eRod runs better without the MAF

[davepl]

I was having trouble with my new eRod running very lean at idle (> 18:1 AFR). On a hunch I unplugged the MAF and sure enough, it now hovers right around 14.7:1.

My MAF sensor is in about the middle of the pipe in the picture. I can't imagine this pipe is too "bendy" for the MAF, but it doesn't run well at all with it connected.

So, any suggestions on how to improve the air inlet?

Or should I just leave the MAF off and run in speed density? I'd prefer to have the MAF operational, but right now it only makes things worse!

Any ideas?

[BR3W CITY]

Its only worth running in speed density if you plan on building the rest of the tune around SD. Its fine in the short term to get it somewhere, but you need to put the time into a dialed VE table for SD to shine and feel as good or better than normal MAF operation.

[Custom 68]

check all your fuses, I had a similar issue and it was a fuse. In my case the same fuse runs the MAF and o2 sensors.
I have gone to speed density for my case but like brew said it is really only if you really need it, your MAF location should be good.

[Wasted Income]

That MAF belongs in the trash can.

Regards,
-SD Tooner

[boostedc10]

Originally Posted by Wasted Income

That MAF belongs in the trash can.

Regards,
-SD Tooner

Troof rite thur!

[davepl]

I would imagine GM has the speed density tune pretty spot on - they sort of have to for emissions. Of course that's an assumption I could be way off. But I just can't figure out what it wouldn't like about my MAF.

On the fuse issue, do you mean fuses in the eRod fusebox? Are there O2 and MAF fuses separate inside of that?

[boostedc10]

Davepl on a more serious note if you are trying to keep the MAF can you get a scan of the car with HP tuners? If so you can email it to me and I can take a look at it.

[davepl]

I loaned my box out but I do have one. I know the MIL is on (it's grounded) and I suspect that's going to be "LEAN O2" on both banks, but we shall see. I'll get the box back and if I can't figure anything out I'll ping you!

[Custom 68]

I apologize I hadn't paid attention that this was eRod, I am also not familiar with that. Mine was a modified harness that has fuses for several of the different parts needing power, one was the MAF.
I just pointed that out as sometimes the simple things elude us. This did me anyway. My truck was still running but was acting off, I kept it going then changed the o2 sensors before I realized it was something simple like a fuse. that fuse block on mine is not right out in the open so I wasn't even looking at it.
Good luck with it.

[davepl]

I connected HPTuners and it showed, as I expected, both banks lean. It also showed a "Pending" MAF voltage low, but I checked and there's 12V present at the IGN pin of the MAF connector. I know the MAF is being read, as you can annoy it by sticking your hand in there, which will stall it.

It's possible the MAF Voltage Low is a hangover from when I had it disconnected, though I did reset DTCs.

I tried a 4-foot extension on the end of the intake pipe to make it a super-long straight shot and that made no difference, so I don't think its my intake tract.

Given that there's full voltage present, I doubt it's a fuse. Since it's reading -something-, just too little of it, I know it's able to read (not a cut wire, for example).

The MAF is cycling around 0.4-0.7 lbs/min (or whatever the units are), as the engine "hunts" and runs lean. Runs near perfect with the MAF disconnected.

Any ideas? Could the MAF be bad from new? That just seems unlikely.

[boostedc10]

What is the MAF code that is pending? Is it a p0102 for the low input?

[davepl]

Yes, P0102. That in turn leads to P0171 and PO174 (System too lean, left and right bank respectively).

I also have "Crankshaft Position Not Learned" yet, but that's to be expected until I get it calibrated.

[Heavymetl]

Originally Posted by Wasted Income

That MAF belongs in the trash can.

Regards,
-SD Tooner

Exactly. I hate MAF and its soggy response. SD all day long.

[davepl]

I see a number of MAF-bashing posts, and it makes me wonder how hard I should be trying to solve this, given it runs well without the MAF.

For a completely bone-stock setup I have to imagine the SD tune is pretty spot on. But I really don't know.

[donandal]

I don't know your e-rod version, but my instructions caution against installing the sensor backwards. i.e. Flowing the wrong direction. When I welded the bung on to my aluminum intake tube I checked the flow direction a dozen times to be sure.

[BR3W CITY]

Originally Posted by davepl

I see a number of MAF-bashing posts, and it makes me wonder how hard I should be trying to solve this, given it runs well without the MAF.

For a completely bone-stock setup I have to imagine the SD tune is pretty spot on. But I really don't know.

This might get long, but hell, I haven't written a short post in a long time.


Fwiw; you mentioned that "GM probably got it right"....and thats not wrong, but its the wrong mindset. When GM engineers do calibrations , they are doing so off of hundreds (often THOUSANDS) of hours of logs, step tests, and failure tests. Designing a VE table that works reliably and has very large margins for safety means that table has to work in thousands of vehicles with that engine (whether its an E-rod or any other GM-built motor). GM isn't building those tunes for power, they are building them for manufacturing goals.

Now, how that specifically affects VE; in a stock configuration (non-custom OS) the VE tables are populated based on those thousands of hours of testing I mentioned above. The VE has a limited resolution (standard ve, not GMVE or VVE), which means that the highest # has to be enough for the highest-expected value @ that particular cylinder volume...and that the lowest # has to do the same. BUT of course we all know that differences of up to 50hp (4-9%) aren't unheard of. So, its possible that the VE #'s populating your table are not only padded, but they could be padded even MORE if you got the runt. AND because you have both closed and open loop, your not spending THAT much time running on VE alone.

