• Thank you for visiting HeavyEquipmentForums.com! Our objective is to provide industry professionals a place to gather to exchange questions, answers and ideas. We welcome you to register using the "Register" icon at the top of the page. We'd appreciate any help you can offer in spreading the word of our new site. The more members that join, the bigger resource for all to enjoy. Thank you!

Mitsubishi 4D32 diesel power unit in CAT 307SSR

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
This isn't an excavator-specific set of questions, but really more specific to the Mitsubishi engine - and this is a gray-market machine for which I've got a parts book (in Japanese) and no other documentation. The 4D32 seems to be a relatively common engine in Asia, but not stateside.

I've got some kind of fuel-circuit problem that's resulted in poor power and missing. I had the tank out due to a pinhole, & this was a long-duration repair. In the meantime, I ran it occasionally to reposition & do some light work, and during that time, fueled from a simple jury-rigged 5-gal bucket setup. Far as I can recall, it ran OK prior to the leak, though I wasn't using it much. I'm certain it took up some air during tank-swapping, but it never really seems to have bled.

Now the real tank is back in, I've bled it to death, and it's still missing like crazy and stumbles/dies very easily. I don't think it's a simple filter-clog, since fuel flows out the top of the filter-housing if I pull the 1/8" pipe plug, and all the injectors spit at me while I crack their nuts open at idle - though I should mention the impact on the idle of cracking each of them is far from consistent or uniform across the four.

Here's one of my big unknowns about the fuel circuit: there are three tank hoses. One's clearly the return. Another is obviously the feed to the normal-service mechanical fuel pump, which has a typical hand-primer pump built in. But the third hose...a short length off the bottom of the tank feeds an electric clicker pump, also mounted on the bottom of the tank, and that pump pushes fuel directly to a port that's on the 'front' (small end) of the IP. That IP port is physically quite near to the return port, but just 'around the corner,' so to speak.

There was no filter in this line, which bothers me a lot, since it's definitely set up to push fuel /towards/ the IP. The electric pump is, I think, only intended for some kind of bleeding assistance when it's run dry - I say this because there's a spring-return, momentary rocker switch that powers it from the cab, so it's not intended to be left on. (Who'd want to pump gritty, sludgy, near-bottom-of-tank fuel to the IP after running dry?)

Or did I get that wrong?

Anyway, I'd never had much luck with the electric pump helping or hurting, but it and the IP hose connections are concerning me as unknowns, now that I've got this lumpy, irregular running.

I'm thinking I might order & replace injectors and see what that does for me - but is there anything else to check first?

Thanks - Dave
 

Mobiltech

Senior Member
Joined
Jan 14, 2014
Messages
1,680
Location
Sask.
Occupation
Self employed Heavy duty mechanic
Do you know what the serial number of the machine is?
 

Delmer

Senior Member
Joined
Jan 3, 2013
Messages
8,864
Location
WI
That's probably a Zexel pump. If it's an inline pump then apparently there's a screen in the banjo bolt at the inlet fitting on the bottom of the pump. That could cause some of your problems. Also the likely water contamination that led to the pin hole in the tank could have caused problems anywhere?

Do you have any pics of where the 3rd line comes into the IP, maybe the rocker switch that controls that 3rd line pump? Never heard of anything like that, sorry.

Also, you could get the injectors tested before deciding to replace them, or remove and observe the pattern on the engine but out of the cylinder. And do a compression test at the same time.
 

lantraxco

Senior Member
Joined
Jan 1, 2009
Messages
7,704
Location
Elsewhen
Emphasis on that screen in the center of the banjo bolt where the feed line comes from the tank, it's a Mitsubishi common issue.
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
Delmer,

Thanks, good stuff. Those are the kind of suggestions I was hoping for.

I'll see about getting the screen out of the inlet fitting, though it's in one of the worst places to try to reach (without removing the cab). I have memories of pulling the pump out once after running too low, and finding enough gritty stuff within the check-valves to have made the mech pump stop working. I'd become numb from cussing by the time I was done, and swore I'd replace it with an electric pump if I ever had to do anything more in that zone. And thinking about it again, it could be the same thing has happened, but hey, wait, I have fuel coming out the bleeder-screw when I crack it (top of filter housing), so wouldn't this be a painful and pointless task? (Not that I don't engage in those regularly.)

