Disconnecting Battery of MEP-002A or MEP-003A While Running: Damage?

[Triple Jim]

I just spent about half an hour entering search strings like disconnect battery while running, regulator damage, no battery, disconnect battery, but didn't turn up what I was looking for. It seems to me I've read that disconnecting the 24V battery while an MEP-002A or MEP-003A is running is bad. Maybe I read that it can damage the regulator, or maybe is was the control circuitry, or both. Can anyone fill me in please?

I'm asking because I'm working on a replacement 24V battery charging regulator for these generators, and would like to collect all the background information I can. Any other suggestions are welcome as well.

Thanks for the help.

 

[Isaac-1]

My understanding of the dangers is that it can damage the regulator or the output stator if the regulator goes out since it is a shunt style regulator and a permanent magnet alternator, but this is going from memory

 

[Wildchild467]

The shut off solenoid is electric and I would think disconnecting it would cause voltage fluctuations. I would think the batteries provide a constant voltage buffer to all the electronics. I don't understand why you would want to disconnect the battery when it is running though. Why not design your regulator and then plug it in and see if the battery voltage comes up to the required voltage... or place a small amp meter inline. This is your Guiana pig project, yes?

Also, what is wrong with the original charging system? Why would your regulator be better than the original? I don't mean that to sound negative, I am just wondering. you never know until you ask

 

[Triple Jim]

Thanks Ike.

The shut off solenoid is electric and I would think disconnecting it would cause voltage fluctuations. I would think the batteries provide a constant voltage buffer to all the electronics. I don't understand why you would want to disconnect the battery when it is running though. Why not design your regulator and then plug it in and see if the battery voltage comes up to the required voltage... or place a small amp meter inline. This is your Guiana pig project, yes?

Also, what is wrong with the original charging system? Why would your regulator be better than the original? I don't mean that to sound negative, I am just wondering. you never know until you ask

Your questions are appreciated. No Guinea pigs will be harmed if I can help it. I've been asked by several people on the board to make 24V regulators available. The reasons given to me include the high price and high failure rate of the originals. If I'm going to design one, I want to know of any shortcomings of the originals so I can try to build one that doesn't have them.

I don't know why anyone would want to disconnect the battery with the generator running, but I remember at least one guy who was trying to jump start his MEP-002A/3A. If my memory is correct, the battery was very dead, acting about like not having one at all, and when the owner disconnected the jumper cables, problems surfaced. I just don't remember the details.

 

[Isaac-1]

I have often read of members buying these generators without batteries then jump starting them and removing the cables, as to the reason there is demand for these is the insane pricing the typical replacements sell for often $150+ for a beat up used regulator and $250+ for new, or at least "like new".

 

[doghead]

I have experienced issues with the control circuit when starting a new to me unit with a slave cable, and disconnecting it while running.

A capacitor on a fuel pump shorted and needed to be removed to get the control circuit to function again.

There were no batteries connected, iirc.

 

[Triple Jim]

A capacitor on a fuel pump shorted and needed to be removed to get the control circuit to function again.

Good information, thank you. That sounds like every small capacitor across the 24V system was effectively trying to filter the very non-filtered DC from the charging circuit output. Little capacitors made for noise filtering on DC aren't designed to handle a ton of ripple, and the one on the fuel pump was just the first one to overheat and blow. Speculation on my part, of course, but that's one logical explanation.

 

[Keith_J]

The charging system is just like that of a motorcycle, permanent magnet field with stationary coils. As such, it generates much higher voltage than needed, the regulator shuts off current when voltage gets too high. This happens many times per revolution, generating much heat which is why regulators for these types have generous cooling.

when there is no battery, voltages can surge high for many times per second and also surge low as magnets swing outside the optimum. This will damage the regulator.

just don't do it,

 

[Triple Jim]

Or make the regulator with that in mind. Funny you should mention motorcycle regulators. Here's one I make:

http://www.lakeserv.net/kaw/regulator/index.html

 

[Keith_J]

Interesting, Jim. Are you ever going to make a 24 volt regulator for MEP 002/003a with a super cap or at least a polypropylene ?

 

[Crawdaddy]

Interesting, Jim. Are you ever going to make a 24 volt regulator for MEP 002/003a with a super cap or at least a polypropylene ?

