Timing on Mitsubishi KE70 engine

Ilikeurtractor · Apr 19, 2017

Anybody ever successfully test/adjust the timing on their KE70 engine (Bolens G15x, Iseki TX1300)? I played around once with this before on a KE70 that originally came with a TX1300F but it seemed like that one checked out ok. I just rebuilt another one in a Satoh S370D and the fuel pump had a paper gasket under it which is a sign someone likely threw away the original metal shim(s) and decided to install a gasket there instead. So I felt compelled to do a timing check to get reasonably close to the 25° BTDC spec. The problem is the timing never seemed to be anywhere near 25° BTDC no matter how I checked it. The repair manual says you can do it a couple of different ways. One is to take out the delivery valve in the pump and put the holder (nut) back on the pump. Then you turn the crankshaft and watch when the fuel stops coming out of the holder and check it against the timing marks. Well, the first time I did this I was getting more like 20° ATDC

and I did this a few times to confirm. Incidentally, I had a 0.026" shim under the pump while doing this. According to the manual, every 0.004" difference changes the timing by about 1°. By the way, if you use this method as soon as you take out the delivery valve fuel will continually gravity feed from the tank and right through the pump so you need to manipulate your fuel line shutoff accordingly. Anyway, I then went on to the "alternate" method of checking the timing by disconnecting the line at the injector nozzle assembly and watching when a drip forms at the end of the pipe. Once again, 20° ATDC. Then I took out the injector nozzle, hooked up the supply line, and rotated the engine by hand until the injector popped. And again, 20° ATDC. Then I started thinking maybe when I put the engine back together I didn't get the punch marks correct on the crankshaft and cam gearing so I double-checked that and it was correct. Additionally I confirmed the intake valve on the #1 cylinder did indeed open at 18° BTDC and it was dead-on. So I'm very confused as to what is actually happening. Looking at the cam lobes, the #1 cylinder intake lobe should be about 180° out from the #1 injector pump lobe (technically it would be 176.5° offset in the normal direction of rotation) and this does not appear to be the case. It is clear the centerline of the lobe has a much greater lagging angle than 180°, closer to 200° which corresponds to the readings I was obtaining. I wish now I would have documented this when I had the camshaft out so I could attach pictures, etc. but you can get a feel for this by just looking through the fuel pump port at the injector lobe locations and relate them to TDC on #1 cylinder via the crankshaft key position which is at 3 o'clock at the same time.

So, unless I'm totally missing the boat on this one (which wouldn't be the first time), then the timing on this particular engine is off from spec by a whopping 45° in the lagging/retarded direction!!! And there would be no way to get it within spec even if all the shims were removed. At best, I could do about 13° ATDC. Is it even possible for the engine to run at this setting??? I suppose it very well could. I don't play around with timing on diesel engines enough to have first-hand experience of how sensitive they are to timing but I think most people who do would scoff at such a claim. The engine does run by the way, and seems to run like the rest I've seen. However, I never knew why the other KE70 I have always knocked so much compared to the others and I wonder if the fine engineers at Mitsubishi decided to adjust the grind of the cam to force the timing retarded to quiet the engine down. My other KE70 is a much earlier serial # so it may not have had the alteration made. Again, probably another stretch but it would explain some things. Just throwing this out there in case others may have seen something similar. I get the impression that generally people don't mess around with the timing if the engine appears to be operating correctly so it may never be noticed and realistically it should never change much anyway as long as the factory shims are intact.

thepumpguysc · Apr 20, 2017

The ones I've checked, {never had to reset one} give a pump camlobe height dimension, not a degree dimension.
You measure the cam lobe at base circle, 0 lift.. w/ a depth micrometer.. rotate the engine to its highest lift & measure again..
There should be a height spec/port closure for the injection pump timing.. & you do alittle math & add or subtract shims accordingly.
Its been awhile tho..

Ilikeurtractor · Apr 20, 2017

I've done the height check, and mine was within reason - I think I measured 2.986" without shims. Spec calls 2.992" (76 mm) +/- 0.002" to the minor diameter from the pump base as a starting point for shims. However, the angular position of the lobe on the cam relative to where the crankshaft is has more influence on timing than the difference in diametrical distances. It is good to check this dimension though. Thanks for pointing it out.

thepumpguysc · Apr 21, 2017

OH.. One thing I forgot to mention & I don't think its written in the manual, & IF IT IS its not highlighted..
The pump rack MUST BE in the run position.. if you have your solenoid in position, the sol. is pushing the rack to "no fuel" & will DRASTICALLY change the timing reading your getting..
I sure hope this helps.. TPG

Ilikeurtractor · Apr 21, 2017

That's a good point. I don't have a solenoid but now that you mention it I don't know where the rack position was on the pump when I tested it. Generally the governor works towards adjusting the rack to more fuel when the engine is stopped but I don't think it goes to full rack on these engines. I do know the pump will deliver some fuel even in the minimal rack position. Then the question is, does the fuel delivery start point change depending on rack position or not (?) I would think the start point timing would stay the same, just the end point of fuel delivery would change.

Ilikeurtractor · Oct 6, 2017

I acquired another Bolens G154 recently that was disassembled so I thought it would be a good opportunity to check the timing on this particular KE70 engine against the last one I did. Instead of just rotating the crankshaft by hand I thought about using a different approach. So armed with a trusty GoPro Hero 4 black camera that is capable of shooting 240 frames per second I headed out to do some experimenting. The results can be seen here:

Full rack injection timing test - YouTube

I believe this confirms the timing to be retarded much more than what it was originally intended to be. The values in the video could have an error up to +5 degrees (each frame is about 5 degrees apart so if the injection started 5 degrees sooner than the first frame that shows it, it would be missed by about that much). It was interesting to find the injection pulse duration lasted about 15 degrees, or three frames from where it can be first detected. Therefore once the pop is over you're 15 degrees later than the start, at least at the speed it was turning at in this test.

I did not verify the shims on this particular engine, but it doesn't appear to have been taken apart in its lifetime so I'm making the assumption it has the original shim(s). To get to the timing I'm seeing would take a fairly thick shim (around at least an 1/8") which would be obvious and I'm not seeing that.

So why is the timing the way it is? I don't know for sure of course, but as I said earlier, I suspect it quiets down the engine for one (but why then are other 2-cylinder diesel engines as quiet with "correct" timing?) I can see where it might make starting easier. It might prevent bent rods (upon starting or low rpm running where the piston might not be past TDC before combustion). The downside is a less efficient engine, more black smoke or smoke in general, I think higher exhaust temps, to name a few. What doesn't seem to make sense though is the engines seem to hit their intended speed mark with no issues. Generally you need to advance the timing as speeds go up and with it as apparently late as it is, I'm surprised they run as good as they do at high rpm. If I get the opportunity I'd like to do this test on the K3 engines, or even other engines. I don't have any of those needing to come apart yet though and I'm not curious enough to strip one down just for this test.