filter corner frequency, since the inner resistance is unknown.

Measuring by voltage drop with the same vibration and different external (parallel to the piezo connectors/cable/electrodes) might be possible but is not simple to do, so we still don’t know the inner capacitance.

A lot of words for telling you we are (almost) no step further, but maybe some of you with a bit of electrotechnical knowledge can understand and verify or contradict my ideas.

Now the not very useful measurements of piezo elements and (passive) piezo pickups:

Elements

Piezo disc (ceramic piezo material)
Diameter of silver plating electrode: 23 mm
Thickness: not measured, rather thin (buzzer)
Capacitance at brass and silver plating 70-76 nF

Piezo disc (ceramic piezo material)
Diameter of the silver plated electrode 14 mm
Thickness: not measured, rather thin (buzzer)
Capacitance at brass and silver plating 10 nF

Piezo rod (ceramic piezo material)
Silver plating electrodes at both ends
Length 15 mm
Diameter 6.8 mm
Capacitance at the electrodes 27 pF (!)

Pickups

Shadow SH-950 (double element bridge wing pickup, Underwood clone)
Thick ceramic piezo disc elements
Electrode distance (thickness): (less than) 4.8 mm
Electrode diameter: (less than) 10 mm
Capacitance at connector 1.3 nF
(Don’t forget that both elements are connected in parallel, so the capacitance of one element is half of the value!)

Yamahiko (single element, probably ceramic piezo rod inside adjuster axis)
Capacitance at connector 540 pF = 0.54 nF

Ehrlund (probably synthetic piezo film pickup)
Capacitance at connector 13 nF

MSP (probably ceramic piezo disk)
Capacitance at connector: 11 nF

Harley Benton (Thomann brand) piezo pickup for tuners
Capacitance at connector 6.9 nF

Keep in mind that this is the outer capacitance from hot to shield, not the inner one which is important for the frequency response under resistive load (input impedance).

I would be happy to get a comment to my ideas from an electronics guy here like @fdeck or @drurb or anyone else with a professional knowledge of electronics.

p.s.: I forgot an inner resistor (in parallel to the inner capacitance) for self discharge of the piezo. There might be more that I have overseen…

I don't claim that what I presented is 100% technically correct...I was a military electronics tech, so my training is vocational. I do not have a degree and I am not an engineer.

The intent of my post was really just to say the input impedance of the driven device has an impact on the frequency response of the pickup. I only covered the HPF aspect, but there are other aspects as well. The point is the frequency response of the preamp is only part of the answer. The preamp and the pickup should be treated as a system...I.E. the frequency response of the system is ultimately what matters.

In researching the post I found various models of the piezo pickup showing both series and shunt capacitance. AFAIK the shunt capacitance is used when the piezo element is treated as something called a charge source. I also came across circuits called charge amplifiers that are related. AFAIK, the HPF model I showed is relevant when the capacitor is considered a voltage source.

You may want to review the technical docs I attached to this post: Sarno Black Box I think a good starting place is p. 36 of the MSI-Techman doc. This will be p. 39 in the PDF viewer...but the bottom of the page is numbered 36.

 

[CryingBass]

I just scored a Parker Fly ( early run ) with Piezo under each string in the bridge. Not a major feature to me, but important.

Now you guys got me worried. Does anyone know if these under-the-bridge piezo elements are as possibly fragile as some of the DB pickups you are referring-to in this post? Sorry for the derail, but you guys know a heck of a lot more than I in this regard.

I have the same thing in my vintage Yamaha TRB 5P, a Godin Ultra 5, etc. I don't gig with them, but am I going to need to treat them like a champagne glass from now on?? ( kinda kidding, but kinda not )

 

[Wasnex]

I just scored a Parker Fly ( early run ) with Piezo under each string in the bridge. Not a major feature to me, but important.

Now you guys got me worried. Does anyone know if these under-the-bridge piezo elements are as possibly fragile as some of the DB pickups you are referring-to in this post? Sorry for the derail, but you guys know a heck of a lot more than I in this regard.

I have the same thing in my vintage Yamaha TRB 5P, a Godin Ultra 5, etc. I don't gig with them, but am I going to need to treat them like a champagne glass from now on?? ( kinda kidding, but kinda not )

I have heard of failures but AFAIK, they are not frequent. May be more likely if the bridge takes some sort of sharp impact.

