MFI tuning and LM-1

[Jim727]

I'm interested in feedback on the importance of exhaust gas analysis in tuning MFI and in use of the LM-1 in particular.

I have been over CMA and found a minor electrical issue that will be fixed, but everything else is good (linkages, timing, etc). Mixture adjustment appears to be a major hurdle. I figure I can have it done on a dyno which is expensive, one-time, and not the same as road use, or I can invest in a portable exhaust gas analyzer and try to budget time I don't have for accurate tuning. What I don't want to do is the old "add a couple of clicks" method.

The LM-1 appears to be a good choice for this - correct?? The LM-1 comes with a single sensor bung, but for a 2.2E it appears that two bungs for independent checking of left and right cylinder banks is necessary - correct?? The alternative is the "venturi" sampler that goes in the exhaust outlet, but that strikes me as restrictive and likely to affect tuning results.

Finally, there is an RPM module that goes with the LM-1. How important is that to proper tuning.

I'm relatively new to the MFI cult and trying to heed Grady's advice to become self-reliant.

Feedback appreciated.
Jim

[frogger]

My opinion of the Gunson gaster that I have is that it is only so-so for tuning MFI. I've often wondered about replacing it with an LM-1 for this purpose. I recall John Cramer tackling this LM-1 / MFI issue a while back. Perhaps we can get an update.

[304065]

Jim, I assume you have read my posts that detail the installation of an LM-1 on my 71E.

You need two bungs, one left and one right. You can swap the sensor side to side, but it's more accurate if you know you have at least half the engine at the right AFR.

Actually, when you have the correlation on one cylinder wrong you can usually SEE it on the LM-1, the last bar of the mixture display flickers at low RPM.

Ideally you would use a pair of LC-1s to see both sides and log the whole thing so you could see if you have the cross links adjusted correctly. This would work well to see how the mixture changes when you come off-idle and transition from the idle mixture to the main mixture adjustment.

RPM logging is a must to verify your mixture settings at idle, part-throttle cruise and WOT under load.

In order to see the value here, you need to have the system working PERFECTLY as well as you can without the LM-1. That means a thorough pass through CMA and the extensions-- every component must be working correctly or you will be chasing your tail. This means no slop in the linkage rods, the best throttle bodies you can find, all gaskets in place, correlation set with a synchrometer accurately, etc., no excuses. Then you can start looking at the AFRs and optimize for idle, part-throttle and WOT with the LM-1.

I'm going to purchase their new 2-1/8" analog gauge so I can have a dynamic picture of what's happening during these conditions. I'm also going to hook up a MAP sensor, these are inexpensive, and maybe temperature logging of thermostat inlet temp (easy to do).

Of course the ultimate is closed-loop but I don't think anyone has made any progress, too much attention on EFI these days.

[Jim727]

Frogger: Thanks for that - looks like Gunson is out.

John: I remembered the thread that mentioned the flickering readout, but couldn't find it in a search; now that I know you were the author I should have more success. Thanks for the reinforcement on the dual bung requirement; seemed intuitively obvious, but "prior planning prevents pi$$-poor performance".

I have gotten it Loud and Clear that meticulous attention to detail is essential in tuning MFI. Skip no step. Cut no corner. Do it right or don't do it. Stimmt!!

Thanks also on the RPM logger; I'll plan for a bigger dent in my bank account. Seems also that with two bungs it would be helpful to get a second sensor (the LC-1?) and rig a switch so readings could be taken under conditions as close to identical as possible - agreed.

Throttle bodies are good and tight; ditto linkages. It's new on the car so the gaskets are also fresh. Injectors are new. The pump was rebuilt by Gus and has all of 4K miles on it. Biggest question mark is that the engine is (now) a 2.2E but the pump is a 2.0S. The consensus on the forum is that doesn't significantly affect the performance of the engine, but if I had a choice, the pump would have been from a 2.2E. I do like the dual solenoid setup, however, and want to make sure that circuitry is completely correct.

Jim

[Jim727]

Originally Posted by john_cramer

Jim, I assume you have read my posts that detail the installation of an LM-1 on my 71E.

You need two bungs, one left and one right. You can swap the sensor side to side, but it's more accurate if you know you have at least half the engine at the right AFR.

Actually, when you have the correlation on one cylinder wrong you can usually SEE it on the LM-1, the last bar of the mixture display flickers at low RPM.

Ideally you would use a pair of LC-1s to see both sides and log the whole thing so you could see if you have the cross links adjusted correctly. This would work well to see how the mixture changes when you come off-idle and transition from the idle mixture to the main mixture adjustment.

RPM logging is a must to verify your mixture settings at idle, part-throttle cruise and WOT under load.

In order to see the value here, you need to have the system working PERFECTLY as well as you can without the LM-1. That means a thorough pass through CMA and the extensions-- every component must be working correctly or you will be chasing your tail. This means no slop in the linkage rods, the best throttle bodies you can find, all gaskets in place, correlation set with a synchrometer accurately, etc., no excuses. Then you can start looking at the AFRs and optimize for idle, part-throttle and WOT with the LM-1.

