Forum member MW magnet repair tool set

xschop

Gardenstater

Dust cap is cut off on this first one and VC is definitely rubbing across from the wires, which is a hopeful development actually. Darn smartphone screen is failing and I can't turn off the flash. Had to hold my finger over it to get any kind of a shot that shows it. I need to make another one of these knives that cuts in the CCW direction lol. The polymer layer is very hard and difficult to cut through. Best not to try.

xschop

Nice dissection skills..



I stopped by the shop to do some stuff today. I found a solid brass round that I'll be able to make 2ea 6500 magnet removal tools...

Gardenstater

(2) 1/8" wide strips of 0.008" card stock get tight when they get more than half way around the pole piece. Allows marking the basket about where pole piece is touching VC, pretty much directly opposite the negative voice coil wire. If it wasn't for the fact that magnet has shifted out of concentricity with pole piece I would probably not disassemble this one.

xschop

Why don't you just use the jig to pull the mag/PP off the base plate?

Center the PP first, then unscrew PP set screws and just barely loosen all 4 Baseplate screws and then pull jig by hand

Gardenstater

xschop wrote: »

Why don't you just use the jig to pull the mag/PP off the base plate?

Next step is to make a removal tool out of the 3" PVC pipe I got today. Will bore out the I.D. to 3.118" on my Taig Micro Lathe. Kind of a pain to set up but I don't want to chip the magnet. Driver is ready. 3/4" oak dowel with a 19mm six point impact socket for the hammer end.

xschop

Can you take off the threads leaving shank of about 1/2 length of 4 setscrews to baseplate? Then use some heat shrink for magnet protection while you pull by hand?

I'll work on a set too this weekend with the 6500 brass tool

Gardenstater

xschop wrote: »

Can you take off the threads leaving shank of about 1/2 length of 4 setscrews to baseplate? Then use some heat shrink for magnet protection while you pull by hand?

I'll work on a set too this weekend with the 6500 brass tool

If you don't mind my modifying the screws I could try that. I like that idea.

PS: Perhaps temporarily replacing set screws with 1/4-20 Nylon bolts for the magnet pulling operation might work also? Maybe your idea is better though. Definitely stronger.

xschop

Yeah, go ahead. Just note that when magnet resetting, they MUST be placed into Pole plate position and threaded ones placed back into base plate position as the Thread major diameter is in the calculation of magnet clearance when resetting magnet.

Gardenstater

Well I didn't have a whole lot of hope for this one. VC is cocked in the spider and is rubbing on the frame piece. Scuff marks on the VC in a few places but it was just barely touching. I'm thinking the only way to save this one may be to drill out the frame piece rivets and reposition slightly and/or cut the spider to VC bond and re-glue with a PVC plug gage inserted in the VC to maintain perpendicularity and concentricity to frame piece as adhesive sets up. Then epoxy frame piece on perfectly concentric.

On the plus side the PVC insert (for MW6500) and tools shown made smashing out the pole piece a one hit wonder. I'll ship it on with the jig since it nests inside and is easily removed.

xschop

See if you can torque coil by pulling on one side of the circumference and pushing on the other. They are very strong and can take this kind of force.

Gardenstater

I can use my thumb inside the VC to torque into a more perpendicular and concentric position but it takes a pretty large amount of force and, of course, it just returns to exactly where it started from. Wooden dowel balanced on the end of the VC pretty much shows the issue:



0.015" on R



0.084" on L



0.069" over 3.114" equals 1.27 deg. 1.27 deg over 0.25" = 0.0055" which pretty much consumes the VC to frame piece clearances with very little to spare.

I'm going to contact Polk CS and enquire as to whether or not is still covered under warranty 🤣

xschop

Double dare ya.
Surely factory defect.

xschop

Wuhan coil is only other option lol

Gardenstater

It's too late to turn back now lol. I will drill out the "rivets" (seem more like swagings?) and see what happens. I want to see how much room there would be to do you know what underneath the spider w/o removing VC.

xschop

Pontificate on my homie.
But save yourself grief and just remove coil anyways.

Gardenstater

How much VC to frame piece clearance is there? Well I inserted strips of 0.0035" paper all around without difficulty and then was able to put a strip of 0.007" card stock in on the side where scraping wants to occur. Maybe it can be trained over time lol.

No grief. This is a learning experience as is everything.

xschop

Was 0.010" on all other MWs

Gardenstater

xschop wrote: »

Was 0.010" on all other MWs

Is that diameter difference or radial clearance (I mean 1/2 the difference). It looks like I've got at least 0.014" dia. difference.

