# Opamp Design for Low Impedance Sources?

I am terribly confused ... I tried myself at building a low impedance guitar pickup some time ago.

### Quick LowZ Pickup Intro

Basically just two loops of aluminium with two magnets, a humbucking guitar pickup with a single loop per magnet.

The three tongues are the terminals, but the outer ones are connected so that there is only one output plus ground. The magnets poles are on the top and bottom, one of them is rotated 180 degrees. This results in the desired humbucking effect.

### The Plan

I hooked up the whole shebang to my audio interface (connected to the Gnd and Hot Pins of an XLR with the Cold shorted to ground) and lo and behold there actually is audio coming from this thing.

Problem is that the output is super quiet and I have to crank the gain to +30db before I can hear something. My plan was to design a simple opamp preamplifier to boost the signal as close to the source as possible.

Now I definitely am no electrical engineer, but this is not my first rodeo. So I looked up the basic inverting opamp circuit and hooked it up on a breadboard. More specifically I used the schematics from here, but modified it to give me about 35x gain (about +30db). Hooked the output up like before, signal to hot, gnd to gnd and cold. This is a TL072 for testing, I meant to switch that out for a ultra low noise opamp once this is working.

(If I remember correctly this is Figure 4 from the Nuts'n'Volts article, just to give you an impression of my breadboarding conditions, I just noticed that it doesn't even show the output connection)

### The Problem

Hum! Instead of sweet guitar sounds I get hum. Definitely no signal in there.

### What I Tried

For testing purposes I hooked up a poti to control the gain and found that the hum disappears at unity gain. The signal is audible again then.

Did a two stage design with a unity gain and a 35x gain stage. Still humming, no signal.

Checked the unity gain bandwidth thingy, but at 3 MHz for the TL072 35x gain shouldn't be a problem.

Tried a 1:35 transformer instead... But transformers are magic to me and I didn't manage to get anything from it.

### The Question

I suspect that it's the super low impedance of the pickup itself. Basically it's a short against ground. But this is where my electrical knowledge fails me.

TL;DR: What am I missing in designing my LowZ guitar pickup preamp?

• long leads on the breadboard will pick up noise Sep 21, 2018 at 15:05
• Maybe my head is on sideways this morning, but in the photo above, the pinout for the TL072 looks wrong on the breadboard. Like maybe you misplaced where you think pin 1 is.... also, power connections? Sep 21, 2018 at 15:11
• @ChrisKnudsen Pin 1 should be in the upper left in the picture, the pins being 1OUT, 1IN-, 1IN+, Vcc-; 2IN+, 2IN-, 2OUT, Vcc+ as per the datasheet. Power connectins -> fair enough ... I should probably take down the picture ... I took it some time during buildup or after teardown apparently.
– fho
Sep 21, 2018 at 15:14
• @jsotola Agreed, but it's more the "put your finger on the signal line" kind of hum. Deep and buzzing. Ground hum.
– fho
Sep 21, 2018 at 15:16
• Interesting design. My suggestion would be to look for preamp circuits designed specifically for ribbon microphones, which have a similar extremely low source impedance. You basically want to think of the pickup as a current source, not a voltage source, so "transimpedance amplifier" might be a useful search term. Sep 21, 2018 at 15:58

Apart from the TL074 requiring a minimum power supply of 10 volts to guarantee working correctly you are facing a big up-hill struggle with this idea. I wouldn't use a site that recommends running this chip (or the TL064) from a 9 volt battery.

Consider that a proper humbucker has maybe 1000 turns on each pup and this means that for a certain string vibration you'll get 1000 times the signal compared to a single turn.

Then, because the basic output signal is so low, you are "fighting" hum and noise from other sources (such as the basic wiring from your masterpiece to the pots and the jack socket). This hum/noise cannot be countered by the hum-bucking principle.

Added to this, you might use an op-amp with a moderate noise performance to keep hiss at bay for a normal humbucker but, for your masterpiece you'll need a noise performance that might be 1000 times better. A typical reasonable audio op-amp has a noise spectral density equivalent of about 10 nV per $$\\sqrt{Hz}\$$ but money won't buy one that is 1000 times better.

Forget this crazy idea and wind more turns and get more signal.

• I should add that this is by no means "my crazy idea". The basic design is sold by companies like Lace (Alumitones: goo.gl/images/ZdpBhn). They don't split the tongues in three, and basically use a current clamp on the shared part. Also there exist several old posts in DIY audio forums where people build this kind of pickup.
– fho
Sep 21, 2018 at 19:44
• Also: my plan was to put the opamp on a little pcb on the inside of the pickup, as close as possible to the source. Smd parts with the backside of the pcb being copper contacts to the tongues.
– fho
Sep 21, 2018 at 20:00
• It is your crazy idea unless you can demonstrate that it is the same as a marketed product/design. The devil is in the detail; I have read all your design details and if someone else proposed the same and told me they could sell hundreds a year to proven satisfied customers then I would suggest that someone is deluded. As I don’t have detailed design info on those that are currently available on the market nor any idea about customer opinion, I can’t comment other than to reiterate that your basic idea is flawed and crazy in the face of my answer. Sep 21, 2018 at 20:03
• No matter how close you place your opamp you can’t override its basic flaws in that it will produce too much background noise/hiss to be anything more than a curiosity. I’ve built pups by the way! Sep 21, 2018 at 20:05
• Andy is correct, your signal is buried in noise with this approach. You might try a transimpedance amplifier circuit to have a hope of any success. Start with a gain somewhere around 1000 - 10,000. Sep 21, 2018 at 21:02

The easiest, fastest, and cheapest solution to this problem is to copy what Lace did with its Alumitone series:

1. short the "tongues" together
2. magnetically couple a current transformer around the middle "tongue"

The current transformer boosts output voltage and output impedance, both of which make this pickup work better with existing audio gear.

You can proceed without the current transformer, but with much more effort.

You know magnetized strings induce a current in your coil. However, you've broken your coil by leaving gaps between the 3 "tongues" of metal. It worked with the XLR input because these are often DC coupled and low-z (by audio standards), so the impedance mismatch was tolerable.

There are 2 issues with the "broken coil" to solve:

1. need to close the loop with a (low-z input) circuit
2. you should ground the center tongue and use duplicate circuits for both outer tongues, if humbucking/common mode noise rejection is important to you

Amplifier suggestions:
With a traditional op-amp, you might be able to get away with something like a photodiode amplifier circuit. You may even be able to adapt the typical inverting op-amp circuit (Fig. 3 from the linked nutsvolts article), but make Rin 0.1Ω, and Rfeedback 5Ω or 10Ω to start. The other values in the circuit can remain unchanged. Definitely this should be DC coupled because otherwise the capacitor will be enormous - And it's probably best to use 2 9V batteries in a bipolar supply with true earth ground in the center, for headroom and op-amp stability reasons.

However, it may be worth exploring "current-sensing" op-amps like the LM3900 or LM13600, as their example circuits are often built with low-z sources in mind.

# summary

I really like these low-z pickup designs, but low-z pickups don't have a wealth of existing designs, DIY guides, and general "lore" like high-z pickups have. You need enough background to know which values to change and how much they should be changed. So don't be afraid of digging in to how transformers work. You can skim (or skip!) the calculus, but get a feel for things like: impedance, inductance, magnetic coupling, where the field lines flow, and Ohm's law.