Converting a MEP-831A to standard generator

[OverkillTASF]

As I stare down the barrel of possibly another MEP-831A with a faulty inverter, amidst thoughts of tearing into the inverter and rectifying the alternator output for high voltage DC use in a solar setup, I wonder... What would it take to make this work more like a MEP-016D...

The first and most obvious problem for me is that I have no idea the signals that the speed controller and governor actuator make use of... But I know the speed controller depends on input from the PMA or inverter, both of which would of course be removed in such a conversion... So, engine shutdown (low oil, etc) becomes the first hurdle. The governor on the engine, as part of the Yanmar engine, can easily handle keeping the engine running at the 3600 RPM necessary for 60hZ power generation... But how do we shut it off... Shutting off the fuel pumps clearly isn't enough... I assume we have to actually move the throttle assembly...

Owners of MEP-016D's, or people just generally familiar with small diesel engine generators... What hopefully simple part enables engine shutdown? Seems like something would mount easily on the existing standard L70 bracket and press down on the throttle level in the same way as the governor actuator... A simple solenoid seems like it would do the trick, but I'm having trouble finding what looks like a suitable part that costs less than $200...

 

[DieselAddict]

There is a knob on the side of the engine that you turn then move to the "shutdown" position. All you would have to do is adjust the mechanical governor so that when the knob was all the way up (run) it would be about 3650 rpm and when you want to shut it down you would just push the knob down and the engine will stop. No solenoid needed unless you wanted it to shut down with the toggle switch on the front.

 

[OverkillTASF]

Oh, I know how to manually shut it down. But I'm thinking about preserving most of the other functionality, most crucial of which would be low oil pressure shut down. So for that, it would need to be able to shut itself down. And not by seizing the piston. Would also be nice to make sure it is resettable so that remote start/stop is still possible.

Thinking about it, it seems likely that the governor actuator is simply a pulse width modulated signal... something like +24VDC is all the way open, -24VDC is all the way closed, 0 is in the middle, and PWM for positions in between. I think my multimeter has a setting that will read pulse counts, may fiddle with it at lunch today. Would be nice to be able to continue to use the actuator!

Edit: Fiddling with the actuator yields a range of 0-24VDC. Not sure if it's PWM because my cheap multimeter seems to be pretending that there is no pulsing happening, but it does seem to set a voltage for a specific position. I think it relies on the spring pressure and the mass of the armature to snap over to the magnet; It doesn't seem to do anything other than push; Not like a servo or something that has positional push/pull. So... it should be simple, maybe, to use the governor... Apply 24 VDC to the actuator to open it all the way, drop it down to 0 for shut off. If it doesn't shut off, do it again; The magnet didn't catch!

We'll see what happens!

Edit: It seems the speed controller, which would also be deleted, just gets power from the fault lockout relay (K12). When K12 opens, the governor actuator gets 0 volts as well. So... connecting governor actuator A6 in place of speed controller A5's terminals F and G should do the trick. That would be easy too, since you could splice them together in the same spot. Elegant, yes!

Edit: Pondering overload protection and the best way to handle that, since the inverter previously handled that and then opened the circuit interrupter as needed. I know most generators don't handle these conditions... but it would be nice to retain all the functions on the malfunction indicator panel. In an overload condition, wouldn't a voltage sag occur? Might be easiest to just monitor voltage and control the circuit interrupter based on that.

Edit: Why so many edits? So that I don't keep bumping the thread to talk to myself.

Edit: Remote start... Apply 2 seconds of "Start" switch position. Release. How can we know that engine is running... Do we have to touch AC power for that? Confirm engine running. Let run for 3-5 minutes (check startup plaquard). Close AC circuit interrupter.

 

[OverkillTASF]

Currently looking at the readily available generator heads in my area... Trying to figure out (without pulling the PWM off of one of my MEPs) whether these engines are compatible with J609A or J609B mounting specs. Will probably buy a 3 or 4 KW head once that's figured out, which will require (on the ones I'm looking at) rewiring the center-tapped windings so I can get the full 3 or 4 KW in 120 VAC out. Woo.

 

[OverkillTASF]

Actually started fiddling with the Arduino today... scavenged a MOSFET of maybe-not-ideal specs from an old PC power supply. Fiddled with the actuator and tried to find some info on it... if it's anything like the other Governor's America Corp. devices I should be able to control it in a fairly straight forward fashion. Made sure the MOSFET worked for as a no-load switch. I stopped just short of hooking things up to the generator to test it out, maybe see if the 1ms pulse widths I can get out of the Arduino are low enough resolution to do what needs doing. Might blow up an Arduino. Whatever. Maybe I'll use an Uno instead just in case... Of my two remaining generators, I'm testing this in the one that has a "bad" Inverter... since I suspect that's the one I'll keep and the other that seems to be all working is probably a better option to sell.

