EMG active pup specs

[erikbojerik]

hey folk

New to this forum and to winding, but I've been building instruments for myself for a few years. I see a few extended-range and multi-scale instruments coming down the pike in the next few years, so I'm starting to wade into this winding thing...slowly. I've read just about every post here, and I'm working on plans for a winder now.

My question: I know that active EMG pickups have some kind of buffer chip/circuit epoxied in with the magnet & coils to bring down the impedence. Has anyone here ever deconstructed one to see what's going on in there? What kind of chip/circuit they use for this?

Their spec sheets also hint at using low-gauss magnets...anyone know what's up with that? What's the idea behind it? Anyone ever measure one?

thanks for this great forum!

[mjmiller]

hey folk

My question: I know that active EMG pickups have some kind of buffer chip/circuit epoxied in with the magnet & coils to bring down the impedence. Has anyone here ever deconstructed one to see what's going on in there? What kind of chip/circuit they use for this?

Their spec sheets also hint at using low-gauss magnets...anyone know what's up with that? What's the idea behind it? Anyone ever measure one?

I'm far less knowledgeable about this stuff than a lot of the folks here so take everything I say with a grain of salt, but I think I can answer your question, or at least confuse you further ; )

The function of a buffer, if I'm not mistaken, is to provide a high input impedance to the source (the pickup, in this case) and a low output impedance to the load (the cable & whatever the cable's plugged into) The reason for using a buffer makes sense if you meditate on ohm's law long enough to understand some of it's ramifications. More on this later, unless I get sidetracked.

Gauss is the unit of measure for the strength of a magnetic field, so a low gauss magnet is... you guessed it, a weak magnet. Why use a weak magnet? Well, the stronger the magnetic field is where the strings intercect it, the more it will affect the vibration of the strings. In other words, weak magnets don't suck heap big tone. I think this is often referred to as "magnetic drag" They (weak magnets) also don't produce as strong a magnetic field where the coil intercepts it, which makes for a lower amplitude signal in the coil.

This goes hand in hand with their low impedance coils, which presumably have relatively few windings, relatively thick wire, or (most likely) a combination of the two. This has the effect of (correct me someone, if I'm wrong here) creating a coil with lower inductance and capacitance, which decreases AC impedance, broadening the frequency response in both the bass and treble directions. It also makes for fewer winding to intercept our (low gauss) magnetic field which makes for... you guess it, a lower amplitude signal in the coil.

Combine these effects and you get a pickup that has very little effect on the strings, a super wide (crisp sounding with powerful lows) frequency response, but jack squat for output level (with diminished ability to drive an instrument cable and subsequent amplifier stage.)

(is it the Heisenberg uncertainty principle that implies that you cannot measure anything without affecting it? These kinds of pickups try to interfere with the strings as little as possible - to have the smallest possible "observer footprint" note: some (many) people like the "warm" vintage sound - less highs, more mids, but if you're Kirk Hammett... bring on the wide band)


Oh, yeah... low output level...

What to do, what to do?

Oh, yeah, buffers.
Don't pay any attention to what I'm about to say - I really don't know what I'm talking about.
The buffer, in this case, is a way to protect the sensitive operation of the pickup from the great unknown of whatever it is you plug into your jack. If they didn't buffer the pickup and you plugged it into a 50' unshielded cable that went to a fuzz face, the results would not be pretty, ruining all their hard work. Placing a buffering circuit in there gives the pickup a consistent, known load that their engineers can count on when coming up with really sensitive, low impact pickups.
So: my guess is that they use a buffering scheme to reduce loading on the pickups (to preserve that extra tone they worked so hard to design into them) and possibly (probably?)apply gain to get the singal level up to standard pickup levels. They may or may not combine these in one stage...
A jfet (junction field effect transistor)in common source mode with a bit of gain and plenty of NFB would, if I'm not mistaken (but I usually am ; ) provide a nice high input impedance and enough gain, along with a low enough output impedance to achieve their goals
They might use an op amp
They probably use an op amp
Hopefully they use a *good* op amp but an ad797 is too much to hope for. I'd settle for an ne5532
Or they might use a BJT (bipolar junction transistor) as an emitter follower(re: buffer stage), followed by one hooked up in common emmiter (re: gain stage), follwed by *another* emitter follower... (this is unlikely)
The list goes on... a darlignton pair, a single BJT in common source...
Depending on the exact specs of their pickup and exactly what they wanted to achieve, they would design an appropriate buffering circuit.

In other words, I have NO idea what is in that buffer circuit, but I know what it's there for, and it's probably *really simple*

Now I'm just TOTALLY rambling... trying to puzzle this stuff out myself as I explain it.