I'm going to purchase their new 2-1/8" analog gauge so I can have a dynamic picture of what's happening during these conditions. I'm also going to hook up a MAP sensor, these are inexpensive, and maybe temperature logging of thermostat inlet temp (easy to do).

Of course the ultimate is closed-loop but I don't think anyone has made any progress, too much attention on EFI these days.

John -

I'm preparing to order the LM-1; looks to be an essential piece of equipment. Could you please post pix of where you have the O2 bungs welded onto your exhaust? It looks like there are only a few options that don't interfere with something.

I like the real-time gauge idea. Seems that the ideal setup would be a dual O2 sensor with individual readout.

Jim

[304065]

Jim, here you go.

LM-1 Installation in Early 911- MFI

That was the original install in a Sport Muffler. Subsequently I have a triple-out Rallye style made by Ben at M&K, with dual sensor bungs welded in the inlets. You don't want the sensors pointing toward the sky, water will collect and ruin them (and they would hit the track anyway) so horizontal is best.

I'm going to be adding an AFR gauge to the dash and a second LC-1 for the other side, this would be a perfect check of side-to-side correlation. . . it's going to be a busy April. . .

[MBEngineering]

HI Jim
another alternative to John's sensor fitments could look like this;



these are fitted on a 2.2 engine.

just a thought.

regards mike

[shbop]

+1 on the Innovate LM-1. Be sure to include the RPM adapter. It's very helpful and doesn't add considerably to the expense.

[Jim727]

Many thanks, guys. Interesting different approaches. Looks like mounting on the collector pipes is more permanent, but mounting on the muffler saves on having to remove the SSIs.

John, I love the dual sensor idea - are you planning a dual-readout gage, or do you plan on using a switch to select the sensor in use?

Mike, it looks like you have bungs in both the collector pipe and in the muffler exhaust. Reason?? Do you get different results from the different locations?

shbop - yes, have already decided on the rpm adapter. It's enough of a price bump that I had to think about it, but it does look very worthwhile.

Jim

[beepbeep]

My 5c:

- Venturi sniffer in exhaust pipe works fine. I use it almost every time. Only drawback is that it *might* show little leaner values at idle than usual. It goes away as soon as youprod the throttle.

- Dual bungs sound a bit overkill unless MFI can be adjusted "per cylinder bank". Also, WBO2 sensors must be heated all the time regardless of use. Which means you cannot just install two sensors and switch inbetween. Disconnected sensor will foul up during the drive.

- RPM is a must. Frankly, I believe AuxBox is a must to. I log at least AFR, RPM and MAP. Just connecting LM1 and logging AFR only will tell you little to nothing.

- I find LM1 a little cumbersome for permanent install. It's best used as diagnostic tool. LC1 is better for permanent install but sucks as diagnostic tool. Nothing but nice light display in cockpit that gets old to look at after two days.


So I believe a venturi sniffer, LM1 + LMA3 is perfect solution. Attach the sniffer, connect LMA3 to log MAP + RPM (and whatever else you like), do few runs and adjust your fuel system until it's OK. Then you can remove it. Nothing will change if everything is OK. Get an extra LC1 and weld a bung if you like to watch colorful lights (but you'll better look at road ahead )

[mjw]

2c added to Goran's 5c:

I use the venturi sniffer, LM1 + LM3; currently I'm just logging AFR and rpm. Also, I use an inductive pickup for rpm measurement. This was fairly straightforward and quick to set up and provides perfect data for making those final part load and idle MFI adjustments. It takes all of the guessing out of it.

As is always the case, this should be the last step in your CMA process.

[beepbeep]

PS.

On non-turbocharged cars, MAP (MAnifold Pressure) is almost the same as throttle position. So by logging MAP you know how much you pressed on the gas pedal.

Knowing RPM, thottle opening and AFR will carryyou long way in order to fine-tune the MFI. Heck, it will be fine for EFI tuning as well

The reason I like LM-1 + LMA-3 is that you can use it on open road as a "pocket dyno" by using accellerometers. Dyno pulls give somewhat synthetic results.

[MBEngineering]

HI Jim

"Mike, it looks like you have bungs in both the collector pipe and in the muffler exhaust. Reason?? Do you get different results from the different locations?"

we have 3 plugs in the exhaust, 2 in the heat exchanger and 1 in the tail pipe, one on the other side in the same poss', so we can check each side and as I only have 1 LM-1 tester we use it in the tail pipe, on the rolling road we use 2 in the heat exchangers, I have not cross referenced the readings from one to the other, will have to check next time at the rollers.

regards mike

[Jim727]

Many thanks.

beepbeep - do I understand that the O2 sensor is electrically heated? I like the venturi idea (though not the price), but I have both a standard (Bischoff) muffler and one modified by SSI to be a 2/2 rally sport. Have you tried the venturi on a rally muffler and seen left/right differences? Also, even though the cylinder banks might not be individually adjustable, it seems that having readings for each bank provides another diagnostic data point.

Mike - I would really like to hear any results you have on comparing the collector readings to the single-out readings.

mjw - yep. Grady has driven it home here: 1) CMA; 2) CMA; 3) CMA.

Jim