Cummins-Onan MDKBM diesel generator won't start

[Zanshin]

I've got a generator that doesn't want to run.

It will almost fire up while I hit the "start" button.

When I check the fuel priming/pump it gives good volume right before injection pump and also at the return line at the fuel tank.

I've just removed the cut-off solenoid (it actually doubles as a regulator with a square-wave signal) and checked it. I also applied a direct 12V while trying to start, and saw that it does get signals. I can't check the square wave without an oscilloscope on board.

What is odd, is that I've cracked the lines at the injectors and nothing comes out. I cracked the line at the pump exit, and also nothing came out when priming or even when starting.

I'm not sure what is going on, particularly as sometimes the engine sounds like it is running (roughly) while starting but stops immediately when I release the "start" button. So some fuel must be getting to the cylinders.

I'm at a loss as to where to look next. Any recommendations?

[skipperpete]

Hi Zanshin, I think that what you called the “shutoff solenoid” is actually the electronic governor. I did a troubleshooting job on this model Onan/Cummins on an almost new boat and the manufacturer sent the owner all the senders and the electronic governor prior to my arrival. On this boat, the generator would cut out under sail.... but only under sail and the brief was to simply replace all the senders and the electronic governor and start the diagnosis after that was done. I did all the routine engine checks, belts, coolant, wiring harness sockets and connections, exhaust system and finally before startup... lube oil level...... only a tiny glistening jewel of engine oil on the end of the dipstick..... the crew had done the first 100hr service and never properly dipped the oil level.... problem solved, it was the low lube oil pressure shutdown when the yacht was heeled over.
My suggestion is to check all the vitals and then all the senders. I only have a casual relationship with the Onan Electronic governor and I don’t know where it gets its sense from ( hitting it with 12 volts might have been a bad idea) but it’s definitely worth investigating. Of course you will need to thoroughly check the fuel supply system from tank to injector pump to completely eliminate the primary fuel circuit as the culprit.
Finally, close the raw water intake to avoid filling the exhaust while dead cranking.
Watching with interest,
Pete

[Tin Tin]

Are you in a cold climate, maybe glow plugs circuit not working.

[cr180]

Pretty sure fuel related. Think you think that to. Thought about the electric priming pump, but then reread post, and that seems to be fine. Maybe air in fuel injection pump, but I think you are doing the right things to vent. Two thoughts. 1. May need to let the engine try to start for a longer period of time with injector line cracked. But as others said, be careful, your muffler will fill up with water and backup into engine (muffler should have a drain plug) or you’ll destroy your impeller if try too long if seacock closed. 2. Could it be a bad fuel injection pump?

[cr180]

Still think fuel related. But another thought. I have a Phasor generator with a Kubota Diesel engine. When starting, I must hold the start button long enough for the oil pressure to build enough to open the low oil pressure switch otherwise will not continue to run. Low oil pressure switch will trip stop solenoid.

[Ericson38]

Originally Posted by Zanshin

I've got a generator that doesn't want to run.

It will almost fire up while I hit the "start" button.

When I check the fuel priming/pump it gives good volume right before injection pump and also at the return line at the fuel tank.

I've just removed the cut-off solenoid (it actually doubles as a regulator with a square-wave signal) and checked it. I also applied a direct 12V while trying to start, and saw that it does get signals. I can't check the square wave without an oscilloscope on board.

What is odd, is that I've cracked the lines at the injectors and nothing comes out. I cracked the line at the pump exit, and also nothing came out when priming or even when starting.

I'm not sure what is going on, particularly as sometimes the engine sounds like it is running (roughly) while starting but stops immediately when I release the "start" button. So some fuel must be getting to the cylinders.

I'm at a loss as to where to look next. Any recommendations?

