Welcome to Solid State Guitar Amp Forum | DIY Guitar Amplifiers. Please login or sign up.

June 21, 2024, 09:35:26 AM

Login with username, password and session length

Recent Posts


Acoustic Guitar Preamp: Is the active DI box the way to go?

Started by smackoj, May 14, 2014, 08:56:00 AM

Previous topic - Next topic


hi friends; I thought maybe a new thread was warranted instead of jumbling up the earlier thread on Phatt's Phabbtone (a real peach of a design IMO). I was asking how to get the Phabbtone in use as a preamp for an acoustic guitar with a passive transducer inside it. These wood block pickups (I use a Schatten built in Canada) only put out a few mv so most of the preamps and some of the DI boxes aren't easily compatible.

Here are a few schemos and a great website review of 2 of the Radial Engineering DI/preamps on the market. I found these in search of an answer to my original question (can I mod Phatt's Ptone to work with 'quiet' acoustic pickup?). The Firefly DI by Radial sounds fab even on my laptop 'crap can' spkrs!  But, the bad news is they sell for 600 USD, yikes!

So, please read, review, and join in if you know how WE might be able to build a better mousetrap or DI box.  :tu:

Radial DI reviews here:  http://www.acousticguitar.com/Gear/Reviews/Radial-PZ-Deluxe-and-Firefly-Review


Re the Vid on the above mentioned link;
Just be very aware that the guitar used in that video is NOT passive pickup.
So you don't get to hear how good it would work with a passive peizo.
If you can afford the tone bone you may not need the PhAbbtone as Tbone has sweepable para mid control.

Re the circuit you posted; You will only need the transformer if you want balanced output.
Just use the unbalanced section for normal use.

Roly's got a similar circuit here;
http://www.ozvalveamps.org/pickups.htm    A boot strapped input with follower 
Down the bottom after peizo section.
You can use a fet for the first active device if you want.

Unless you want para EQ you won't have to spend much to get the job done.

J M Fahey

Small details about both schematics posted above:

1) the transformerless Op Amp one has only 50K input impedance.
Standard for Piezo pickups is 1M or higher

2)the second one is slightly better at 200K, still short of the usual spec and besides requires a VERY expensive transformer.


+1   :dbtu:

High enough input impedance is important with wound pickup guitars (1M min.), but with piezo elements, particularly small area ones like bridge-slot strips, having a high enough input R is critically important.  The old standard for piezo record player pickups was 2M7, and even that was a trifle low.  Having too low a pickup load causes a serious loss of bass output, where "bass" can be high in the mid-band, making it sound very thin and tinny.

470k load

A few sums suggest that the two-BJT bootstrapped front end (originally from a Miniwatt Digest yonks ago) might have an higher input impedance than a single FET (MPF102, as well as being less prone to electrostatic damage).

From LTSpice;

Vbias = 0.968Vsig
Vbase = 0.976Vsig

0.9760 - 0.9680 = 0.0080V 8mV

I = E/R, 8mV, 180k
(8*10^-3)/(180*10^3) = 0.00000004A  0.04uA

for Vin = 1V, 0.04uA represents;

R = E/I

1/((8*10^-3)/(180*10^3)) = 22,500,000 or 22.5 Megohms

The capacitance of any cable connected directly to a piezo must be small (e.g. AM car radio foam centre co-ax).  If the pickup has ample output you can try using a capacitive divider right at the element.


I've done a bit of 600-ohm balanced line stuff, particularly around radio stations, and used a few trannies for wide band audio.

Cheap trannies we normally wouldn't think too much of, such as 600-600, 600-600+600, 600-150, etc ohm used around modems and the like for phone line isolation, actually work very well for guitar.  The trick is that they are all measured for bandwidth at maximum rated signal level, and if run below this level will do a lot better - the more iron the better, but keep them away from hum fields coz that also make good pickups.

