How important is injection pump timing in an -002a/-003a, really?

[sewerzuk]

The TM says it is "critical."
For those that have tried to load test and/or operate their sets with a known timing problem, how did your set run? Was your timing too early or too late, and by how much?
I have swapped quite a few IP's on -002's and -003's, but I've always taken the time to swap in the correct button to get the timing withing 1/2 of a degree mark. I have an -002 in my shop right now that is running 1.5 marks (3 button sizes) too late. I was just going to wait for the correct button, but I thought I might load test it tomorrow just to see how it actually runs. Under no load, it seems to run exactly like a correctly timed -002a...
Is 1 button size OK? Two? Where is the line where the set simply doesn't run well enough to be considered running properly?

 

[Triple Jim]

That's the spirit! I'm interested in the load test results. Wild guess: late will reduce overload capacity, early would tend to smoke at high load, it will do pretty well.

 

[ETN550]

The TM says it is "critical."
For those that have tried to load test and/or operate their sets with a known timing problem, how did your set run? Was your timing too early or too late, and by how much?
I have swapped quite a few IP's on -002's and -003's, but I've always taken the time to swap in the correct button to get the timing withing 1/2 of a degree mark. I have an -002 in my shop right now that is running 1.5 marks (3 button sizes) too late. I was just going to wait for the correct button, but I thought I might load test it tomorrow just to see how it actually runs. Under no load, it seems to run exactly like a correctly timed -002a...
Is 1 button size OK? Two? Where is the line where the set simply doesn't run well enough to be considered running properly?

On a low BMEP prechambered diesel like the 002 there may be no noticable difference. Retarded timing in general will increase smoke, increase exhaust and exhaust valve temps, decrease peak cylinder pressure, decrease fuel economy, and create harder starting. A lot of the time only operation at the extremes will make these effects observable. Such as running at the max load, max altitude and max ambient temp. Conversely, more difficult cold starting in extreme low temps. I'll bet you see no obvious difference. But do put an infrared temp gun on the exhaust elbow and compare to correct timing under same load and inlet air temp. In the lab on a test stand, under controlled conditions fuel consumption, Bosch smoke, and temps can be monitored precisely and the Engineers can hone in on the best timing.

I had an old 12 valve Cummins in a Dodge and advanced the timing from 13 to 16.5 and saw no measureable difference, EGT, or seat of the pants. Fuel pump was lightly modded.

 

[Triple Jim]

I found this informative page: http://www.oliverdiesel.com/tech/timing.htm

 

[storeman]

When I first started salvaging generators I didn't understand the need to get the closest numerical combination to the IP and button being replaced. Either I got lucky or it didn't make much difference.

 

[sewerzuk]

Well, I really like theory. Knowing what should happen is always a good thing.
I also place just as much value on experience! Knowing what ACTUALLY happens often differs from what I expect, because I overlooked some little detail or didn't think about some interaction.

So, I plan to load test that generator with the late timing this evening. I will make a note of the peak power it can output before bogging down, EGT's under a steady 100% load, and visible smoke.
In a couple of days, I will re-time it to the correct specs, and do the same exact tests. While not scientific, I'll bet that I learn something!
I decided not to compare it to other running -002a's, simply because there are too many other variables at play to call a specific difference related to timing. That, and I currently only have one other running -002a.

 

[Triple Jim]

Sounds scientific to me, just make note of the ambient temperature both days.

 

[sewerzuk]

Done!
All measurements were taken with a Fluke DVOM. Ambient temp was 48 degrees F.
I slowly added load until I got to 58 amps (about 7kw).
I started at 61 hz, set ended up at about 56.8hz (a little more droop than I am used to seeing, but not completely unreasonable).
Zero visible smoke from the exhaust under all loads.
I sat with an IR thermometer pointed at the exhaust elbow, and watched it until it stopped climbing (about 5 minutes). The highest temperature I could find was 402 degrees F.

Maybe it is just my imagination, but the exhaust sounded like it had a little more "bark" than normal (makes sense, but then again...maybe I just noticed it because I was expecting it).

At any rate, if I didn't KNOW the timing was 3 button sizes too late, I would never have noticed. The set ran just fine, and was able to pull well over its rated load. Of course, I wouldn't know about any long-term effects of running this way (I would assume accelerated exhaust valve/seat wear would be the most likely thing).

As soon as the correct button arrives, I will install it to correct the timing, and run the same tests. It will be interesting to see what changes!

 

[LuckyDog]

opcorn:

 

[sewerzuk]

In case anybody is really biting their fingernails...I haven't forgotten about this project! I did get the correct button, and I have re-timed the engine. It is now about 1/4 mark from PC. I get the air purged out of the lines and engine started, but didn't have time to run the load test due to a few other pending projects with deadlines. I should have time to repeat the load test with the correctly timed IP sometime this weekend. I will be sure to post my findings as soon as I have them.

 

[Triple Jim]

Thanks for the update. I'm looking forward to the results.