I posted this in the wrong place, was trying to assist you-

https://www.cruisersforum.com/forums...wn-268218.html

[Zanshin]

@skipperpete - It is the governor, which also functions as a cut-off solenoid, that I was talking about. Hitting it 12V to test full movement wasn't a mistake, that came straight from the test page on the service manual. I've primed the system for a very long time and have already turned off the water cooling to avoid flooding the engine.
@Ericson38 - I'm hoping it isn't the fuel injector pump assembly, that is the single most expensive piece of hardware on the engine after the motor block!
@cr180 - the manual says that low oil pressure sensing is turned off during the starter process, and I've checked the oil level and it is in range. So if the sensor is broken that wouldn't affect this problem. And I've removed the sensor and measured the resistance specified in the manual and it is in range as well.
I should add that this generator has an extensive list of error codes (indicated with flashing lights) and none of those have gone off; except the "excessive cranking", which I know about.
I've checked the air filter and it was, as expected, pristine - it only gets air from the generator compartment and that gets air from the cabin and engine compartment; both of which are quite clean compared to a normal car air intake.
I'll make it to St. Martin next week, and there is an authorized Cummins-Onan dealership/shop there so I might need to take this to a professional. There's a chance that the controller board is dead; which I cannot debug on my own.

[skipperpete]

Thanks for the update Zanshin, it seems like you’re well ahead of the basic troubleshooting routine. Do you know where the governor actually does get its sense from, I’m interested for future Onan callouts. Initially I typed that if I was there I would consider running the engine with the governor removed and the manual mechanical stop function verified but I considered it too risky for the average cruiser so I deleted it. From your response it seems like you’re familiar enough with the genset to try this. The only electrical or electronic fault that would stop the engine must be manifested at that governor unit. Did the plunger retract or extend with the 12volt DC across it?

[Zanshin]

@skipperpete - the manual says that the PWM signal comes from the quadrature winding output. I have an Arduino on board and could program that to display the PWM signal since I have no oscilloscope, but the multimeter showed varying values that indicated a signal was being made.

The manual states that when not energized, the spring-loaded plunger is in the "no fuel position". So when I applied a pure 12VDC to it, it should have been fully open but even then the engine sounded no different and wouldn't start.

What I haven't checked are the glow plug and starter relays - but I'm in the tropics so the glow plugs shouldn't be important and I would think that if the starter relay were broken the starter wouldn't engage.

Oh, and I've bridged the starter battery to my engine starter battery to make sure that there's enough power; that, too made no difference.

I might remove the governor again and see what happens when I try to start the engine without it in place. That was a tough piece to get in an out despite only having 2 bolts, but one is in the rear and almost inaccessible. It took me close to an hour to remove it earlier today. It requires an Allen key that needs to be inserted by feel and gets less than 1/8 of a rotation per attempt.

[Ericson38]

Originally Posted by skipperpete

Thanks for the update Zanshin, it seems like you’re well ahead of the basic troubleshooting routine. Do you know where the governor actually does get its sense from, I’m interested for future Onan callouts. Initially I typed that if I was there I would consider running the engine with the governor removed and the manual mechanical stop function verified but I considered it too risky for the average cruiser so I deleted it. From your response it seems like you’re familiar enough with the genset to try this. The only electrical or electronic fault that would stop the engine must be manifested at that governor unit. Did the plunger retract or extend with the 12volt DC across it?

Pete,

This K1 relay is not a governor, but a relay (fuel cutoff). Also known as the shut down soleniod-

https://www.smokstak.com/forum/threa...-others.89953/

[Ericson38]

Originally Posted by Zanshin

@skipperpete - the manual says that the PWM signal comes from the quadrature winding output. I have an Arduino on board and could program that to display the PWM signal since I have no oscilloscope, but the multimeter showed varying values that indicated a signal was being made.

The manual states that when not energized, the spring-loaded plunger is in the "no fuel position". So when I applied a pure 12VDC to it, it should have been fully open but even then the engine sounded no different and wouldn't start.

What I haven't checked are the glow plug and starter relays - but I'm in the tropics so the glow plugs shouldn't be important and I would think that if the starter relay were broken the starter wouldn't engage.

Oh, and I've bridged the starter battery to my engine starter battery to make sure that there's enough power; that, too made no difference.

I might remove the governor again and see what happens when I try to start the engine without it in place. That was a tough piece to get in an out despite only having 2 bolts, but one is in the rear and almost inaccessible. It took me close to an hour to remove it earlier today. It requires an Allen key that needs to be inserted by feel and gets less than 1/8 of a rotation per attempt.

So if you manually pull the plunger back connected to the threaded end of the solenoid plunger (with is connected to the rack through a ball and linkage), it should start. In your case, if you let go of it, it should die, since the keep alive windings are most likely open circuited. That you can test with the procedure from Smokestack link.

