DiMarzio PAF Master, Analysis and Review

[antigua]

www.dimarzio.com/pickups/humbuckers/paf-master-bridge
www.dimarzio.com/pickups/humbuckers/paf-master-neck




I haven't installed this set yet, but I needed a PAF-type set with distressed covers for a Zemaitis that came stock with very high output distress covered humbuckers.

I was curious to see if the "PAF Master" actually had real PAF parts, or if it's an "Air Bucker" like the 36th Anniversary , and bizarrely, the neck is an "Air Bucker" with an AlNiCo 5 magnet, but the bridge is not an "Air Bucker", it contains a metal keeper and an AlNiCo 4 bar. See pictures below. The "Air Bucker" patent can be seen here www.google.com/patents/US5399802 . 

The result of this unusual difference between the neck and bridge PAF Masters is that even though the bridge has a weaker AlNiCo 4 magnet, it actually delivers more flux to the strings due the fact that it has a metal keeper around the screws and the magnet makes contact with the slugs. The neck pickup has a stronger AlNiCo 5 magnet, but less flux at the strings due to the "air bucker" setup, which simply means the magnet does not make contact with the screws or pole pieces, and is held in place by plastic washers, thereby decreasing the flux intensity in the screws, slugs and ultimately the strings. The neck PAF Master also contains additional steel slugs inside the bobbins, which increase the inductance. The patent DiMarzio filed for these in-between slugs can be seen here www.google.com/patents/US5908998

I like all the innovation involved in these moderate output pickups, but it's all around very strange that a pickup set called the "PAF Master" is so steeped in non-PAF spec appointments. My skeptical side is telling me that maybe DiMarzio is really just trying to make a pickup that is a little more tricky to duplicate.

As far as RLC "hotness" goes, the neck and bridge are very close  together. The bridge pickups has only 200mH inductance over the neck, and neck pickup actually has a loaded peak resonance that is higher than the neck, though the difference is only 30Hz.


DiMarzio PAF MAster Bridge DP261N8

Series resistance: 7.39k ohms
screws (green & white): 3.72k ohms
slugs (black & red): 3.67 ohms

Series inductance: 4.508H @ 120Hz
screws (green & white): 1.948H
slugs (black & red): 1.843H

Gauss: 340G slug, 385G screw

Unloaded:               V: 3.5dB f: 5.06kHz (black)
screws (green & white): V: 1.9dB f: 6.73kHz (green)
slugs (black & red):    V: 3.3dB f: 6.66kHz (gray)
Loaded (200k & 470pF):  V: 1.5dB f: 2.47kHz (red)




DiMarzio PAF Master Neck DP260N8

Series resistance:      7.63k ohms
screws (green & white): 3.83k ohms
slugs (black & red):    3.82k ohms

Series inductance:      4.296H @ 120Hz
screws (green & white): 1.856H
slugs (black & red):    1.824H

Gauss: 215G slug, 295G screw

Unloaded:               V: 2.1dB f: 5.42kHz (black)
screws (green & white): V:-2.5dB f: 6.81kHz (green)
slugs (black & red):    V: 2.2dB f: 7.46kHz (gray)
Loaded (200k & 470pF):  V: 1.0dB f: 2.44kHz (red)



The heavy damping of the screw coil is unusual, and still being investigated.



. 







Nickel Cover Effects

The DiMarzio "distressed" nicked cover appears to cause slightly higher losses than usual. Unloaded, the peak drops be 3.4dB with the cover, where 3dB has been typical with chrome plated brass covers, and with load, it drops 1.2dB, where as 1dB is typical. These differences are superficial, though, as the minimum difference that is  detectable by the human ear is only 1dB.  Therefore, taking the cover off all together changes the level just enough to be perceived, but the difference between this distressed nickel and typical chrome plated brass is too small to hear. 















This is the PAF Master neck pickup, a patented "air bucker", maybe more appropriately "plastic / wax bucker":





Here is the bridge pickup with an AlNiCo 4 bar, a metal keeper around the screws, but no wood or plastic spacer, as is usually expected with PAF clone humbuckers:




Here is another demonstration that the slug coil, and only the coil, effects the screw coil in a way that the screw coil does not effect the screw coil. In the plots below, the coils are still side by side with the driver coil over both of them, with all metal removed. Each coil is then measured my itself, and it can be seen that the screw coil has the double dip, but the slug coil does not, and just to reiterate, if the slug coil is moved away, the screw coil no longer shows that double dip:

[stratotarts]

Considering the eddy current losses in the screw coil side, is it possible that there is a short in the coil windings? Are the slugs the same size on both bobbins?