My #1 since about 1995 has been a TRB6P. When I was still on active duty, I had a military TRB6P and also bought a used TRB6P for private gigs. No piezo problems with either instrument

 

[Ric Vice]

I also have a Godin Acoustibass with a under saddle piezo. I've owned that bass for 30 years, without any issues so it's "a horse of a different color". Unlike a double bass pickup, the Piezo elements are protected by the bridge and body
of the EB. Just my take of course. Those early Parker Fly Basses were not manufactured for very long. Why that was I don't remember.

 

[CryingBass]

Thanks guys! My takeaways are 1. I should not worry about such things and 2. If I do fall asleep while playing, just don't let my head land on the bridge...

Really, that's comforting feedback. A rarity these days

 

[DoubleMIDI]

The string tension and bridge pressure on a guitar should be a lot lower than on a DB, even with two more strings.

It also depends on the piezo material. If it is ceramic it would be sensitive to an impact and can break.
If it is a synthetic material, it will survive medium impacts, but it might (!) dissolve after ten to twenty years with he synthetics. A lot of synthetic material either dissolves or gets extremely hard and breaks easily, similar to the ceramics, after a long time or under UV radiation.

 

[DoubleMIDI]

…
AFAIK, the HPF model I showed is relevant when the capacitor is considered a voltage source.

You may want to review the technical docs I attached to this post: Sarno Black Box I think a good starting place is p. 36 of the MSI-Techman doc. This will be p. 39 in the PDF viewer...but the bottom of the page is numbered 36.

The problem is, that the piezo isn‘t purely a voltage source. It would take away a lot of problems if it would be.
If it would be a voltage source, then there would be an unlimited current that can flow and input impedance would be irrelevant. But since there is a limit or non-linear flow of electrons travelling through during a certain time span, high voltages can break down if they stay for too long before changing polarity which will result in a distorted waveform.

I thought myself to have a look at the MSI-manual but didn‘t had the time to browse through it again. (The last time had been many years ago…)
But I can remember that the highest input impedance needed for one of their sensor foils was 22 MegOhms.
I might have even measured one of the sensors I got from them but cannot remember where I have placed them.

 

[Wasnex]

The problem is, that the piezo isn‘t purely a voltage source. It would take away a lot of problems if it would be.
If it would be a voltage source, then there would be an unlimited current that can flow and input impedance would be irrelevant. But since there is a limit or non-linear flow of electrons travelling through during a certain time span, high voltages can break down if they stay for too long before changing polarity which will result in a distorted waveform.

I thought myself to have a look at the MSI-manual but didn‘t had the time to browse through it again. (The last time had been many years ago…)
But I can remember that the highest input impedance needed for one of their sensor foils was 22 MegOhms.
I might have even measured one of the sensors I got from them but cannot remember where I have placed them.

MSI makes a bunch of different piezos for different applications. If I remember correctly the 22meg requirement is not typical.

The post I linked also includes an instruction sheet for a shielded MSI piezo that is suitable as a musical instrument contact pickup.

A quote from the document:

The transducer behaves like an “active” capacitor, consequently, loading of the signal by the input impedance of the measuring device must be considered.​

Recommended impedance is 10meg, but the instruction sheet says:

Due to the thinness of the films, the associated capacitance can be sufficient to give adequate low frequency response into standard 1 MΩ loads
​

I am not making any sort of assumption that the capacitance of the piezo, behaves exactly the same as a capacitor in an HPF. But AFAIK it does apparently behave very similarly.

Here is the response plot that shows the low frequency response with 10meg and 1meg.

I don't see any info that actually specifies the break point of the HPF. But the document says:

Frequency response is inherently flat into the MHZ region with only the R-C rolloff at low
frequencies distorting the profile.​

Also, the piezo manifests resonance above 10MHZ...but of course this is not a problem for our application.

I didn't see capacitance listed anywhere in the instructions, but it's listed as 2.78nF in the associated datasheet.

The datasheet has a better response plot IMHO:

If I put 2.78nF into the HPF calculator: A 1meg load results in a 57hz HPF, while a 10meg load results in a 5.7hz HPF. Not exactly the same results as the charts I provided in this post.

