Old gibson magnets

[EFK]

Is there any way to determine the Alnico grade of old Gibson magnets? I have a pile of 50s P90s as well as a PAF and a few late 60s/early 70s T-tops, and aside from the T-tops which I'm pretty sure are all A5 short mags, I'd love to know what the other mags are. There are no 'real' markings on them except one of the P90s has a single black magic marker line on the middle of each end (both ends of each mag) and the weakest set (won't charge over @ 450 mT) has two thin little marker lines on the outside (N side) of each. Other than that they are all over the map.

Is there any *predictable* inductance pattern, i.e if I use a single humbucker as a control and switch out only the mags, is there an accepted order of inductance which parallels Alnico grade? Example, for simplicity's sake: A2<(induct. less than)A3<A4<A5.

Anyone?

[Mike Sulzer]

Is there any way to determine the Alnico grade of old Gibson magnets? I have a pile of 50s P90s as well as a PAF and a few late 60s/early 70s T-tops, and aside from the T-tops which I'm pretty sure are all A5 short mags, I'd love to know what the other mags are. There are no 'real' markings on them except one of the P90s has a single black magic marker line on the middle of each end (both ends of each mag) and the weakest set (won't charge over @ 450 mT) has two thin little marker lines on the outside (N side) of each. Other than that they are all over the map.

Is there any *predictable* inductance pattern, i.e if I use a single humbucker as a control and switch out only the mags, is there an accepted order of inductance which parallels Alnico grade? Example, for simplicity's sake: A2<(induct. less than)A3<A4<A5.

Anyone?

I do not think you can do this with inductance. A fairly sensitive test for inductance changes is to look for changes in the resonant frequency. I just did this test with a humbucker with enough capacitance across it to set the frequency at about 5 KHz. There is no signifiant difference between magnet in/magnet out; so I do not think you can tell the differences between magnets.

Alnico has lower permeability than steel; so it is not surprising that it has little effect on inductance since the magnet is not inside the coil and is poorly coupled to the cores.

Maybe different alnicos have different conductivities. I have not looked into that.

[Possum]

...

The only true way to determine what they are is to have them destroyed by vaporizing them for a spectrum analysis, which is pointless because then the magnet is gone.

I came up with a method, but this is a very flawed method and not really reliable for anything. Get a single humbucker that you can slide magnets in and out of. Figure out to keep it at a precise temperature, any variation in temperature will make the tests invalid. You must measure the DC resistance of every measurement to make sure the pickup hasn't warmed up or cooled off. Charge all your magnets full. Get new, modern magnets of each grade and charge them full. Use your Extech LCR meter and take readings of AC resisance, and inductance at both test frequencies and write those down (series mode). This will give you some idea of what each grade's readings will look like. Then do the same for all your vintage magnets and match them up with your modern magnets.

Why this method is very flawed is that there is no standard for any alnico mix and the modern magnets you use will be different from each manufacturer, so that screws this all up. I have alnico bar magnets that are really really dark and some that are really bright, for instance. The other problem is that alnico grades like 2 and 4 give real similar readings, hard to tell them apart. Most vintage magnets are going to be closest to alnico 3, 4 and 2. 3 is fairly easy to spot because they are the weakest gauss. But those vintage magnet mixes were real sloppy and some of them are probably inbetween modern alnico recipes.

You might have fun screwing around trying this out but don't put any stock in how accurate it is because its not accurate at all, just a real rough method I came up with years ago and abandoned...

[JGundry]

You can find out what grade they are without destroying them by having a BH curve run of each of the magnets. A place that makes Alnico can do this and they can tell from the BH chart what Alnico grade the magnet is. The problem is each grade of Alnico can have a very wide variation and still be considered an A2 or A5 magnet.

I had BH curves run for several magnets. Below are the results.

Long magnet out of 52' Les Paul P-90 (A2)
Long PAF Magnet (A5)
Short PAF magnet (A2)
Short mystery magnet which was tumbled (I think it came from a Shaw?) (A2)

I also had a vintage Fender rod magnet tested (A5)

All of these magnets sounded different in the same pickup and they all had very different BH curve charts.

A magnet place can run the charts and compare them to their own chart of a particular grade of Alnico. They can tweak their own grade to come close to your vintage magnet by duplicating the BH curve. But the only way to absolutely nail it is for the place making the magnet to cut the magnet in half, grind down and polish the magnet so they can scan it. Some magnet places do this anyway as a control for their own mixes. I had this done with the short A2 PAF magnet with great results.

Obviously the place that runs the BH curve is going to expect a follow up order.

What I found most interesting about having BH charts made was how meaningless from a tonal standpoint giving a name to a particular grade of Alnico is when it comes to vintage magnets. They just varied too much. Probably due to different mixes by different magnet makers Gibson bought from.

I do know for a fact that Gibson made short magnets in both A5 and A2 as recently as 1968. I have seen the dated drawings Gibson drew up and they specified both A2 and A5.

[StarryNight]

It seems if you had a few vintage magnets you could test all of them out (sonically) and if you found one that was oh so sweet you could have that one destroyed and duplicated - it might be a good business strategy.

[belwar]

The only true way to determine what they are is to have them destroyed by vaporizing them for a spectrum analysis, which is pointless because then the magnet is gone.

Thats not totally true. Xray Spectroscopy is non destructive.. BUT the sample has to fit in a testing cup. The testing cup at the place we have used is smaller than the size of the pickup, so the part had to be hacked apart. Plus the Xray has an apeture. The place I used has a 5mm, 10mm, and 15 mm apeture. The 5mm is close enough that it can do a strat magnet. So the test it self is not destruction, but you have to destroy a part to fit it in the machine! :>

[EFK]

Why would they need to cut it in half to polish it? Couldn't they just polish up a little bit one of the already-ground edges and 'scan' that?