On old style 600 ohm to grid mic input transformer, found in some valve PA's, make a great isolation tranny for Line Out.
If you say theory and practice don't agree you haven't applied enough theory.


thanks for the really helpful and interesting info. I am going to have to do some reading on trannies before I can expect to reach my goal. Transformers seem somewhat mystical and I don't want to waste money doing garage 'experiments' with bunches of them.

adios for now friends.  jack :tu:


Hi Jack,
Quote from J M Fahey on the other thread;
"Any realistic transformer won't have input impedance into the megohms which you need."

Which roughly translated means you don't use a transformer for a peizo pu.  8)

Can I ask, What goal exactly are you hoping to achieve with the aid of a transformer?


I'm nothing if not cheap.  But there's a logic (apart from getting parts free from dumpster-diving).  Some parts are cheap, like 600 ohm - 600 ohm trannies, because they are used in vast numbers, and they are used in vast numbers because lots of designers use them, and they select them, perhaps over more "ideal" components because they are cheap, widely available, and likely to remain so for a long time.

If you get into them deeply transformers are quite complex little beasties, but then so are transistors, and we mostly get by on very simplified models that ignore most of the detail as insignificant.

And so it is with trannies; the full equivalent circuit of a transformer is daunting, but for most applications we can consider it to be a sort of gearbox.  While it is called a "600 ohm to 600 ohm" trannie the primary will actually reflect whatever impedance you connect to the secondary.

If you hang a 10k resistor across the secondary of the trannie then for AC signals the primary will also look like a 10k resistor.

The ones with tapped/split secondaries are perfect for fully isolated in-built, balanced, DI output, driven from an emitter/cathode follower.

To expand a little on my previous comments;

Larger trannies like this are often specified with a maximum level of "0dB", "+10dB", etc.  For a 600 ohm circuit this is implicitly dBm - with respect to one milliwatt in 600 ohms.  After a bit of number-crunching this comes to 0.7746V or 774.6mV (across 600 ohms) - close enough to Line/Aux level.

(for reasons I won't go into) the bandwidth of a transformer gets wider as the drive level is reduced, so the measured bandwidth will be significantly wider at -10dBm than at 0dBm, or again than at +10dBm.

A tranny is the wrong kind of component for a very hi-Z input.  A trannie basically operates on current, and what a piezo has is lots of voltage but at a very high source impedance (it's effectively a large voltage generating small capacitor) so current is just what we don't have.  Piezo's are much more electro-static than electro-magnetic (like a speaker), all about voltage, not current (unlike a transformer).

Nominally a FET would be the ideal device for sensing this signal voltage, but as a bit of math above demonstrates a FET can actually be bettered by BJT's in a "bootstrap" arrangement using exactly 100% feedback (or more like 99.98% in reality, strictly the Alpha of the transistor).
If you say theory and practice don't agree you haven't applied enough theory.


<what is the goal you are trying to achieve with the help of a transformer?>

I guess it's kind of hard to describe my 'goal' because it seems I got a little confused with all the options for preamps-boosters-di boxes available today. My original idea, after building Senor Phatt's Phabbtone preamp, was to adapt it so it could be used with an acoustic guitar (reference the latest discussion under the "Phabbtone" thread on this forum). But, not just any acoustic-electric guitar, but one fitted with a 'wood block' sound board pickup, aka piezo pickup. This type of pickup, as I now know, wants to see an input impedance of about 1 M to 3.3 M. 

So, I guess I thought that trying to convert the preamp to a preamp/di box made sense and a few of the designs I found use transformers in conjunction with the FETs or Op Amps. I presumed that this type of device was better when using a transformer to adjust the impedance? And that is how I arrived at this place where I'm talking about a DIY preamp plus DI box. I know we talked about using a buffer which would take the place of matching Z with a jfet or IC chip. However, I have always been intrigued with transformers carrying guitar signals and it just drove my imagination to seek that as a remedy rather than a FET buffer. I am also starting to build a Roly designed GroMo aka Gronk o Matic. I will pass along what works and what doesn't work so good as time and parts allow.  :tu:

thanks amigos