 

[DoubleMIDI]

Putting this (external) capacitance in the HPF calculation is wrong. That‘s what I basically said before.

This capacitance is in parallel to the “voltage source”, not in series as you need for your calculation and this is the reason why using this value doesn’t fit the diagrams.

It is the inner capacitance that you cannot measure by simply connecting a capacitance meter to the piezo that is responsible for the HPF effect because it limits the flow of electrons that is responsible for the HPF effect depending on the resistive load of the amp input.
But the outer capacitance does have an effect on the signal level as it builds a capacitative voltage divider with the inner capacitance.

Since MSI did extensive measurements, you might be able to calculate the inner capacitance using the data you can read from the diagram.

I know (and I think I also said) that 22 MegOhms was the highest impedance needed for a certain sensor from MSI.
I wouldn’t say there is a typical impedance, rather a wide typical impedance range for the sensors of MSI and 22 MegOhms is the upper end.
(I also looked at schematics of condenser microphones some time ago, the input impedance from capsule to FET/valve is in the one GigOhms range to avoid low frequency cut if I remember correctly.)

I have read about the SDT1, but it is a bit small for a DB bridge foot (might fit a cello bridge foot well) and was a bit expensive compared to the foils.

I think the Shadow SH-965NFX is a better solution than a (double) SDT1 plus a (double) impedance buffer, since the FET of the impedance buffer sits directly on the sensor foil of the Shadow pickup, keeping the high impedance wire extremely short. It is also reasonably priced if you notice that there are two sensor foils coming with it (as well as the battery adapter for the impedance buffer) and large enough to fit my 4/4 5-string DB bridge.

So that’s the reason I didn’t order a SDT1 when I was experimenting with MSI piezo foil sensors.

BTW, I killed one sensor from MSI by trying to remove the user applied copper adhesive from it to rearrange, because it peeled off the unprotected silver coating from one side.

 

[Wasnex]

Putting this (external) capacitance in the HPF calculation is wrong. That‘s what I basically said before.

This capacitance is in parallel to the “voltage source”, not in series as you need for your calculation and this is the reason why using this value doesn’t fit the diagrams.

From page 6 of the MSI Tech Manual

This is the same formula used by the online RC Calculator.


From page 38.

If you insert the capacitance and resistance value into the calculator, you get the same answers shown in the example:

High Pass Filter Calculator (learningaboutelectronics.com)

 

[DoubleMIDI]

I just doubt that the capacitance that can be measured at the two electrodes of the piezo are the capacitance in series with the voltage source.

But I need to think about it and read about the capacitance of plates and distances and how to calculate the value.
The brass plate and silver coating of a piezo disk build a capacitor from hot to ground (so in parallel with the voltage source) but the value might be much smaller than the measured one. So I’m unsure now how it works and which one influences more or less.
I’m also rather sure that this is a simplification. Maybe a useful one, at least in some situations.
I simply don’t have the time to look into the manual for the next three months. I have a fixed deadline and no idea to do all the work until then.

I think that the voltage source is not an ideal conductor when no voltage is produced but rather breaks the connection (no current flow). And therefor I cannot see how the (inner) capacitance in series with voltage source can be measured at the electrodes.
That is my main problem.

But thank you for looking things up for me, I know that I’m not always right, we can let it stay that way until I have a bit more time to get into that a bit deeper.

If you are right, we have values to calculate the input impedance needed for some pickups now.

 

[DoubleMIDI]

My assumption that the voltage source does not conduct if not active was wrong. I found it out by looking at the piezo as an actor (speaker), not as a sensor (pickup).

A piezoelectric material not only generates voltage when compressed or bent, it also deforms (bends) when an electric signal is applied.

If the voltage source wouldn‘t be conductive, that effect wouldn‘t happen, so it must be conductive and Wasnex and the MSI manual are right.
Conceptually I was right that there is an external capacitance from hot to ground (at the electrodes of the piezo) by the conductive plates of the electrodes, but their value is much smaller than the inner capacitance, so the influence is also very small.

Well, on the way to find the right thing, one often gets stuck or misleaded. That’s a normal thing, and if one finds the way back, it is worth the hassle with the wrong routes.
There is always something to learn.