Onan stop solenoid 307-2546-00 used on DKD, MDKD and others

https://pjpowerinc.com/p/307-2546-on...solenoid-valve

[skipperpete]

Originally Posted by Zanshin

@skipperpete - the manual says that the PWM signal comes from the quadrature winding output. I have an Arduino on board and could program that to display the PWM signal since I have no oscilloscope, but the multimeter showed varying values that indicated a signal was being made.

The manual states that when not energized, the spring-loaded plunger is in the "no fuel position". So when I applied a pure 12VDC to it, it should have been fully open but even then the engine sounded no different and wouldn't start.

What I haven't checked are the glow plug and starter relays - but I'm in the tropics so the glow plugs shouldn't be important and I would think that if the starter relay were broken the starter wouldn't engage.

Oh, and I've bridged the starter battery to my engine starter battery to make sure that there's enough power; that, too made no difference.

I might remove the governor again and see what happens when I try to start the engine without it in place. That was a tough piece to get in an out despite only having 2 bolts, but one is in the rear and almost inaccessible. It took me close to an hour to remove it earlier today. It requires an Allen key that needs to be inserted by feel and gets less than 1/8 of a rotation per attempt.

Again, thanks for the clarification, disregard my suggestion re the manual stop, it was originally where the square plate is directly behind the governor but it’s removed for the electronic governor.
If the fault is not related to the electronic and electrical system it must be a way less annoying mechanical or fuel problem. Does your generator have the little “bleed to tank valve” on the injector pump inlet? is it closed? Maybe try a preheat! and based on a recent experience with an engine that defied all attempts to start it, can you rotate it manually , I know that’s not easy with the Kubota shaft extension but might be do able with the alternator pulley.

[Zanshin]

Originally Posted by Ericson38

So if you manually pull the plunger back connected to the threaded end of the solenoid plunger (with is connected to the rack through a ball and linkage), it should start. In your case, if you let go of it, it should die, since the keep alive windings are most likely open circuited. That you can test with the procedure from Smokestack link.

Onan stop solenoid 307-2546-00 used on DKD, MDKD and others

https://pjpowerinc.com/p/307-2546-on...solenoid-valve

The link is for a different generator. This one, the MDKBM, uses the solenoid not as a shut-off, but as a speed modulator. Once installed, it cannot be manually moved. There is no mechanical linkage for speed control.
I am going to remove the modulator, which should "open" the injector pump valve, and then see if I get fuel flow to the injectors. If that doesn't work, then I'll give up and pack things together so that I can have it checked in St. Martin.

[skipperpete]

Zanshin that’s an interesting situation with the shutdown position of the electronic governor and it makes perfect sense if viewed in the context of a loss of power/signal in the speed control circuit, without this feature there would be an overspeed in the absence of a mechanical governor to dynamically control the rpm and frequency . Most mechanically governed engines ( but not all of course) stop in the maximum fuel position and some even have a button to further release the rack to “overfuel “ position for cold starting so if your conventional cassette injector pump gets a stuck plunger in the interval between shutdown and next start the result would be no fuel rather than the more frequent and destructive uncontrolled overspeed.... this would apply even if your engine uses the combination of the original mechanical governor controlled by the electronic servo.
And the result of removing the servo won’t change the situation if the rack is still stuck in the no fuel position..... still no fuel from HP pipes.
Its just a theory but possibly worth pursuing and it could be dismissed by removing the blanking plate and viewing the position and movement of the rack linkage rather than removing the injector pump. If you do venture into the “stuck rack” territory, play it safe and do all of your testing with the injector lines open.
Pete.

[Zanshin]

Update:

I removed the governor, tried to start while manually moving the throttle lever with no luck.
I then removed the fuel lines between the injection pump and the injectors. Cleaned them and checked for blockage. None found.

A "cracked " injector shows very little (i.e. 1-2 drops) of fuel when priming or cranking, so I'm assuming

1. Fuel pump. Looks good but perhaps not enough pressure.
2. Governor. Perhaps some strange configuration problem.
3. Injection Pump. Unfortunately, this is the most expensive and most likely cause.

I've put the engine back together in preparation for my passage to St. Martin. There is a dealer/shop there, Electec, who I am going to hire to do the rest. I've reached my level of incompetence (The Peter Principle).