I found these links on another forum. It claims to the secret EMG pickup preamp circuit.
What do you have say about this.
I have heard that opamp in EMG pickups is something like EMG01. But circuit say Lm4250. Do you think its fake.
And what these loops on output from ground means.
It matches the concept that David Schwab has posted in the past, but with the component values shown the gain will not be equal for the two coils, which is suspicious.
The circuit diagram does not match the circuit of the pictured circuit board.
Are the two pickup coils identical?
If yes, the circuit diagram isn't quite right, but is probably close. I would expect a balanced differential amplifier merged with a DC voltage divider to set the DC operating point (~ground) at about one half the battery voltage, allowing operation off a single 9 volt battery.
If no, the circuit still doesn't quite make sense because the gain of the plus channel will depend too much on the AC resistance of that coil. One would expect a series resistor like the 30 Kohm resistor in the minus path.
I wonder if the schem is correct about it being a LM4250
Checking the data sheet it seems that low power is the only thing a LM4250 has going on, it's a general purpose OpAmp.
(not a low noise job)
It's a shame whoever took that pic of the PCB didn't flip it over and give us a look at the other side, might be more SMT over there.
I know where this schematic came from, and it's the same source that sent me pictures of the inside of an EMG 81 (see below). I got the photos over two years ago, and schematic this past July.
Yes, this is supposed to be accurate. I haven't traced it from the photos I have though, but the same people did both (see below). It took them a while to trace the PCB from inside the pickup. The only thing they didn't know was the exact op amp, since it had no number on it, besides a private EMG part number.
The other circuit from the booster switch board is different. That photo is not where this schematic came from. Here's the board attached.
I'd guess the unbalanced coils are part of the tone modeling thing they do. One of their older pickups, the EMG-58 was too noisy (hummed too much), so they redesigned it as the EMG-85 which is quieter. They also have very fine brass screening acting as a Faraday cage for shielding.
The source that did the schematic said the coils were matched, but I think that's hard to say since they had to destroy the pickups to find out.
But the specs were listed as follows:
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)
IC unknown, marked as EMG001
Electric shematic will be able in a few days.
Last edited by David Schwab; 09-08-2008 at 07:53 AM.
National Semiconductor's datasheet examples suggest
that EMG started with component values from NS'
gain-of-5 differential amp schematic, and set the current
bias to ~1 microamp to optimize the noise power figure.
A 10k output buffer resistor is also prominent throughtout
the datasheet specifications.
That lack of a 30k resistor on the non-inverting pin
looks like an error instead of a design decision.
This circuit is certainly not balanced: inv, gain = 5, non-inv gain = 6. So, yes, maybe a resistor in series with the non-inv input has been left out. It would have to have a value of 13.7K to lower 6 to 5. (The lower leg of the voltage divider is the two 137K resistors in parallel.)
The inv. input input impedance is 30K; this is kind of low for a high impedance coil. And the input impedance the other coil sees is 68.5K. This does not make sense.
From the EMG site (italics added)The inv. input input impedance is 30K; this is kind of low for a high impedance coil. And the input impedance the other coil sees is 68.5K. This does not make sense.
Also remember that the two coils are not wired in series, so loading it down like that will make it darker sounding.Tone Modeling
One of the most important aspects governing the tone of a pickup is the resonant frequency. EMG Pickups use "Impedance Modeling" to manipulate the two coils. This innovation allows us to shape a mix of the reactive slope and resonance from each of the two coils. The idea is to achieve a complex mixture of each coils phase and frequency response resulting in a richer tone from the pickup. This means the sound is vibrantly alive with more harmonics than from conventional passive pickups. EMG Pickups like the EMG-S, EMG-SA, and the EMG-60 use this technique to its fullest, while the EMG-81 uses modeling in only a small way. Modeling might work well for a single coil pickup, but not for a design such as the EMG-ACS Acoustic Sound Hole Pickup. As each pickup design is approached differently, it all depends on the final result we're searching for.
A 6/5 gain ratio at low frequencies will not do a good job of canceling the hum from magnetic fields. EMGs do cancel hum well, and so I think something is missing. Perhaps the coils are different as someone mentioned.
It is true that actively adding the outputs of two different resonant circuits can produce responses that cannot be achieved with two identical resonant circuits. If the two coils are the same, they would have the same resonant frequency, but different Qs. If the coils are different, both parameters could be different.
The differences that can be achieved with such an active circuit are greater than can be achieved by putting the two different coils in series. If the cable capacitance dominates over the inter-winding capacitance, as it can, then the two inductors pretty much just add in series.
I think every EMG made has the coils in parallel. That's one thing I don't like about the EMG-P.
I have an old Bartolini Hi-A pickup with one coil being intentionally 1K different from the other, and it's dead quiet. I had it installed in one of my '74 Ric basses, and along with total copper foil shielding of the bass, I was able to remove the string ground with no noise whatsoever.
I make some bass pickups with similarly large offsets with no hum problems.
EMG's have the entire pickup in a Faraday cage made from very fine brass screen. As I mentioned, some EMG models that had very mismatched coil circuits and did indeed hum. The EMG-58 was one of those. We used to use them in the American Showster AS-57 tailfin guitars, and they were the best sounding EMG humbucker, but they were a tad noisy. They replaced them with the 85, which is quieter, but lacks the harmonics of its predecessor.
Either way, take a look at the photo I posted of the circuit board inside the 81. That's where that schematic came from. The IC and one surface mount part is missing, but you could probably trace the circuit assuming they didn't use a double sided board. I've been meaning to do it but haven't had the time.
