I have concluded my project of charging up magneto magnets and offer the following conclusions based on the methods I used:
-After 80 days since recharging, magnetic strength went down approx 1% stored with keepers; 4% without keepers.
-magnet strength is not affected by temperature (10-212 deg F)
-reverse charging the poles does not result in a weaker charge.
-tapping during charging does not help.
-it takes about 40 gauss to lift 1 lb with a T magnet
See the following discussions about how I did the study:
Many thanks to Dean Yoder for providing the magnets for this study.
Robert, thank you for conducting the tests and for sharing the data with the forum.
The state of magnetism is effected by temperature but it must get high enough to do so. The temperature in which permanent magnetism changes to induced magnetism is known as the Curie temperature. It would be interesting to know how close an engine that is running hot gets to the Curie temperature.
Another piece of data that would be interesting to know is a chart of magnetic strength versus charging voltage and current applied to a coil of a given coil (wire gage, core, number of turns, etc.)
What interests me was the strength with and without keepers. To those who subscribe to the keeper theory, magnets without keepers loose 4 times as much magnetism as those with keepers. To those who do not subscribe to the keeper theory there is little difference between 96% and 99%. I've seen guys on a Hit and Miss Engine forum say that they carefully slide a magnet off of the recharger onto the keeper and off of the keeper and into the magneto so it doesn't suddenly loose its magnetism. From Robert's study above, it sounds like the Hit and Miss guys are going to a lot of trouble for nothing.
I think before your magnets reached the Curie Temperature, your engine would be on fire. IIRC, CT is around 1500 degrees or more.
Robert, In your tests did you compare using separate coils with steel charging the magnets on the sides to coils wrapped around the magnet ends?
I store the spare magnets in pairs, glommed onto each other, like this: <>, N-S, S-N.
I figure that's as good as or maybe better than a keeper. I may be wrong.
Robert, the Gauss required to lift a 2 pound magnet is a very slippery number. I have 3 different meters to do direct Gauss measurements and the readings vary so much that I have never been able to establish a Gauss number for a 2 pound weight.
The measurements vary greatly and are much higher around the magnet pole edges than they are in the center.
A much smaller Gauss value will lift the 2 pound weight, if both contact surfaces are clean, smooth and polished.
I have also discovered that the 2 pound weight test will provide a useful Magneto, but that is a minimum and not even close to a maximum possible magnet charge.
Ken, You may very well be correct but I have seen the Curie temp for Neodymium magnets stated as 600 to 750F yet know smaller physical size magnets can sustain significant permanent loss of magnetization at just 175C, well below boiling.
Don't know how hot the magneto magnets could get in normal or adverse operation. I agree there is probably a lot of margin before the magnets suffer permanent reduction in magnetic field strength due to temperature but still would be interesting to see the data.
Well, they actually did lose 4 times as much strength without keepers, loosing from 1% to 4%. But we would never see that change in the magneto output. I should have mentioned that most of the "loss" occurred during the first day after recharging; after that, it didn't matter if they had keepers or not. But if the magnets are stored with a bunch of other metal parts, keepers might be a good idea.
I just did a no-no by removing the magnets from my hit&miss engine magneto to recharge them; I'll report the results shortly.
Robert, I would be interested in your methodology. My experiments with a gauss meter are something like James's. Even with one gauss meter I had a hard time getting accurate repeatability even on the same magnet. I wound up building a jig for aligning the meter and the magnet to get somewhat repeatable results, but I am sure even then my margin of error was close to four percent.
Paul, I did not use a charger that clamped to the side, which is more common method.
James, you are right in that the measurements can vary greatly. That's why I changed from my mechanical puller device to an electronic probe. And yes, I found the strength is the highest at the edges of the ends. The small probe gave reasonably repeatable measurements on the side of the magnets (as shown in the pic in the link). The figure of 40 gauss/lb was for the pull required to separate a keeper from a rough as delivered Ford magnet.
I also placed a square metal clamp used to hold the magnets to the flywheel on the side and measured the strength .030" above that. This is what the coils should typically see, and it was about 30% less then the magnet side strength (end strength being even higher).
Just took a bunch of pictures but my PC is doing some thing different gave up for now on posting them.
Weight tested one magnet that was setting ends down on a machined plate for about the past four years . The magnet has just a slight grind on the legs to shine it for a good contact with what I will call a flux bar while charging. A simple 17 wrap on each leg was used with a heavy duty battery hit about four 1/4 second times with a vintage floor switch
The magnet held 8 1/2 pounds before being pulled from a machined plate.
I will see if pictures post. Magnet on top of kettle was tested. Notice how just a little of the end was ground for contact with the flux bar at the end of the charger. The black coil was used the white one was a later try with no improvement.
Even a minor grind as shown in the pic will make a noticeable difference in its pulling strength. After recharging my magnets averaged 345 gauss, which using my 40 gauss/lb would hold 8.6 lbs. But this is with unfinished ends.
Tom, you are right; my testing methods would never approach 1% accuracy. I used 6 magnets (12 poles) for the study. I marked each magnet in the same spot on the side that was about where the center of the clamping plate would be. I placed the probe on the marks for the measurements. I then took the average for each set of measurements to calculate the change in the strengths. I didn't know how to do a proper statistical analysis of the data. I would be glad to send you (or anyone) my spreadsheet data.
In the end it is the magneto output that really matters. My test would indicate no change in performance of a freshly charged set used within one hour (maybe possible in the Ford factory) or a set that has been sitting around without keepers for a few months. The question remains why some magnets end up weak and do benefit greatly from a recharge.
Thank-you Robert for the extra info. I never tested the keeper/non-keeper situation because as I see it, the magnets on a T flywheel are unkept. (Is that a word?) Not only are they unkept, but they are held in tension (which seems to me to be like "unkept" on steroids) due to like poles being next to a each other.
My tests on polarity swapping jibe with yours.
Thanks again for posting this.
I agree with you Robert. until todays information I might know why! My setup pushing the coils up setting them close to one side of the magnet might charge the magnet more on one side then the other. using a round wind over a rectangular magnet end might put different charging around the edge of the magnet. You caught my interest so now I will machine some hard wood about 3/8" over magnet size each way, knock the sharp corners off and wind up a rectangular coil that four steel shims cold slide between the magnet and the coil. If it works I will post it.
In your last post - did you mean 175 F ?
Ooops, yep! 175F, below the boiling point of water. Thanks Bud.