 

[sewerzuk]

second load test done! Here are the results:

58A resistive load
Ambient temp: 61 degrees
Frequency droop: 61hz (no load) 59.4hz (full load)
Zero visible smoke
Highest EGT recorded 396 degrees
noticeably less "bark" in the exhaust note

For reference, here were my findings with the engine when it had the late timing:
58A resistive load
Ambient temp: 48 degrees
Frequency droop: 61hz (no load) to 56.8hz (full load)
zero visible smoke
highest EGT recorded: 402 degrees

The obvious biggest change I noticed was the change in frequency droop. This tells me that, with late timing, the engine required the governor to provide significantly higher correction. It obviously had less horsepower, and would be using more fuel.
The EGT difference, corrected for ambient temperature, was about 19 degrees. As expected, the engine had higher EGT's with late timing. This may translate into premature exhaust valve and seat wear.
Also, the "bark" I noticed in the exhaust tone was gone with the correct timing. It's hard to make a definitive judgement without some way to measure this, but even on different days I could tell the correctly timed engine was quieter.

So what does this mean? A correctly timed engine does indeed run better. There is a measurable difference in horsepower, fuel consumption, and EGT's between an engine with timing that is 3 buttons too late and one that is within 1/2 button. However, an incorrectly timed MEP engine will still run fine and produce its rated power. If I get the chance with one of the sets that are currently in my shop, I will test how a larger timing mismatch affects engine performance. I would also like to find out how advanced timing may affect the engine. To be continued...

 

[ETN550]

second load test done! Here are the results:

58A resistive load
Ambient temp: 61 degrees
Frequency droop: 61hz (no load) 59.4hz (full load)
Zero visible smoke
Highest EGT recorded 396 degrees
noticeably less "bark" in the exhaust note

For reference, here were my findings with the engine when it had the late timing:
58A resistive load
Ambient temp: 48 degrees
Frequency droop: 61hz (no load) to 56.8hz (full load)
zero visible smoke
highest EGT recorded: 402 degrees

The obvious biggest change I noticed was the change in frequency droop. This tells me that, with late timing, the engine required the governor to provide significantly higher correction. It obviously had less horsepower, and would be using more fuel.
The EGT difference, corrected for ambient temperature, was about 19 degrees. As expected, the engine had higher EGT's with late timing. This may translate into premature exhaust valve and seat wear.
Also, the "bark" I noticed in the exhaust tone was gone with the correct timing. It's hard to make a definitive judgement without some way to measure this, but even on different days I could tell the correctly timed engine was quieter.

So what does this mean? A correctly timed engine does indeed run better. There is a measurable difference in horsepower, fuel consumption, and EGT's between an engine with timing that is 3 buttons too late and one that is within 1/2 button. However, an incorrectly timed MEP engine will still run fine and produce its rated power. If I get the chance with one of the sets that are currently in my shop, I will test how a larger timing mismatch affects engine performance. I would also like to find out how advanced timing may affect the engine. To be continued...

Excellent work. Completely believable results. The frequency drop at full load retarded is strong indication that a lot more fuel is needed to maintain the set point.

I don't know much about the 002 and 003, and every design is different to a degree, but if they have aluminum pistons too much advance under high load, maybe even moderate load can melt a hole in the top of the piston. Despite lower overall exhaust temps, the peak firing temp and pressure in the cylinder will increase. The retarded "bark" from the exhaust opening might be replaced by increased cylinder knock, especially when cold. I saw these results all the time at CAT when customers tried advancing timing on the smaller truck engines 3306 in particular, in an effort to run with the bigger engines. Other than that detail, and increased NOx emissions, (not smoke) I don't think there is a down side to advancing the timing. Given the conservative nature of the engine it is hard to say if the piston problem would be real or not. The frequency drop is a great indicator of the injection quantity required to maintain the constant load.

Since the military engines are non emission compliant it is possible that their settings are optimized for all around performance including climatic conditions, ability to start, fuel economy and longevity.


Based on your valuable information, I would not be afraid to run a unit retarded until I was able to get the right parts and make the timing correct, even if it took a while. The retarded exhaust temp is still very very low and should not be a concern long term. JMO.

 

[Jimc]

This is a very interesting thread. Great info. Just curious also if the actual pointer was checked to see if it was set right. I assume they are all set at the factory and not in a standard location. One would assume they should be set perfectly but with how accurate they need to be you never know i guess. Only has to be off a fraction and that can translate into 1-2-3 button sizes off. I set the dial indicator up on mine just to be sure. I changed the flywheel out and you would think the marks would be the same but it was off and i had to readjust. I have 2-002's and 1 more 003 to do now and im gonna throw the dial indicator on them just to be sure and see how accurately they are at a factory setting.

 

[steelypip]

So advance the timing too far and blow holes in the pistons. Makes sense, I guess. That's the same thing that happens to an Otto-cycle (sparkplug) engine when you advance the ignition timing too far as well.

So the question is: will you keep getting a lower BSFC (better fuel consumption) if you keep advancing the timing or is it optimized? I suspect, given that Onan designed the engine to be a generator (constant 1800 RPM) and given that fuel consumption is regarded as important by generator manufacturers, you won't see a significant further improvement in fuel economy (reduction in droop) with further advancing of the timing.

But I am very curious to see what the evidence shows, because we all know the old saying about opinions. Pity it's such a pain to swap buttons compared to loosening bolts and turning things or adjusting a lifter...