I think not a water-in-fuel issue (hopefully) because the way I noticed the leak was a growing stain in the soil under the excavator, eg fuel leaking out, not so much water leaking in, or so I hope. On draining the tank, I definitely got some rusty crap, but not more than I think is typical from condensation-accumulation. The rust-hole came from the outside, under a hidden buildup of leafy/soily debris that was against an inner face of the tank, piled up against the bottom weld for the last X years...

Here's a scan of the parts-book page showing the IP - maybe you can determine if it looks like the Zexel you mentioned. I just sent that to a Japanese-native friend to see if he can translate the descriptions for the three fuel connections. injection pump components_1.jpg


Here's a photo of the 'front' of the pump where two banjos live:

IP front view.jpg

it's a very constricted space and none too clean, but maybe you will be able to see what's going on .

Rocker-switch, too, shown below...funny, the 'universal' symbol on the switch is an icon of a typical gas-station pump.

cab switch panel.jpg

Prior to pulling out the tank, I'd assumed that the electric pump either just bypassed the mechanical, or drew FROM the IP, so as to pull air out of the rest of the system, but there's no way it could have been installed thusly.

I should own a diesel compression-tester, but don't yet. I guess I have to hit ebay for that.

Question: to aid in diagnosis and bleeding, I keep wanting to install - at the bleed-screw on top of the filter-housing - a tee leading to a 1/8" valve and a snubber/pressure-gauge. Wouldn't that be immensely helpful in checking fuel filter isn't clogged and that reasonable pressure is reaching the IP?
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
Emphasis on that screen in the center of the banjo bolt where the feed line comes from the tank, it's a Mitsubishi common issue.

Thanks lantraxco...See my followup about the screen - if I'm getting fuel out the filter-top bleed-screw, doesn't this rule out a problem with that screen?

--Dave

(so now I have to google "Morooka," along with the rest of my Japanese manual...?)
 

Delmer

Senior Member
Joined
Jan 3, 2013
Messages
8,864
Location
WI
Thanks lantraxco...See my followup about the screen - if I'm getting fuel out the filter-top bleed-screw, doesn't this rule out a problem with that screen?

No, you're getting low power and missing. Assuming both of those are with the engine running... then you need to have flow and pressure, at the same time. A pressure gauge is an excellent way to diagnose fuel issues.

So the far right switch is the starter? Then is the "gas pump" an extra fuel start aid, hydraulic activated, the unfiltered fuel not mixing with the filtered? hard to believe they'd go to that much trouble but what do I know?

You can do a rough "compression" test by pressurizing each cylinder with a rubber nozzle air gun and seeing if you can detect air leaking out the intake, exhaust, or crankcase, the places where the compression will eventually leak.
 
Last edited:

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
No, you're getting low power and missing. Assuming both of those are with the engine running... then you need to have flow and pressure, at the same time. A pressure gauge is an excellent way to diagnose fuel issues.

So the far right switch is the starter?
No, I *think* that's glow-plug switch...I began using this machine before I even knew it had glow-plugs, but never used them even after I checked it out a little more - simply since I never had a problem starting it (still don't)...

Then is the "gas pump" an extra fuel start aid, hydraulic activated, the unfiltered fuel not mixing with the filtered? hard to believe they'd go to that much trouble but what do I know?
From all wiring-checks, disassembly and device-noticing duringmy pokings and proddings over the years, there's no associated hydraulics - it just switches on the electric fuel pump for as long as you can stand to hold the switch down.

You can do a rough "compression" test by pressurizing each cylinder with a rubber nozzle air gun and seeing if you can detect air leaking out the intake, exhaust, or crankcase, the places where the compression will eventually leak.
Ah, so the point wasn't to quantify, but to look for leaks, then?

What about the IP - is it familiar?

