Help with fuel system tuning/troubleshooting

[SQLGUY]

Hi folks,

Now that I've got my basic ignition set up well enough to get the engine running, I could use some (maybe a lot) of help with understanding the fuel side of things.

Here's where I am so far:

Yamaha SECA 750 (air cooled, transverse four).

I'm using port injection with injectors and a fuel rail taken from a junk yard 1999 Hyundai Accent. They are Bosch, or Bosch-style, injectors rated at 14.25 lb/hr. I have fabricated an intake manifold from copper pipe, and attached a throttle body from a 1997 Ford Escort. The fuel pump is new, it's an in-tank Bosch unit originally intended (according to the eBay seller) for Harley Road Kings, but marked Saab. The fuel pressure regulator is the one that came with the Hyundai rail and injectors.



EGO is currently disabled. IAC is set to PWM warm up with default settings. I have an oil temp sensor but it is not yet installed in the crankcase, so the ECU always thinks the engine is cold. IAT is hooked up.

I was initially not able to get the bike running. Some of that was ignition problems (now solved) and some was fuel problems. What I found is that if I set the injector flow rate to 5lb/hr or less I can get the bike running and it will idle OK (a bit uneven, but stable). Of course, this is about a third of what the flow rate should be for these injectors. I have verified that the bike will not run if I recalculate required fuel based on 6lb/hr or more.

With the 5lb/hr, the bike idles a bit high (around 1800 RPM) and this will actually drop if I try to give it more throttle. Too much throttle and it will start misfiring.

If I set the required fuel based on 3lb/hr, the idle comes down a bit, but the engine will respond to throttle, a little bit, still not able to rev much.

I pulled the injectors at one point and test fired them. They seem to spit more than emitting the fine spray I would have expected. I also pulled another set of the same injectors from the junk yard. I test fired one from that set as well, and it also spat the same way. The plugs are new, but after a while running they are fairly sooty. I have not yet measured the rail pressure (as I don't have a gauge and was hoping to avoid buying yet another tool for this project, but...)

Ambient pressure, per the MPX4250, is about 79kPa. When idling, the map pulls down to somewhere in the 65-68kPa range. I really don't understand how I'm supposed to use these numbers in the generate VE table tool. Does any of this help me come up with a better number to put in for map at idle in the tool?

Suggestions, thoughts, ideas from here?

Thanks,
Paul

[SQLGUY]

OK, I still don't understand what the VE table tool is looking for, but I do know what my main problem is: the hose from the fuel pump to the tap on the tank had come off. Basically I had 6" of gravity, and the offset of manifold vacuum, providing pressure to my injectors. I'm pretty amazed it was possible to get the engine running at all.

Tomorrow I will redo the tap system I had set up for the tank so that I can make a more reliable connetion of the in-tank hose. These in-tank fuel pumps may be nice for manufacturers, but they're really a pain for retrofitters!

Also, for future reference of anyone else searching "fuel injector testing": spitting, rather than spraying, and only going two inches or so, are not good signs when test firing injectors.l

[SQLGUY]

Good news and bad news.

The good news is, after changing my fuel tank setup, I am now getting good pressure on the rail. The bad news is it didn't make nearly as much of a difference as I thought it would.

If I tune required fuel based on the theoretical flow rate of the injectors (14.25 lb/hr), the engine seems to flood immediately. If I tune based on something like 24 lb/hr, I can start and idle well, but get no throttle response other than misfires. I can see the pulse width increasing slightly as I turn the throttle, but there's no corresponding rise in RPM.

Ignition looks OK.

The EGO and CLT are now hooked up. EGO will sit at .47 or so at idle, but will dive to 0 when the throttle is opened at all.

[Matt Cramer]

Sounds like it doesn't have enough acceleration enrichment. You also might want to add a wideband oxygen sensor instead of a narrow band.

[SQLGUY]

Sorry for keeping two threads going, but I'm still having difficulties both with ignition and with fuel.

The ignition side is OK for the time being by bypassing uS and tacking back in the stock TCI module.

The fuel side still has me puzzled. The EGO shows that I'm always running lean (about .25V), but my plugs are suity black, and the exhaust certainly smells plenty rich. The EGO is a new Bosch four wire; it's installed in the blending box about half way down the exhaust.



I understand that a wide band would make things easier for overall tuning, but shouldn't this narrow band be accurate at idle with no load? I am still setting things up here, so the bike has not been used under load at all yet, it's just been on the center stand in the garage.

Also, just as more information, which may not be related to mixture: if I rev the engine to about 2500 RPM, it settles back down to idle (1050) pretty normally. If I rev it over 3000RPM, it will take off and stay at about 5000, unless I reduce the idle setting... then it will settle down further and die. Any ideas what this might mean? Could it be caused by insufficient compensation for voltage variations? Keep in mind that the battery is quite small on this bike, and the alternator is only rated for 19A.

Thanks,
Paul

[Matt Cramer]

That blending box could very well have an air leak that creates a false lean reading.