I've got (MW6500, Silver Basket, 09/19/79):

Pole Piece dia. = 0.995"
Pole Piece Ht. (not incl. base) = 0.840" max, 0.838" min.
Pole Piece base Ht. = 0.250"
Frame Piece dia. = 3.114" max, 3.109" min.
VC Former Ht. = approx. 1.07"
VC Former I.D. = 1.010"
VC O.D. = 1.055"
Magnet I.D. = 1.281 approx.
Magnet Ht. = 0.542"
Magnet O.D. =3.336" max, 3.334" min.

Clearly some sort of steel filled adhesive was used because all the charcoal grey adhesive flakes are sticking to the magnet. Maybe an epoxy similar to JB-Weld or 3M Structural #2214 one part Heat cured epoxy (assuming they used it before spider/cone/surround/voice coil were installed). If so, it shows epoxy can lose its bond with repeated temperature cycles etc.. The flakes are very thin (0.003") so significant clamping pressure was used.

xschop

What's 6500 pole piece shaft OD?

Gardenstater

xschop wrote: »

What's 6500 pole piece shaft OD?

I just added to above. It was 0.995"

xschop

Good to hear the magnet ID matches all other MWs, so the kit's centering dowel is good to go.

And you would need a replacement voice coil of ID 26mm. EBay has them at 35mm length too...bonus

Gardenstater

My VC ID is 25.6mm and length is only 27 mm approx. I guess what matters is the length of the wire coil section, along with the DCR and Inductance.

xschop

A 25.5mm VC is not enough gap clearance, the reason why I lathed off .020 from the factory 6502 pole piece when I used the 25.5mm ID JBL VC.

Gardenstater

xschop wrote: »

Good to hear the magnet ID matches all other MWs, so the kit's centering dowel is good to go.

Well.....I just tried the centering dowel, beautifully machined and knurled wow, which fits my pole piece shaft ( ) beautifully, but unfortunately it doesn't quite enter the hole in my magnet. Turns out the hole is not ground but is apparently as molded/sintered and isn't round. I did the best I can at measuring w/o a telescoping gage/micrometer and using my not non-magnetic hardened stainless steel Mitutoyo calipers and it looks like it is 1.2785 min, 1.280 max.

Gardenstater

Well I got the 2nd Silver Basket MW6500 (OEM in my 7B's purchased in 1979) de-glued and de-dust capped. First off, a little background on this one, which is that it passes the manual 3 fingers push test to listen/feel for any scraping sounds. But I knew there was an audible problem when playing very low frequency test tones even at only moderate volume levels. That's why I took this one, and the other one with the spider repair out of the speakers a couple years ago.

The Gorilla Max Strength Clear was not a *total* nightmare to remove. It was adhering tenaciously still, a couple of years later. I tried to do a peel test but the stuff adhered too well and every time I pulled it just broke the adhesive rather than the bond. I'm sure glad I didn't JB-Weld epoxy this one. You have to even get into the crack because there is a sort of a taper at the edge of both pole plates which allows some adhesive to penetrate.

So it was immediately apparent when looking at it that there were some scuff marks when the VC pulls back a fair distance.......probably on those low bass notes. I think the pics show it this time because I fooled my defective smartphone's camera into not using the flash by taking the pictures outside in sunlight. Next step is to try to smash the pole piece off of the magnet as done on the other one. I might try to center it up first to break the bond but I'm leery of that because of the SNAP that occurs all too easily when it is removed from the jig. That can damage the VC and former when it gets creased by being pinched between the framepiece plate and the pole piece shaft.

PS: My angled Xacto knife I made works fantastic. By the 2nd dust cap you will be cutting them flush to the cone almost and without fear. I recommend this.

Magnet is off center to the Pole Piece plate. Did they never center them or did mine shift.....



Scuff marks are pretty darn visible to the left of the former gap

xschop

Alot of the MWs magnets I encountered were not glued concentric from factory although they never shifted.

Gardenstater

xschop wrote: »

Alot of the MWs magnets I encountered were not glued concentric from factory although they never shifted.

It must not matter for the proper functioning of the magnet motor? I tend to believe they were not idiots. Sure would make it easier to visibly tell if there has been any shifting but they probably did not anticipate that would ever be a problem. They should have scuffed up the plating on the pole pieces to get a better bond. Even a house with a set back thermostat in the winter can represent an awful lot of thermal cycles over the years......

Gardenstater

Well this one also has a cocked voice coil but it isn't really rubbing at the moment on the frame piece. Gets pretty close though and there are signs that it has been scuffing. I hope this isn't occurring during my removal of the pole piece. How would we know? the marks on the inside that correspond in location with the scuffing on the voice coil wires is suspicious of that. Still Ohms out fine at 6.5 Ohms. VC is definitely crooked and not centered perfectly in frame piece however. Could be manufacturing issue. Could be old spiders and surrounds. No way to know for sure. The plot deepens.

Scuffs on wires



Suspicious crease caused by end of pole piece corresponds in location with scuffs on wires



Same cocked VC issue as other one

Kryten

Hmm, as the plot thickens, novices are getting nervous............