Magnum 18 Alternator disaster fix

[Damien Walker 246]

My C175 is fitted with a Magnum 18 (original KT17 threw a rod) and after arriving at a friend's field with the tractor in the trailer, I tried to start it to go and do the job I had come to do. Clunk! Engine rotated a few degrees and came to a dead stop.....something was stopping it from rotating. Removing the flywheel showed that the alternator had lost a magnet or two which had dropped into the works jamming any further rotation...pulverising the magnets in the process. Problem: those magnets are not available as spares (well, I couldn't source any) and the only option available from Kohler was a complete new flywheel at great expense.

 

Repair was possible however because luckily, I have a very well equipped workshop and so here's what I did (see photos):

 

You can buy very powerful Neodymium magnets in all shapes and sizes (though I couldn't find the crescent shape required by this alternator). So, I machined up a Tuffnol  ring that mimicked the original magnets and then  milled recesses into it to accommodate the magnets. The magnets were glued in place using 24Hr epoxy and then the whole assembly was glued into the flywheel. The Tuffnol ring means the inner face of the flywheel is smooth...nothing can drop off and jam any more. The magnets are positioned with alternating poles, which means that the generating frequency will be much higher, due to there being many more magnets. I probably should have grouped them into groups of 6 of the same polarity (there's 36 in total), but I didn't think of that at the time. My approach was to simply fill the existing space with however many magnets would fit!

 

The result is a more powerful alternator which may even be too powerful for the regulator as the ammeter always shows significant positive charge. I'm on the same battery (and regulator) some 9 years later. though..and it is still working well.

[8ntruck 7,016]

Interesting.

[Tuneup 1,433]

You had plenty of fun times there, I'm sure. Wonderful job.

[gwest_ca-(File Mod) 11,047]

Wondering if 2 regulators wired in parallel would handle the current?

Left and right AC terminals would need to be matched on identical regulators since one AC side is tied to the DC output.

 

Garry

[Damien Walker 246]

I'm an electronics engineer by profession (so should find this sort of thing easy!) but as I don't know what is actually inside a Kohler regulator, a solution is not so clear. Your two in parallel idea does make a lot of sense however. I also really should do some measurements to see what voltage and current it is now capable of producing.

 

One major problem is that being a permanent magnet generator, these machines produce power continuously, unlike a road vehicle alternator where the regulator controls the field current and so can control the output between full and zero as required. These clearly require a different sort of regulator. On that basis, doubling up on the regulator then would probably only increase the current capacity of the regulation system (ie would protect a single regulator from burnout) but it wouldn't protect the battery....it could actually allow more current into the battery due to reduced combined regulator internal resistance.

 

If you have any knowledge of the regulator internal circuitry, I'd be very interested to know how they operate!

[R. L. Addison 299]

I have a flywheel with missing magnets also, and like you I don't like the thought of buying one I am going to have to do something. I have access to a Bridgeport but no

index, so I'm on the lookout for a used flywheel at this point. I too looked to find magnets, but had same luck.

[Damien Walker 246]

 

39 minutes ago, R. L. Addison said:

I have a flywheel with missing magnets also, and like you I don't like the thought of buying one I am going to have to do something. I have access to a Bridgeport but no

index, so I'm on the lookout for a used flywheel at this point. I too looked to find magnets, but had same luck.

 

If you have access to a Bridgeport it should be easy. Indexing doesn't need to be very accurate, particularly if you use the same magnets and arrange the poles in groups of six. Just work out the circumferential spacing between whatever magnets you use and mark with pencil or use dividers to set the next position. All you'll need then is an angle plate arrangement and a clamping bolt. I used a Tuffnol blank (which I happened to have to hand), roughed out with a hand router and trammel and then turned in the lathe to size. I then used a wooden blank (surplus kitchen worktop!) to act as a mounting mandrel for the indexer which I also turned to size on the lathe. I can't lay my hands on the couple of magnets I had left over just at the moment to check their size, but I think they were probably 20mm x 10mm x 5mm....I'll keep looking and let you know if I find them.

[R. L. Addison 299]

It is a shelf engine at this point, so I think I'll just keep an eye out for another flywheel to buy used.

[Evguy 556]

I know this is an old thread, but the same failure has occurred on recent mower purchase. I figured the flywheel would probably be hard to find, especially since my hydro pumps are driven off flywheel face. 
I decided on a small alternator using a pulley that wasn’t being used. 
Mounting doesn’t seem to be too difficult, but a heat shield will be added to deflect heat away from alternator.