Honestly, just wait for David Schwab to come along and correct me - he actually knows what he's talking about. And check out his web site - he makes pickups that imho are voiced better than EMGs (you can listen to sound samples)

Hope I created more questions for you than I answered ; )
(that's when you know you're learning and thinking)

Michael Miller

[David King]

As to what preamp chip they're using, we know it's going to be reasonably low noise and low current draw and it's going to cost not more than a dime or two. My guess is 062, 072 or 082 series. They probably change up every 10 years or so. Current noise specs are 96dB on their best humbuckers and have been for quite a while. Audiophiles they are not.

[erikbojerik]

Not being an electrical engineer, I don't quite understand how all that results in a broader frequency response, but I'll buy that. It seems to be one of the defining features of EMGs.

Aside from that, I follow the logic up to the "low output" part of the discussion. I guess the buffer also works as an amplifier...?? But how exactly do you think they can boost the output without also boosting the noise? EMGs have a rep for being low-noise....

David, I think you have a sign error there...the noise specs on a 89 (tri-coil) are MINUS 90s....-94dB for the single coil, -99dB for the humbucker. Those seem like pretty good noise figures to me, but I'm very new to this, so bear with me if I'm off base...

[Joe Gwinn]

Not being an electrical engineer, I don't quite understand how all that results in a broader frequency response, but I'll buy that. It seems to be one of the defining features of EMGs.

Aside from that, I follow the logic up to the "low output" part of the discussion. I guess the buffer also works as an amplifier...?? But how exactly do you think they can boost the output without also boosting the noise? EMGs have a rep for being low-noise....

The trick is to use a low-output coil (fewer turns, so lower inductance and higher resonant frequency) and use the buffer amplifier to raise the output to the typical range for passive pickups having five or ten times as many turns on the coils. Such coils are quite flat in the audio range, so all tone shaping is done by filter components built into the buffer amp. The EMG website actually explains this, but not very clearly.

Part of the low output is due to use of small magnets with little string pull, reducing wolf tones.

David, I think you have a sign error there...the noise specs on a 89 (tri-coil) are MINUS 90s....-94dB for the single coil, -99dB for the humbucker. Those seem like pretty good noise figures to me, but I'm very new to this, so bear with me if I'm off base...

It's true that these are minus, but in common usage this is understood but not stated.

In guitar service, having noise 90 db down means that the pickup is noiseless in practice. Only in a dead quiet recording studio would there be even a chance to detect the noise in such a device.

[David Schwab]

I tried to comment to this the other day but the forum was not responding...

EMG uses an op amp of some kind ... it has their name on it, so I guess it's something made for them.

I was always under the impression that they used low impedance (low output) coils, but it seems they actually use pretty normal high impedance coils.

Their magnets aren't any smaller than normal, at least for the 81. The EMG-40DC I have home has as much pull as the pickups I'm making with normal sized ceramic magnets, maybe even more.

I'm basing all this on some photos someone send me once. They dissected a non working 81 to see what was inside. Looks like a fairly normal humbucker with a alnico magnet, twin blades, and a bunch of wire on the bobbins. The coils were wax potted before the pickup was encapsulated. The preamp was on the baseplate, which was made from a PCB.

The one thing about EMG's is they use a differential input to the buffer and run each coil separately. They also use the buffer to mismatch the response of the coils, and probably some EQ also. They call this "tone modeling".

I used to think they were pretty hi-fi bass pickups until I started making my own. They do have a very consistent tone though. If you like their sound, you always get it. You don't get much else though.

(as a disclaimer I'd like to point out that I have no interest in copying EMG, and merely found the photos to be interesting from a "oh that's what's in there" basis.)

[David Schwab]

Not being an electrical engineer, I don't quite understand how all that results in a broader frequency response, but I'll buy that. It seems to be one of the defining features of EMGs.

The more turns of wire you use, the louder the pickup, but also the higher the inductance.. after a while you start to lose high end, and low end too. The coil starts to behave like a band pass filter.

Let's use Alembic as an example. I think we can all agree they make very wide response pickups. They use about 1500 turns of 40 gauge wire wound right on a ceramic magnet. They have a matching dummy coil with no core.

These pickups have very little output, but very wide frequency response. You just boost up the level to normal high impedance levels, and you have very clean and clear sound.

EMG's don't really have that wide a response. They are about what you get with a buffered high impedance pickup (which is what they are).

[mjmiller]

Not being an electrical engineer, I don't quite understand how all that results in a broader frequency response, but I'll buy that. It seems to be one of the defining features of EMGs.

I'm not an Electrical Engineer either, but if you want to know about the principles that give any pickup its frequency response, I'll give it a better whack:

There is one thing that determines a pickup's frequency response: Impedance. That one thing is determined by 3 things: DC resistance, inductance, and capacitance. DC resistance is a property of the wire in the coil - its material and thickness, Inductance is a property of the coil itself, and capacitance in this case is known as "parasitic" capacitance - in an "Assume the Cow is Spherical" world, it would not be there, but in the real world it is. All those turns of wire with insulator between them form a capacitor, whether we like it or not, and one end of this capacitor is connected to ground, because one end of the coil is connected to ground Your coil is an inductor AND it is also a capacitor connected across itself.
Bear with me, this will all make sense. (I hope)

Inductors impede AC - we say they have "AC impedance" and this impedance increases with frequency.