Still don't quite understand, when everything is working properly, how the governor/actuator play together to shut it off. Since the actuator only pushes the throttle open, I don't see how it ever overcomes what seems like the "idle" position of the mechanical governor (while powered off... it's been a while since I fiddled with one of these running since they're in my basement) in order to capture the magnet. In any case, DieselAddict's idea of setting the mechanical governor in the "shut off" position would take mechanical governance out of the equation and make the default position "off". That would work, but then RPMs, even at idle, would be controlled entirely by the actuator. Not sure it's designed for that or works that fast. Guess I'm gonna have to run one of the buggers to refresh my memory on how the engine gets shut off when you hit the "off" switch... Or find someone who has decoded the signals sent from the speed controller, and unlock the secret "bounce back to the magnet!" command.

 

[DieselAddict]

The secret is to flip the power polarity to the actuator.

 

[DieselAddict]

The actuator doesn't have a lot of push or pull. Not enough to overcome the full tension of the governor to kill the motor if you set it up to run manual full rpm.

 

[m715]

I don't have a 831 but I do have a 6kw generator with the yanmar diesel as sold by central Maine diesel. That gen set has a shutdown that uses a fuel pump with a solenoid that shuts the fuel off.

 

[Dwnorton1]

The secret is to flip the power polarity to the actuator.

I agree that this is what has to be happening, because I have been chasing a intermittent fault in my k12 fault relay circuit. The gen would start but would not get the reference to ground that the the fault module provides through the k12 relay. The engine will start but I have to pop the hatch and force the governor actuator over to the magnet to kill the unit. That indicates to me that the actuator has to be helping out or the linkage arm as it would never be grabbed by just strength of magnet alone. When unit is functioning properly it grabs target and shuts off unit. I hope you get this figured out as I have a burnt up inverter on one as well.

 

[OverkillTASF]

The secret is to flip the power polarity to the actuator.

I actually did that... it went in the "open" direction at both polarities, and my notes from testing a running one were not that it ran negative voltage... does yours behave differently?

I don't have a 831 but I do have a 6kw generator with the yanmar diesel as sold by central Maine diesel. That gen set has a shutdown that uses a fuel pump with a solenoid that shuts the fuel off.

All of mine will happily run even with the fuel pump shut off, albeit poorly. Wish it were that simple! Kill all the electrics on one of these and it will go until it breaks or runs out of fuel!

 

[Dwnorton1]

All of mine will happily run even with the fuel pump shut off, albeit poorly. Wish it were that simple! Kill all the electrics on one of these and it will go until it breaks or runs out of fuel!

That is exactly what mine does when I start it and the issue with k12 gets me. It will run at really low rpm with switch off and without magnet capturing target. I suspect that the reverse polarity may just be a pulse to get target heading toward magnet where it can make its capture but not positive on exactly how shutdown works though.

 

[OverkillTASF]

If anyone could check the following with a basic multimeter at the speed controller it would be helpful... I can't run mine until I clean up the basement enough to skid the bastards out...

1. Voltage between J / H
   1. On switch "On", prior to starting.
   2. On switch "On", prior to starting, battle short enabled.
   3. If possible, on switch to "running", while trying to start. Should probably disconnect the starter for this...
   4. Engine running, idle, no load.
   5. Engine running, no load, governor actuator locked "up" in start position. This should be 3600 RPM.
   6. Engine running, no load, pressing down on governor lever, slowing motor slightly below idle. The actuator should start lifting.
   7. Engine running, with load.
   8. Engine running, with load, pressing down on governor level, slowing motor slightly below where it wants to be. Actuator should start lifting.
2. Bonus points for: Current over J or H during the above scenarios.
   1. (Answered one myself) 24 volts direct from the battery results in 2.2 amps current to the governor.
3. Extra bonus points for: Duty cycle over J/H during the above scenarios? Wish I still had my smelloscope...

Messing with it tonight... I'm still kind of struggling with what is actually happening during shut down, and if the actuator can ONLY push out away from the magnet and is dependent entirely on its built in spring tension to return to equilibrium. Also see some odd behavior moving to "On" with and without battle short enabled, hmm...