[antigua]

Oct 11, 2016 7:11:58 GMT -5 stratotarts said:

Considering the eddy current losses in the screw coil side, is it possible that there is a short in the coil windings? Are the slugs the same size on both bobbins?

I bet you're right about it being a short. I measured it the first time and it didn't do this, but subsequently it did it every time, so I attributed this first plot to some kind of error on my part:





I checked the DC resistance and inductance of the pickup before making the bode plots, and the inductance now reads 1.855H versus 1.856H prior. That 1mH drop might be attributable to margin of error, though. The DC resistance reads 3.83k ohms, identical to what it was when I first measured.

If this is a short, it goes to show that there can be a large drop in Q factor without greatly effecting inductance or resistance, which makes this a type of flaw that can only be easily seen the Q factor. The bode plot is an obvious way to see it, but looking at the Extech, it  shows a Q factor or 0.366 at 120Hz and 1.771Q at 1kHz for the presumably shorted screw coil, versus 0.3618 at 120Hz and 2.010 at 1kHz for the good slug coil, so it appears that the Extech's Q measurement can also indicate a short. 

I've ordered a new neck pickup without a cover, so I can compare the two. It doesn't look like the short lowers the Q factor of the loaded series plot that much, but it's now a defective pickup as far as I'm concerned. 

[antigua]

So I received a new PAF Master neck, and it shows the same damping specifically on the screw coil. This is very strange, especially how I didn't get that damping on the earlier plot. I'll mess with them and see what I can find out.   


, 

[antigua]

It seems that there are at least two factors that causing the screw coil to show heavy damping.

The first is the particular cover that came with that PAF Master neck. I swapped the covers on the neck and the bridge, and the bridge cover on top of the neck pickup showed less damping than the original cover. 




The second issue is that the screw coil seems to have some intrinsic damping, as you can see below that the screw coils of both the original and replacement  PAF Master neck pickups have much lower Q factors than the slug coil. They also have strange double peaks:




Here's what happens with the screw coil if the covers are reversed, so that the holes are over the slug coil, but every thing else is the same. You can see that strangely the solid half of the cover, without the screw holes, causes less damping than the when the cover is oriented correctly, and in both orientations, the "good" cover from the bridge pickup shows less damping:




To see if the screws are to blame, I removed the stock screws, which have been intentionally rusted to look "aged", and swapped in screws from a Gibson 57 Classic, and they caused the damping to decrease only slightly. I also measured it with no screws at all, and surprisingly, that second resonance is actually higher in amplitude than the first peak, when there are no screws in the coil:



Removing the magnet decreased the inductance by a small amount, but otherwise had no effect on the curve. 

Here is what happens when those embedded slugs are removed from the coil. I had to drill holes into the top of the bobbin in order to push the slugs out. I'll take pictures of that tomorrows, since these things are rarely seen by anyone not in the DiMarzio factory. They increase the inductance quite a bit, so I'll get numbers on that, but as far as the plot goes, it just caused that second peak to become even more extreme:




I'll continue to do more tinkering to see what else can be found out. I know that the 4 conductor cable can cause double peaks, but it's not grounded in this testing, so there's no reason it should be doing that. 

[stratotarts]

Strange about the double peaks. It reminds me of some results I got when I put a solid cover on the A3B1 humbucker. At first I got an extra peak in the response:

Then I put kapton tape on the pole slugs and it went away:

[antigua]

What the tape on the slugs to prevent them from being grounded?

My next step is to bypass the the 4 conductor cable and tap into the coils directly. I don't like having to do that since it can wreck the coil if that lead wire gets yanked out somehow, but there's a mystery and it must be solved.

[stratotarts]

I put the kapton tape in to prevent the slugs from contacting the cover. My theory is that the slugs formed a bypass circuit through two slugs, the baseplate and the cover. I'm not sure now why it happened, I was just happy to eliminate it because I wanted to move on to another experimental step. I suggest trying it on the screws and slugs. However I admit that the effect that you are seeing is stronger. It is a real mystery.