I never asked about the line tying the filter outlet to the return-line banjo...presumably there's some kind of spring-check valve in that line at some point, maybe a flow-control valve - something, anyway - that prevents it from just short-circuiting fuel back to the tank, dumping all the pressure there?

--Dave
 

Delmer

Senior Member
Joined
Jan 3, 2013
Messages
8,864
Location
WI
Glow plugs makes even more sense. I have NO idea about the 3rd line, was just thinking that it could possibly activate something hydraulically inside of the injection pump. Makes no sense, but neither does priming with unfiltered fuel when there's already a priming pump there.

I'm no pump expert, that one doesn't look like the Zexel on a 4D34 that I'm trying to figure out, but it is a mirror image to a Kiki that I took a look at. Mine has that fitting going to a third port on the fuel filter base. All of the "extra controls" on mine are mounted on the back of the pump, so that would seem to rule out my "hydraulic fuel enrichment" notion. Then again, my engine has a throttle plate and NO cables going to the IP, so what help is that.

For all I know of the workings of these pumps the ports that are connected could be the "waste" side, and it wouldn't matter if the fuel is filtered, it's going through the filter before it goes into the injectors anyway? I wouldn't worry about the original engineering, if it doesn't look Jethro'd in there then it's probably fine.

Lantraxco seconded the banjo screen so do that first. Then maybe install a fuel pressure gauge. Then I'd take a look at the injectors and compression.
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
I'm still stuck in my tracks, so to speak, but I've got some more (confusing) data:
1) I didn't pull out the mech pump or check the screen that's said to be somewhere on the inlet to it (it's really hard to pull due to location, and I think nothing is clogged)... but I did test it by substitution - I bypassed it with a lay-around electric pump. With the electric pump running the fuel flows without a problem through the circuit, and it can be seen/heard splashing back into the tank, so I take it this is as good of an indication as one could hope for that there's no screen or filter blockage.
2) As if that were not enough, I put a gauge at the IP fuel inlet, and it jumps past its 10psi scale as soon as the pump is on, so I have no question the IP is being fed with fuel.
3) Running conditions are, abruptly, WORSE. It went from crappy, stumbly running, to not starting at all after I'd posted a few times about this problem.
4) I was wrong in my supposition about that far-right switch being for glow-plugs. Testing revealed that they get energized by a relay that the start-switch engages; I don't know what switches them off. (I seem to recall now that the far-right switch is an unlikely power-limiter device...)
5) I've bled everywhere repeatedly, right up to the injector tube nuts, and though I get fuel out, I now get no ignition at all.

I'm hanging all my hopes on bad fuel right now, and that's not entirely wishful thinking. I've run this machine on biodiesel for the most part for some years of intermittent use, but the last batch I bought recently smelled like rotten meat. The producer isn't a small-time operation, and says it should be fine since they run their vehicles on it, though. And that may NOT be the problem, since I refilled another machine with 10 gals of the same stuff and have been using it intermittently for about a month now. Still, my machine went from stumbling and running poorly on the previous fuel, to not starting at ALL on the newer stuff (I didn't mix them when I put the rebuilt tank back in). With nothing else that's easy to test, I figure I will buy 5 gal of petrodiesel and see what happens when I replumb things to run that fuel instead. Not a lot to lose there.

Failing that - compression test, I guess?

Delmer, you mentioned some means (maybe theoretical) of pulling the injectors and reconnecting them to their fuel lines. I don't know of a simple way to do that. If there is, please let me know, since that would be far easier than running around town trying to find a willing diesel shop to test them - though I'd do that right off if I had some indication that it was worth doing.

Thanks - Dave
 

Delmer

Senior Member
Joined
Jan 3, 2013
Messages
8,864
Location
WI
An injection shop will test the opening pressure, spray pattern, and no dripping. You can tell if the spray pattern is way off by hooking up the injectors outside of the head. Either with an extra injector line, or one of the originals might be able to turn enough. Usually the toughest part is getting the injectors out (in general, never done a Misubishi). At that point, might as well get them tested and do a compression test... Short of water getting into the cylinder, the compression shouldn't drop from sitting...