My musings on the mechanical governor... wants, badly, to run the engine at 3600 RPM. Ostensibly with everything shut off the mechanical governor is trying to reach its full throttle position. The actuator lever's bit of spring tension keeps it from getting there. Even when starting? On the one I can get to with the bad inverter, the actuator doesn't lift all the way in "On" unless I turn on Battle Short". I would expect when starting it would demand full throttle though, and lift all the way. In any case... once you start running, upwards pressure on the governor decreases as the governor is a little closer to the 3600 RPM that it wants... so the actuator is lowered a bit, and the magnet gets closer to the target... The governor still wants 3600 RPM though. When the speed controller sees load on the generator, it applies some upward force to the actuator, allowing the governor to push up a bit more, thereby raising RPM. But what happens when you hit the kill switch... Doesn't the mechanical governor still want to get to 3600 RPM, so its pressure would be upwards on the actuator, pushing it further from the magnet? Doesn't the sudden loss of actuator power (assuming actuator can only be moved one direction) mean that the only motive force towards "Shut off" is the spring in the actuator, so it's really just dependent on that spring providing enough of a sudden "snap" backward to capture the actuator and hold the governor lever in the "off" position long enough to shut down? Applying 24 volts to the actuator and then letting off it (when everything else is off) doesn't have enough oomph to snap back and hit the magnet... but that might be because the engine isn't running and so maximum force is being applied on the governor lever...

I could certainly rely on the mechanical governor to keep the engine at the 3600 RPM necessary for 60 Hz power... But it doesn't seem like, if I did that, the actuator has any way of actually shutting things off. And leaving the "speed selection knob" set to the stop position doesn't alllow the actuator to do anything, since it seems like the actuator can't demand RPM, but simply caps the max. If I can't shut it off electronically, that makes the project simpler but more boring and with fewer bells and whistles.

What a rant.

I wish I knew how other implementations of the L70 handled remote start and shutdown. This arrangement with the actuator seems pretty silly. Though I REALLY want it to work... because I can think of cool/fun things to do with it if I can control the RPMs between idle and full speed, even with a standard generator head.

 

[DieselAddict]

Lets start with some basic information.

Pull start the engine without turning on the main switch. Make sure the electronic governor arm is free to move (not on the magnet and not "pinned" fully to the right). What RPM does the engine run? It should be a low idle. If it idles at 3600 rpm you have a mechanical governor adjustment issue.

The actuator does not have enough torque to overcome a misadjusted mechanical governor. If the mechanical governor is not adjusted correctly the electronic governor has zero hope of working correctly.

 

[OverkillTASF]

Thanks DA. I'm sure I'd be able to answer some of my own questions through experimentation, but I can't run the genny where it is right now, and can't get it moved out of the basement with the mess I've created. Still got one generator to sell and can't even get to it yet. Heh.

Sounds like what you've stated is about what I would expect from the system. If the actuator IS pinned, you would expect the mechanical governor to run the engine at 3600 RPM, right?

 

[DieselAddict]

Yea, in the pinned position it should be in that range.

I want to say (from memory) that the mechanical governor needs to have its absolute top speed set for about 3750-3800. With that the electronic governor will be able to push it up to about 3600 without needing too much torque from the actuator.

I'll double check when I'm home at the end of the week.

 

[Dwnorton1]

The governor control puts out 5.3 vdc at no load and 11vdc at full load according to TM. Mine is actually like 5.9 and 11.4 or somewhere there abouts if memory serves. Had one with failed governor (24VDC+ output from J&H) and it would definetly wrap the engine up to high rpm. I think most of us that are vocal about the 831's must travel for a living. I can check mine this weekend as well. craigc would be another to check with but alas he very well could be on road as well.

 

[OverkillTASF]

The governor control puts out 5.3 vdc at no load and 11vdc at full load according to TM. Mine is actually like 5.9 and 11.4 or somewhere there abouts if memory serves.

Ah-ha! DWNorton, thanks man.



I had been through the TM several times but somehow missed this. Hope I didn't fry my actuator hitting it with 24 volts for brief moments. If the positioning of the actuator is just straight up voltage, that should be a piece of cake. Strange that, even though this actuator is gov specific, it's 12 volts on the high end.

 

[DieselAddict]

Just as a general FYI, I'm using a DC Brushed motor driver to drive the actuator for the custom speed controller I've been working on.

https://www.pololu.com/product/2992

 

[OverkillTASF]

Just as a general FYI, I'm using a DC Brushed motor driver to drive the actuator for the custom speed controller I've been working on.

https://www.pololu.com/product/2992

Interesting... if the voltage range of the actuator in operations is 5.3-11VDC, are you having any issues with that driver? Or does some other factor of your design negate that?

Edit: Also, is what you're working on top secret or what? Just a private project or... Be a shame for me to reinvent wheels you're already working on!

 

[DieselAddict]

The actuator if I remember correctly can withstand a good bit more than than 11v. I'm using B+ to drive it (~27v). Since its a PWM output I just clamp the duty cycle on the PWM to limit the current. Ultimately when I'm done I may choose a less expensive driver and put in one an additional power supplies. I'm already using one for the 5v stuff on the Arduino.