[Damien Walker 246]

Hi evguy, I'll be interested to hear how you get on...there's not much room available but that looks like a very neat unit....do you have the spec for it please? I run an electric sprayer on one of my horses and capacity of the stock alternator is a bit marginal.

 

The flywheel is easyish to to get though: couple of bolts to remove your pump (I'm guessing you can let it dangle off the pipes) and then you have remove the cowl.....that's quite a few 3/8 (I think) bolts around the periphery and the tin work just pulls off leaving the flywheel and cooling fan exposed. Single bolt/nut in the middle and you will almost certainly need some kind of puller to get the flywheel off because they usually fit onto a tapered shaft. The magnets are in a recess in the back of the fly wheel.

 

You could make up the necessary mounting using a router. Use a trammel to cut out the ring and the route the slots...it would be easier if you could set up a jig but it is quite an easy job really...just take it one step at a time!

 

Good luck either way!

[Evguy 556]

The alternator is rated 20 amps output.

 I found a preowned unit on feebay.

I could see that all the magnets are all gone, probably damaged the stator as well. 
I’m mounting the alternator on rear right side, behind the cylinder. I can get some pictures as I progress. 
I have sorted the brackets and belt, just a few holes and wiring and should be good to go.

 

[Damien Walker 246]

I think the standard alternator is 15A, I shall have to look it up, so 20A would be a handy increase.

 

It occurs to me too that it would be best to 3D print a magnet support ring these days. It's too big for my printer but I could do the CAD work and prepare the .stl print file if anyone is interested....could print it in sections I suppose.. 

[Evguy 556]

Today I put the after market alternator on to replace exploded magneto. I made a heat shield to deflect heat from cylinder head. Wiring was straight forward, connecting up to original regulator module where old magneto was connected.I did upsize the wire to handle expected 20 amps.

I tested and it didn’t work 😡 , found a bad fuse, once replaced all was good, battery is charging.

Cost wasn’t bad, used alternator 58. Belt 12.

I had stuff to make shield, spacers, wiring.

My battery wasn’t too discharged and I saw 12.7 amps charge and battery went from 12.4 to 13.6 pretty quickly. 

[Damien Walker 246]

Neat job evguy!

 

Hopefully there will be enough airflow past the cylinder with your arrangement...looks as though there should be, but it might be worth measuring some temperatures after a long run.

 

I'm guessing that it is a permanent magnet alternator, just as the original Kohler is...and so hopefully the stock regulator will work but I really don't understand how these regulators function  as they can't control how much the generator generates....only how much of it you use. That to me means 'resistance', except that high current and resistance together implies dissipated power and a lot of generated heat.

 

 

 

[Evguy 556]

I checked the airflow, it was highest at to top, where I had trimmed shield around alternator. Yes, permanent magnet, looks like small version of original stator setup.

 I did learn the regulator output is zero if B+ isn’t present. I had read that if battery is too low it won’t charge. My guess is some diodes to rectify the AC, then maybe a power transistor and regulator circuit. It’s the same principle as a standard power supply I assume. I know its mounted in air stream from flywheel for cooling.
 

[Damien Walker 246]

You are correct: if the battery is too low (or no battery at all!), the regulator doesn't work, presumably due to the lack of electrical bias on the regulator circuit. Your guess about the circuit components is probably correct, but therein lies the problem. The average (simple) voltage regulator circuit works by diverting some of the generated power into the load and the remainder is returned to earth. When more power needed by the load, less is returned to earth and v.v.. The problem then is that if you have charged the battery (no power required) ALL the generated power has to be directed to earth via the regulator circuit which means is has to be pretty hefty to handle it....much heat being the consequence! A 'proper' field energised alternator does not suffer from this problem as it only has to cope with the (small) field current and turn it up and down with the demand on the secondary side of the generator..

 

I think the original regulator is pretty amazing as a result (or it could be that I have no idea about voltage regulators!). These days you'd probably use a MOSFET transistor for low internal resistance but I'm not too sure they were readily available when this circuit was designed. I'd love to see the internal circuit diagram!

[Evguy 556]

 think the original regulator is pretty amazing as a result (or it could be that I have no idea about voltage regulators!). These days you'd probably use a MOSFET transistor for low internal resistance but I'm not too sure they were readily available when this circuit was designed. I'd love to see the internal circuit diagram!
 

 

I would like to see a diagram of the regulator as well!

 I’m tempted to take a spare and run some bench testing, checking currents and voltage as well as heat from the regulator. 
For now, as long as it gets the job done, I’m happy with it.