That's the whole purpose of this thread. He's trying to engineer a better one that doesn't have the issues the stock piece has. And when it comes out, I'm game for one!

 

[Speddmon]

Jim,

If you decide to make a regulator, and I pray to God that you will. Please make it relatively inexpensive, beefier than the OEM and maybe at least a 10 amp regulator. If the stator would handle it, which I think it might, it would allow for faster charging. I read somewhere that with the 6 amp charging system currently in use, you need something like 4 hours of run time to replenish the drain just from starting the generator.

 

[Triple Jim]

No plans for a super capacitor that at this time. The capacitor I'll use will be an electrolytic that's big enough to filter the output of the diode bridge so that the control system is happy if the battery gets disconnected. I'm not sure what type of polypropylene capacitor you're referring to.

 

[Keith_J]

Bear in mind the stop solenoid has a good inductive capacity and has a snubber diode so it is biased. I only know this from repair of an 003a that had been reverse connected, causing one of the Canon plug wires to become unsoldered.

The solenoid may make for some interesting system dynamics.

 

[Triple Jim]

Jim,

If you decide to make a regulator, and I pray to God that you will.

OK, that got me laughing, thanks.

Please make it relatively inexpensive, beefier than the OEM and maybe at least a 10 amp regulator. If the stator would handle it, which I think it might, it would allow for faster charging. I read somewhere that with the 6 amp charging system currently in use, you need something like 4 hours of run time to replenish the drain just from starting the generator.

Price is a always a big concern of mine. The trouble of course, is that as far as products go, the demand for the things I make is very small, so there is no sending it to China to get them made for $25, for example. I make everything myself, with office help from my wife, and assembly help from my son and daughter when they're home from college. It's a tough balance between keeping the price down and having the item worth making. On the bright side, all my circuit boards are made in the US, and sheet metal parts I have laser cut are done locally. I'll probably be able to offer the regulator for a bit less than the $159 of my motorcycle version, because it won't need the hand made (by me) stainless steel housing.

The permanent magnet alternators I've worked with are actually the limiting component of the charging system. The rectifier/regulator converts the AC to DC, and if the battery needs charging, all the output goes straight out to it. Only when the output voltage reaches the set point does the regulator come into play and shunt unneeded current to ground.

The 12V motorcycle regulator I make uses 45 amp Schottky diodes in the rectifier bridge for low voltage drop and low heat production, and the HexFET is rated at 64 amps, so there's really nothing in it that would limit output current. I plan to use similar parts for the 24V regulator.

I've had my 003A in service for a couple years now, and most of its running is either a monthly 15 minute test, or a short outage of an hour or so. I've had only a couple outages that were several hours long. I don't find that the battery runs down from lack of charging. Of course if it's very cold out and the engine has to be cranked for a long time, the charging problem you mentioned could come into play.

 

[steelypip]

The biggest changes I've seen in motorcycle regulators in recent years is the adoption of MOSFETs in place of the older SCRs. The documentation and I've seen suggests that the MOSFETs allow much finer control of current flow and shunt a lot less of the usable power to ground in the normal 'batteries mostly charged' case.

 

[Triple Jim]

Yes, the HexFET I use is a variety of MOSFET. I find its main advantage is very low voltage drop, like a couple tenths of a volt. This leaves a little more for the battery, but more importantly it doesn't make much heat.

 

[Isaac-1]

Jim, have you considered a design that might also work on the MEP-701a? I think it also uses a permanent magnet alternator.

Ike

 

[Triple Jim]

I had not, but only because I don't have a 701A here. Do you know the rated output current from the alternator? If so, I'd need a failed regulator so I can design a drop-in replacement that mounts and connects like the original.

 

[Crawdaddy]

I had not, but only because I don't have a 701A here. Do you know the rated output current from the alternator? If so, I'd need a failed regulator so I can design a drop-in replacement that mounts and connects like the original.

I have a failed 016B regulator that I can send. The 016B/701A regulator is potted with potting compound that I recall people saying is very tough, so I don't know what all you can glean from it, but it's just taking up space in the shop.

I installed a battery tender on the 16b that seems to work fine.