[antigua]

I've determined that the double peak is caused by the adjacent slug coil being where it is. I've seen this sort of thing before, but I had forgotten about it. The strange thing is that the slug coil doesn't exhibit the same behavior in relation to the screw coil. 

For all the tests in this post, those embedded slugs that DiMarzio puts in many of their pickups have been removed, eliminating them as a potential cause.

Here's a plot with the screw coil connected, and the slug coil is not connected, and in a nut shell, the height of that lower peak comes down to how closely the slug coil couples with the driver and screw coils. The green line, which shows the lower frequency peak most prominently, is where the driver is over the slug coil, though the screw coil still the one being measured, causing the most dramatic amount of coupling and interference from the slug coil. 






In this test, I had the screws removed from the screw coil, since they had appeared to be unrelated to the cause of the dual peaks, and it turns out they were not specifically related. If I replace the screws, the same shapes emerge, but at lower amplitudes.


Here's a picture of the driver coil perpendicular to the screw coil with the slug coil out of the way (gray plot line)






Now we see that the slug coil does not experience any "double resonance". The plot below represents a role reversal, and a lot of strange peaks result, but none have that double resonance. In other words, the slug coil can mess with the screw coil, but the screws can't mess with the slug coil:




And here is proof that it's the slug coil that causes the dual peaks in the screw coil, and not the slugs themselves:




The single peak is the slugs along side the screw coil. You can see it makes now difference. The dual peak plot is the slug and screw coil side by side, neither coils has screws or slugs in them.

Here's pics of this test:

Slugs, no slug coil:


Slug coil, no slugs

[antigua]

OK I'm starting a new post because I hit upon a fairly clear conclusion, although why this happens is a mystery. 

You have the PAF Master neck humbucker's slug and screw coils side by side, with all the metal removed, so there is just plastic and copper coil wire. The test driver coil is laid on top of them like this:




The three plot lines below show something rather interesting: when the oscilloscope is hooked up to the screw coil, and the slug coil is beside the screw coil, but not hooked up, the screw coil shows two different peaks; it's own peak, and the slug coils peak (black line):




When the slug coil is hooked up, there is no double peak, you only see the slug coil's peak (red line).

When the screw coil is hooked up, but the slug coil is moved aside, the screw coil now has only one peak, it's own (green line). 

Important clue: when the slug coil is a closed circuit (has it's leads twisted together), this effect disappears, and it's as if the slug coil has been moved aside. The effect only occurs when the slug coil is open (leads are not touching). 

So there is clearly some sort of mutual resonance that occurs between the in-circuit screw coil and the open slug coil. 

Not only do I not know why this is happening, but I did't know such a thing was possible. It looks to me like capacitve coupling is at the heart of it, since the effect only occurs when that slug coil is open. It's also very strange that the screw coil does not have the same effect upon the slug coil. 

[newey]

That is odd, but unless you were able to replicate this effect with another PAF, it may just be an anomaly with this particular pickup. But you may be right about capacitative coupling, the mystery then becomes why it doesn't work the other way around.

[antigua]

That's a good point. I have seen these double peaks before, usually when I oriented the driver coil incorrectly, so I think I viewed it at experimental error, where as in this case the effect happens even with the driver coil where it is supposed to be, so it might have  been that putting the driver coil in the wrong place merely emphasized the effect in the other humbucker. I'll see how closely I can duplicate this effect with other pickups. 

I don't think this PAF Master is defective, because I bought a second PAF Master neck on the belief that the first might have had a series coil short that would make it sound dull, but they show more or less the same pattern (pink and green lines):

[newey]

You could move the driver coil slightly away from the slug coil, and see if the anomalous second peak is reduced, or disappears entirely. We could then confidently conclude that it is an experimental factor with this pickup. Perhaps the two coils here are ever-so-slightly closer together, causing this coupling effect.

[antigua]

I'll do a plot with the two coils slightly removed, and I do expect that if they are proximate, but not touching, you will still see the effect, just to a lesser degree.