I think the mags w/ the black marker slashes on the end are A5 as they'll take a pretty stout charge, although A4 can charge up pretty well so that's a possibility too. And the really weak pair I suspect are A3, although I've found over time (with a lot of these old magnets) that they do not seem to take as high a charge as comparable modern-made mags.

[JGundry]

Why would they need to cut it in half to polish it? Couldn't they just polish up a little bit one of the already-ground edges and 'scan' that?

I think the mags w/ the black marker slashes on the end are A5 as they'll take a pretty stout charge, although A4 can charge up pretty well so that's a possibility too. And the really weak pair I suspect are A3, although I've found over time (with a lot of these old magnets) that they do not seem to take as high a charge as comparable modern-made mags.

They have to test the magnet in a vacuum and I believe the want the sensor perpendicular to the magnet surface. The surface they want to get a reading from needs to be free of oxide so they grind it down and polish it. Why they wanted to cut it in half I'm not sure. Maybe it was so it would fit in their vacuum chamber or maybe they wanted a spot to sample that was far away from the molded edge.

I thought the long magnet from my 52' Les Paul was probably an A3 because it did not hold a big charge and it sounded quite a bit different than modern A2. But it is an A2 magnet with the B end of the BH curve much different than what you would seen in a modern mix. Trying to judge them from the charge they will hold can pretty much only nail down A5 IMO. And even then if it is unoriented A5 the charge is going to be much lower than regular A5.

[EFK]

Well something else to think about - when you charge the magnet already installed in the pickup, already installed in the 'magnetic circuit,' it obviously takes on a higher charge than if charged individually and then inserted into the pickup. My question is, to anyone who has tried this (Possum I know you have because I read it in one of your posts quite a while back), does it hold this higher charge long-term [provided it is not removed] or over a period of time does it settle back to a "standard" charge? This could be useful for some of the weak A2/A3 if they actually hold the higher charge.

[Possum]

....

The BH curve thing isn't 100% accurate because as you noted the tolerances and what standard for identifying, plus vintage magnets were sloppy mixes. Mike a humbucker is a really poor choice to test for resonant peak freq because of the ton of metal in them, short coils, they have very low resonant frequncies compared to single coils and very slow curves at the peak. Especially if its handwound. A single coil would be more useful but there's not enough information there because you're only looking at one thing, with the inductance method you are looking at two frequencies. But the inductance method is only maybe 70% accurate if that. The Xray thing is pretty close but I was told by the lab I used that the most accurate method is by destruction vaporization. They gave me a choice of Xray or spectrum so I went the destructive method. I personally don't think just copying a formula is going to give you the same tone as a vintage magnet because I'm sure the method they were made with back then is different than the high efficiency methods used today, just a theory, you'd have to talk to an old timer magnet maker to know for sure I guess.

Yes I magnetized some buckers as a unit and initially there was a higher charge, but I never tested to see if it held. Obviously if what Jon says is true that Gibson magnetized buckers at some point, they don't stay highly charged. Probably because there isn't any high carbon steel in them to help hold it. There is an old myth that if you take a humbucker apart and put it back together it won't sound the same because the magnetic unit has been fooled with. Maybe there is some truth to that.....

[JGundry]

A magnet maker can tell the grade of Alnico by the BH curve for sure. But I think for the BH curve to work perfectly you need to have the curves run on the same test rig for all of the magnets. For the xray spectrograph to be fool proof I think you need to have the repro magnets made at the same place doing the scan. If magnet company does the scan they already have it calibrated as a control for their own in house mixes.

The old mixes varied but magnetics places even now will have maybe 3-4 or more variants of a standard A5 mix. The curves will be different for all of these but they are all A5 and they can tell by the BH curve which is which. A magnetics place will have a base mix of say A5 and they add to that in different proportions to create sub mixes. They will do a full heat of A5, A2 whatever and tweak that to make the sub mixes.

[Possum]

....

Makes sense. This is why I buy alot of different manufacturer's magnets, I have about five different versions of alnico 5, all sound different. I used to buy from Earth Panda but they don't seem interested in small orders anymore and since I lost my contact there I don't get emails answered. They made some A3 that is really dark and syrupy sounding, all the other A3 I have is much brighter. I have some A4 rough cast that holds a charge as high as most A5 does, yet has a nicer top end. Its nice to have so much variety but then it can be hard to get consistency over a long time period.

[JGundry]

I'm sure I will get some of this terminology wrong. But the way an engineer explained the sub mixes to me was like this. A magnetic circuit has a load point which can be calculated. The magnet has an energy point that you can find on the BH curve. Ideally you want to have a magnet that has the correct energy point for any given magnetic circuit. The sub mixes provide different energy points. Very often the sub mixes are tweaked even further for the customer to match up the energy point with the load point of their magnetic circuit.

[Possum]

...

Interesting. Bet they've never been presented with a humbucker circuit. 3 different steel alloys, 2 nonferrous but conductive enclosures, and a magnetic coil. Engineer's nightmare :-)

[JGundry]

Interesting. Bet they've never been presented with a humbucker circuit. 3 different steel alloys, 2 nonferrous but conductive enclosures, and a magnetic coil. Engineer's nightmare :-)

Actually I think they have. The place I talked to wanted me to send them a pickup to test. I didn't do it though. I think an engineer would think efficiency. But I'm not sure efficiency would translate into good tone. I wanted a clone of a magnet and wanted to focus on that. Gibson did use different sub mixes so maybe they went through this testing exercise with new designs? This place does have a history of working with guitar pickups.