Inductor = resists treble / admits lows

Capacitive circuits have infinite DC resistance (which is why we use them to block DC between amplifier stages) but their impedance DECREASES with frequency, allowing signal to pass through. The higher the capacitance, the more easily it passes lows.

Capacitance = resists bass / admits highs

BUT remember, this parasitic capacitance is actually connecting our signal to ground, so the inductance resists highs and the capacitance allows the highs to go down the drain, so they work together.

Now these two combined effects only mean something (meaning you can only do the math) when there's a DC resistance involved. If a coil has a lot of DC resistance, it becomes far, far, easier for treble to go down the drain
If a coil has fairly little DC resistance, not as much of it takes the easy road down the drain, because there's an easy road to the amp, too.

Put these things together and you discover that the less inductance, the less capacitance, and the less DC resistance your coil has, the flatter its frequency response will be.

Does that make sense? Sorry, I was going to explain better in my last post, but I ran out of steam, I guess

I guess the buffer also works as an amplifier...?? But how exactly do you think they can boost the output without also boosting the noise? EMGs have a rep for being low-noise....

Yes, I'm sure the buffer is also an amplifier and they can't boost the output without boosting the noise. Another benefit of low impedance pickups is that they are naturally, inherently, lower noise devices than high impedance pickups. They also pick up 60hz buzz just as weakly as they pick up guitar strings so if you hook them up to an op amp and do a little filtering with negative feedback, you get a nicely sized signal and not much noise. If you boost jack squat for noise by 10 times, you still have jack squat for noise

Schwab: you posted while I was posting so I may be proven wrong on some points before I even posted
lol
Great info - thanks. I was thinking of buffering my pups and I wasn't sure what it would do - now I have a better idea Great pics, too



Also:
Joe - this is the second time I've heard the term "Wolf tone" to describe something other than the most deformed 5th in Just temperment.
Obviously they're different things... but I don't understand the mechanism of the wolf tone you're referring to... is it a sharpening of the fundamental due to magnetic pull? Does it alter the harmonics of the string... effectively creating a magnetic bessel curve by tightening the string only along a certain portion of its length? Wow... this is reminding me of work I've done with Shakuhachi bore shaping... (japanese end blown flute)

I think I just answered my question

Michael Miller

[Sock Puppet]

Wolf tones

A wolf tone is defined as an interval in a scale that's noticeably out of tune because of the limitations of the scale in use.

"Stratitis" is another term for the discordance created when a guitar pickup magnet is adjusted too close to the string and particularly afects the thicker more magnetic strings when played up the neck.

Imagine the magnet(s) creating an floating anchor point part way along the string, strength varying as the string vibrates... then look at the patterns you can get with a harmonograph.

Better still, try it on your guitar.

S.

[Sock Puppet]

Just saw the "Joe" part,

Apologies Joe, need new glasses.

S.

[erikbojerik]

Thanks guys, and thanks also for the photos David.

I imagine there's more on that circuit board than just an op amp, eh? Any SMDs floating around on there that you know of?

Would this buffer/amp circuit be similar in principal to a typical on-board active preamp?

[David Schwab]

I imagine there's more on that circuit board than just an op amp, eh? Any SMDs floating around on there that you know of?[/QUOTE]

Yes, you can see the SMD's in the last photo. I'm actually surprised the op amp isn't one.

Would this buffer/amp circuit be similar in principal to a typical on-board active preamp?

Yeah, more-or-less. As I said they claim to do tone shaping and noise reduction at the preamp.

You don't need all that though. Look up some JFET or op amp buffers online. There are a bunch. Wind a low output pickup and use a buffer.

[madialex]

Hey DS, don't forget you can still have low impedance with a full coil of wire. Look close at the wire, it looks pretty hefty to me. Maybe 38 AWG, that would make a very low impedance coil I would think.

[madialex]

Never mind, I think I mis-read the post.

[David Schwab]

Hey DS, don't forget you can still have low impedance with a full coil of wire. Look close at the wire, it looks pretty hefty to me. Maybe 38 AWG, that would make a very low impedance coil I would think.

Yeah, it looks a little heavy, but I think it's the wax. Larger diameter does give lower resistance.

This is what the guy wrote. He's from Poland so his English was a bit spotty. I forgot he wrote specs... What would .06 mm be in AWG?

EMG81

Magnet: Ceramic (cuting) 56x3x13mm
Wire: 0,06mm (PE)
Core: 54x3x12,5mm (silicon steel?), solid steel.
Coil: 4,18KOhm (one coil), wax potted, aprox. 5500-6000 turns, h=7,5mm
Bobbin: 64x13x9mm (or with "tube legs" 12,2mm)


Coils conection:


-opamp--------^^^^^^^^---------^^^^^^^^^^---------opamp+
ground

IC unknown, marked as EMG001

Here's another view.