Here's an EMG SA, which is also very mismatched:
Sure, you can use different numbers of turns on the two coils. Cancellation of magnetic hum using two coils requires that the two coils have the same overall sensitivity. Since the sensitivity depends on both the number of turns and the magnetic core, you can get good cancellation with different number of turns if the cores are different in a complementary way. Simply making the number of turns different with two identical cores decreases the degree of cancellation. This is how it is in theory and how it works in practice. It might still appear to be good enough most of the time, but in an environment with really high stray magnetic fields, you need really good cancellation.
Shielding with a conducting material is for electric fields. Both kinds of hum are a problem, and each has a different solution.
So is anyone who wants to make clone of them.
Perfect matching is not required to obtain adequate hum cancellation. Here is a simple numerical example:I have an old Bartolini Hi-A pickup with one coil being intentionally 1K different from the other, and it's dead quiet. I had it installed in one of my '74 Ric basses, and along with total copper foil shielding of the bass, I was able to remove the string ground with no noise whatsoever.
I make some bass pickups with similarly large offsets with no hum problems.
Consider each coil as two voltage sources in series, the music source (Vm) and the hum source (Vh). The coils also differ in that they don't have the same number of turns and thus induced voltages, the ratio being called k, where k=1 if the coils are identical. All of k will be allocated to coil 2. The coils differ in that the algebraic signs of Vm and Vh are the same in coil 1, and opposite in coil 2, so:
Vc1=Vh+Vm but Vc2=k(Vh-Vm), the difference being the sign and the factor k.
A differential amplifier amplifies the difference between the two inputs, so (assuming a gain of one for now) we get
Vout= Vc1-Vc2= Vh+Vm-k(Vh-Vm)= Vh(1-k)+(1+k)Vm
If k=1, we get double the music voltage and complete cancellation of the hum voltage. But k is never exactly unity, so what is the effect of mismatch?
The cancellation ratio is (1-k)/(1+k). How large can k be to achieve 20 dB cancellation? For 20 dB, the cancellation ratio is one tenth, so 0.1=(1-k)/(1+k), and k= 0.8181, and 1-k=0.182. This means that if the coils are geometrically similar and differ in turns count by 18% or less, 20 dB or better cancellation will be achieved.
For 30 dB cancellation, 1-k= 6%.
The usual with surface mount cards is to use two-sided boards, perhaps with components on both sides, unless the circuit is so simple that it's cheaper to implement crossovers with wire bridges and drilled holes. So, I'd photograph both sides of the circuit board, and be sure to identify one corner with a magic marker such that it's visible in both photos.Either way, take a look at the photo I posted of the circuit board inside the 81. That's where that schematic came from. The IC and one surface mount part is missing, but you could probably trace the circuit assuming they didn't use a double sided board. I've been meaning to do it but haven't had the time.
The EMG-P is in parallel, but not the humbuckers. But I meant to say they aren't in series as you would expect.
Now that I think about it, the other side of the board is the back of the pickup, so there are no parts...The usual with surface mount cards is to use two-sided boards, perhaps with components on both sides, unless the circuit is so simple that it's cheaper to implement crossovers with wire bridges and drilled holes. So, I'd photograph both sides of the circuit board, and be sure to identify one corner with a magic marker such that it's visible in both photos.
Yes, I agree with your numbers. Another way of saying it is that you can approach 6 db better than implied by 1-k. That is, 1-k = .1 gives 25.58 db rather than 20 db, and 1-k = .01 gives 45.98 db rather than 40 db, while k-1=.31622 gives 14.53 db rather than 10 db, but only 4.53 db better than implied by 1-k.
But how much is good enough? Do you plan for the average environment, or the really bad one that might not be too common, but is still out there?
As a rule of thumb, 10 dB is quite noticable, 20 dB is pretty good (and would probably make the hum inaudible in a noisy venue), 30 dB is almost perfect, and 40 dB isn't any better than 30 dB in practice.But how much is good enough? Do you plan for the average environment, or the really bad one that might not be too common, but is still out there?
When people intentionally mismatch coils, the difference in turns count seems to be of order 10%, which would allow for very good cancellation: 20 Log10[(1-0.90)/(1+0.90)]= -25.6 dB.
This assumes that the electrostatic shielding is adequate. Perfectly matched coils are still susceptible to electrostatic pickup if not shielded.
clearly it does.
And Deb has a hair trigger for kicking people off lately...
Because they don't sound the same with metal covers, which is why people started taking the covers off. You can get better noise reduction without changing the tone by using shielded plastic covers.It is interesting that most humbucker users do not use a cover. But the base plate, even though it covers just the bottom, does have a significant shielding effect. So it could be worse.
Traditional humbucker parts are really quite archaic. Covers are popular these days because of the way they look.
Achieving these levels of cancellation is another matter entirely. Although matching turns counts to within 10% doesn't seem that hard to me.
Was it something you said?Thank you for writing that: one of my pet peeves. In fact, writing that and arguing that it is correct contributed to me getting kicked off another forum.
What was their counterargument?
It's a tradeoff to be sure. Most covers do muffle the sound.It is interesting that most humbucker users do not use a cover. But the base plate, even though it covers just the bottom, does have a significant shielding effect. So it could be worse.
But making very long rod magnets out of shorter rod magnets has little effect. (magnetized along the long dimension) Do you remember what we were discussing?
It took Deb several years to kick me off, but yeah, the final event really was hair trigger, and kind of surprised me.
I think the discussion was about the effect of chrome plating on the tone of a pickup.
This photo should refresh your memory.
There are currently 2 users browsing this thread. (0 members and 2 guests)