For now, either you're getting fuel into the cylinders or not. Is there any mist/fog/smell coming out of the exhaust when you're cranking the engine?

You have overwhelmingly suspect fuel, but you don't think the screen is clogged? I still do. Trying another pump that doesn't get the engine to start is not a conclusive test in my opinion. The 10PSI reading is something, I can't argue with that. Did you bleed the injector tube nuts AT THE INJECTORS? with no air bubbles? Then hook them back up and see if you get mist out the exhaust. Sorry, I'm not more help, my experience is limited to outside the pump.

You do know that age and humidity turn GOOD biodiesel into nasty slime, right?
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
Thanks, Delmer...

There is a light-gray mist coming out of the exhaust as I crank, but not even a hint of ignition. I finally thought to check my stop-solenoid, thinking it had failed in the stop position, but it looks like it's operating.

Screen clog: I can't see how it could even be a factor with my bypass. It's said to be at the lift-pump inlet, right? I've got the lift pump completely out of the equation - pulled off and blocked the outlet hose, bypassed with an entirely separate electric in-tank pump and 10' of hose going directly to the filter-head inlet. Did you or I misunderstand where the screen is said to be? (Even if I did, there's a steady flow of fuel through the filter/return circuit, returning to the tank constantly, so I can't see how there could be any still-relevant problem with the lift-pump, screen, or existing hoses: all of them, other than the return-hose, are bypassed for testing.

Yes, been bleeding right at the injector-tops, and occasionally at the filter-top, though I'm convinced that's not necessary because that circuit has an active "flow" now from the electric pump - no air there at all.

The smelly biodiesel is only a month old. I've actually not had a problem with biodiesel going bad, even after years of sitting in a tank - though it does "polymerize," leaving chicken-skin-like deposits on items like filter-screens, and a brownish gunky coating on metals - like the brown stains you get on the outsides of old frying-pans, also from oxidizing. But the liquid portion still burns fine. Inside of the tank on my loader probably looks like Steven King's worst nightmare, but it fires up every time.

I came home this evening with 5 gal of petrodiesel, and tomorrow I will swap over to that with a new filter and see if there's any difference. Seems to me the deterioration is indeed too fast to be anything "wearing out" or the injectors getting gunky, but this is new territory for me.

Thanks again - Dave
 

Delmer

Senior Member
Joined
Jan 3, 2013
Messages
8,864
Location
WI
The smelly biodiesel is only a month old. I've actually not had a problem with biodiesel going bad, even after years of sitting in a tank - though it does "polymerize," leaving chicken-skin-like deposits on items like filter-screens, and a brownish gunky coating on metals - like the brown stains you get on the outsides of old frying-pans, also from oxidizing. But the liquid portion still burns fine. Inside of the tank on my loader probably looks like Steven King's worst nightmare, but it fires up every time.

Fires up every time, until now that is. I was wondering why Pumpguy hasn't set us straight yet. Now I know, he saw this coming before I did and is down at the bar crying in his beer over another poor abused biodieseled pump.

With the mist, you're getting fuel so everything should be working up to that point. Have you checked the glowplugs? disconnect the electrical connector from each one and take an ohm reading in place, it should be obvious that they're all in the same range or open. And are they getting power when you crank? letting them warm up 10-20 seconds before cranking? One wouldn't think you'd need glowplugs, but if they had been coming on everytime it started and now the injectors are messed up maybe they're needed now???

I agree compression shouldn't go bad overnight, but worn gas engines will sometimes "wash down" the cylinders with gas, causing aggravating low compression/no start when working on other problems. I've never seen it, but it sure could happen to a diesel, might involve some sticky rings as well. With fuel that "green" gunky injectors could happen at any time, it's not like this was running fine one minute and then wouldn't start the next.
 