I don't think this effect is caused by the test setup either way, though.  That would be to suggest that the driver coil facilitates the coupling of the two coils. I can test that theory by testing again with the driver coil above, but farther away from both coils. 

[newey]

That would be to suggest that the driver coil facilitates the coupling of the two coils.

No, that wasn't what I was thinking. My thought was that, if the particular coils in this pickup were capacitively coupled, perhaps by being just a fraction of a mm closer than a "usual PAF", then placing the test rig directly over the one coil may induce a signal in the "unconnected" coil- which is then picked up in the other coil via the theorized coupling. And, that by moving the driver, one could verify the effect was dependent on the driver positioning, since the induced signal would vary. But this is little more than spitballing on my part, since none of this explains why it shouldn't work the other way around.

[antigua]

Oct 14, 2016 20:08:44 GMT -5 newey said:

That would be to suggest that the driver coil facilitates the coupling of the two coils.

No, that wasn't what I was thinking. My thought was that, if the particular coils in this pickup were capacitively coupled, perhaps by being just a fraction of a mm closer than a "usual PAF", then placing the test rig directly over the one coil may induce a signal in the "unconnected" coil- which is then picked up in the other coil via the theorized coupling. And, that by moving the driver, one could verify the effect was dependent on the driver positioning, since the induced signal would vary. But this is little more than spitballing on my part, since none of this explains why it shouldn't work the other way around.

That's a good point about the driver coil exciting the interference in the secondary coil. 

Although I didn't work out all the math involved, the underlying cause of the asymmetry is that the slug coil has a much higher capacitance than the screw coil,

This plot is created with all the metal parts removed from the coils, and the coils are nowhere near each other:



Both of these coils have near identical inductance, 924mH for the slug and 918mH for the screw coil, so for the slug coil to have a resonant peak that is almost 3kHz lower  than the screw coil requires that it have roughly an additional 120pF capacitance. That's an extreme amount of capacitance for a humbucker coil in general, especially one with a DC resistance of only 3.7k ohms with 42 AWG. 

The topic is covered more here, guitarnuts2.proboards.com/thread/7769/damping-caused-unused-splitting-humbucker , but the coils couple like a transformer, and whether the secondary is open or closed changes the RLC resonant equation of the primary coil. The probable reason why you don't see this with most other PAF type pickups is because they usually don't have dramatically imbalanced capacitances. 

I don't know why the slug coil would have  such a high capacitance, but the science says its caused by closer proximity of the wire, so that might be caused by higher wind tension or a thinner insulation. Given the fact that unwanted capacitance is usually regarded as a bad thing, I'd say it could be considered a manufacturing defect. 

[newey]

the underlying cause of the asymmetry is that the slug coil has a much higher capacitance than the screw coil,

Aha! I'd consider that "case closed", then.

[antigua]

I just noticed something interesting. DiMarzio says in their "tech talk" copy:

"We chose Alnico 4 for the PAF® Master Bridge Model because of its high iron content and energy coefficient. AL4 lends more body and attack to the sound than you might expect from a pickup with relatively low output voltage and DC resistance specs." www.dimarzio.com/pickups/humbuckers/paf-master-bridge

I naturally assumed this was all marketing talk, but then I noticed:

DiMarzio PAF MAster Bridge DP261N8
Series resistance: 7.39k ohms
Series inductance: 4.508H @ 120Hz

DiMarzio PAF Master Neck DP260N8
Series resistance: 7.63k ohms
Series inductance: 4.296H @ 120Hz


the bridge actually does have a higher inductance for a lower DC resistance. Maybe they aren't lying about this. I'm not really sure that A4 does have a higher iron content though, I can't find a reliable reference to back this up. They also say "relatively low output voltage", but if the inductance is higher, so too will be the output voltage, but I'll give them the benefit of the doubt and assume they misspoke.

[reTrEaD]

They also say "relatively low output voltage", but if the inductance is higher, so too will be the output voltage, but I'll give them the benefit of the doubt and assume they misspoke.

I agree that it sounds a bit misleading. But they did say "relatively". So the question is, low output voltage relative to what?

[antigua]

It should be relatively high output voltage. If you increase the inductance, but the output voltage is still low, then you've somehow accomplished nothing. It's like saying you've created a higher octane fuel, but your engine is still weak.