Mobiltech

Senior Member
Joined
Jan 14, 2014
Messages
1,680
Location
Sask.
Occupation
Self employed Heavy duty mechanic
Does it have a diesel burning glow plug in the intake manifold? Some 307s did and it can cause air in the fuel if the valve in the glow plug is not seating.
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
Does it have a diesel burning glow plug in the intake manifold? Some 307s did and it can cause air in the fuel if the valve in the glow plug is not seating.
No, no 'devices' in the intake manifold. There are four glow-plugs in the standard top-of-cylinder locations, though. From testing yesterday, they're getting power, but only when keying over to the 'crank' position on the keyswitch. I do need to get out there and check their resistance, though, as Delmer mentioned.

On that subject - another of the gray-market mysteries of this machine is a device that used to be installed amongst other controls, next to the starter-switch. It's got to have something to do with the glow-plug circuit (I think so because it heats up under some circumstances, eg trying to start), but it's only connected by 18-ga wire. Photos show the thing. It's as if it was some indicator that glow-plugs were energized, blended with a bizarro cigarette-lighter...it doesn't get hot enough to melt a plastic-bag, though - I'd had it wrapped up to keep it from getting too dirty or wet in the bottom of the console when I hadn't realized it even got hot, and the bag's still OK.
dash device c.jpg
dash device b.jpg
dash device a.jpg
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
More data, as I dive into the rabbit-hole:

1) did the fuel-swap (positioned a temp tank with new petro-diesel & fully flushed rest of system) but this produced no new results - still no hint at ignition
2) glow-plugs are indeed getting power at key-over
3) glow-plug resistance ranges from 3.4-4.5 ohms, so I think not a problem (I could pull them and verify that they 'glow,' but it seems like a remote possibility that all plugs suddenly failed, right?)

Since I see nothing else to test at this point, I dived into pulling out stuff. I'm doing some googling to see how to get the injectors pulled; they're held 'down' by a bolted-on rocker assembly, and look as if once that's gone they should pull out manually, but if they do, it's sure not easy. I can turn the injectors all day with a little 10mm wrench, but they apparently aren't threaded, since they just keep spinnin'...more pulling force needed?

Yes, I will get a diesel compression-testing setup ASAP, but that's prolly going to be a mailorder proposition. Another option: if I've got the adapter fittings to make it work, I am wondering if a leakdown tester would have value here - it's intended for gasoline engines, but it seems like the same concept applies, where my compressed air would be injected into each cylinder in turn, to monitor how quickly & where each leaks. Anyone? I also have an assortment of accessory pressure-gauges I might be able to plumb in without waiting for a real kit to come. What rough pressure-range would be right for this?

Failing all of the above to reveal a problem - is there a way to test the IP while on the engine? What are typical test parameters?

--Dave
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
I was able to tease out one of the injectors - just a matter of rigging some pulling leverage, something I'd forgotten had to be done, even. It's a Denso pump, now that I can read the data-plate.

Local shop gets $57 to test four, and would charge $80-something to rebuild each if need be. That wouldn't be an issue if I knew it would fix the problem, but yeah, I don't...and I suspect I can get new ones for less.

I suppose compression-testing first would be the way to go - but since these are not threaded injectors, I guess I will have to reinstall them and the retainers, & compression-test via the glow-plug holes, yes?

Any way to test the IP...for pressure? Volume?

--Dave
 

repowerguy

Senior Member
Joined
Mar 18, 2015
Messages
810
Location
United States southern Ohio
Occupation
mixer truck mechanic
I had a Kiki pump on a SD-33t Nissan once that did what you described. It had set 5 years and it had a plunger stuck I believe. I rigged up a temp. tank with 50% atf and diesel and let it run for about 2 hours. Finally she let go with a belch of black smoke and ran fine from then on. Getting it started was another story, did you know you can push start a Funk Reversomatic?:rolleyes:
 

Pants

Well-Known Member
Joined
Mar 6, 2013
Messages
143
Location
Honolulu
... I rigged up a temp. tank with 50% atf and diesel and let it run for about 2 hours. Finally she let go with a belch of black smoke and ran fine from then on. Getting it started was another story...

Wait...what do you mean by "getting it started," if that's not what immediately preceded the part about "...let go with a belch of black smoke and ran fine from then on..."?

I wish I owned a Funk Reversomatic, just to tell chicks about it. Gotta google that.

--Dave
 
Top