Seems like I'm saying GM doesn't know what they are doing, but here's the catch. In stock config, the ECM isn't using VE alone; the MAF and the 02's act as an input-feedback loop. Its saying "I did this, so you should have that"...and if it doesn't see that, it makes changes....or throws a code if those differences are beyond what it can account for. Those "padded" VE #'s can be compensated for by that loop, and the engine is happy. To add to that, the IAT is in the MAF (usually) and the IAT data is further used in combination with the MAF reading and the 02 sensor trims to calculate air density. This means that whether your in 112* ambient temps at sea level or 8* at the treeline, your getting an accurate air density calc, and thereby getting proper fuel control.

SO, then why do many of us preach the virtues of thine Speed Density tuning? (aside from just fitment reasons)
In SD, you no longer have the input-feedback loop. The tune tells the engine systems what to do, and they do it, no back talking. This means that much of the safety net is gone, but the user now has a more direct control.
Even in stock form there can be (small) improvements made to trim the fat, or in some cases richen up area's which were set too lean (usually emissions related).

When you start modifying things like cams, heads, valvetrain geometry, forced induction....you effectively change the way your engine moves air. Sometimes the MAF is a choke point, sometimes it simply isn't designed to meter the volumes of air that highly modded motors can move. In the case of FI cars its not capable of effectively metering air going from engine vacuum to 2-3 bar. Changing to SD means your only on those VE tables (with 02 loop if desired, which is a good idea). This fine control means you can run a tighter ship, making more power more efficiently...ideally. For guys running non standard intake types (carbed, ITB, nelson's crazy alien intake thingy), its the only option.

Here's the rub; all of the perceived benefits of SD are also its pitfalls. The amount of time it takes to do an SD tune properly is 2-10x longer than a regular MAF tune. That ultimate control over VE also means that the person pulling the strings BETTER have a good idea what they are doing. A bad VE cell in a MAF tune could be overlooked and "smoothed" over by the trims from the MAF/02 loop....a bad VE cell in an SD tune could mean pegging lean and detonating. If you were overlanding or going on Safari, where baro pressure and temp could swing wildly during operation (like HUGE), then SD would be at a disadvantage over the more flexible MAF setup.

TL;DR
SD can make more power and in some cases burn more efficiently, but is only a bridge worth crossing if you really need to....and the person at the helm needs time, patience, and a slightly higher-than-normal grip on the science going on to do so at peak effectiveness.



Now as for the E-Rod in specific; its tune will be great for the way it was sold, but it was sold as a package with the emissions stuff. It'll still have a VE table built for safety. Which ECM is coming with the E-Rod package? If its running the GenIV or newer stuff it may be using GMVE or Virtual VE (works the same, but way way more tables and a much more complicated multiplier being used). If thats the case, there are a few more possibilities for whats happening.

[davepl]

We should clarify terms - SD can still be:

SD closed loop - VE + MAP, IAT, CTS, etc plus O2 feedback
SD open loop - VE + MAP, IAT, CTS and hope it works

I will still be in closed loop so that allows up to 25-ish percent of correction, and I'm only seeing 304% STFT and LTFT.

So from your post I'd say:

- The SD tune is "close" out of the box
- The O2 feedback and trim tables gets it within the variance of eRod vs Camaro vs Corvette variances

Also it's GenV, I believe, with the big calculus integrator equations that I don't fully understand yet. Part number 19257231 or 19257235 depending on automatic or manual (mine's the automatic ECU but with a manual trans for now, I need to flash the tooth count).

I wonder why GM went to MAF on the 94 Corvettes when the 93 was fine. I know they made the change to Sequential at the same time, but I don't think it was related.

It sounds like a MAF is the cat's butt for modified engines, but I'm not really sure why they bother on stock applications when closed loop O2 should get them close quite quickly. Maybe not quickly enough for the EPA I guess!

[I've double and triple checked the MAF mounting to make sure it's correct. I do assume the arrow points in the direction of airflow (and not towards the filter or something goofy. ]

[Hart_Rod]

Originally Posted by davepl

Yes, P0102. That in turn leads to P0171 and PO174 (System too lean, left and right bank respectively).

On my last LS swap I was getting a P0102 code when I first started it. It turns out that 2 wires were "mis-pinned". I swapped the wires and the code went away. Also, did the air intake come with your Erod setup? If not, It will still require a VE tune in order to calibrate for your exact setup.

[BR3W CITY]

Dave, for the most part you got it. The only thing I'll add is that I'm not 100% sure if E-rods are coming with an "erod tune" or if it is in fact just the Camaro tune. I don't have a copy of an E-Rod tune to look at, as no one I know has to pass emissions

Its not direct injection tho....right? My only exposure in person on GENV Ecm's is a procharged Escalade, and the direct injection was.....different.
GMVE/VVE is cool for what it is, but I really don't understand why they had to make it so damn complicated. It does work around that whole resolution issue I mention in the long post.

[Hart_Rod]

Originally Posted by BR3W CITY

Dave, for the most part you got it. The only thing I'll add is that I'm not 100% sure if E-rods are coming with an "erod tune" or if it is in fact just the Camaro tune. I don't have a copy of an E-Rod tune to look at, as no one I know has to pass emissions

Its not direct injection tho....right? My only exposure in person on GENV Ecm's is a procharged Escalade, and the direct injection was.....different.
GMVE/VVE is cool for what it is, but I really don't understand why they had to make it so damn complicated. It does work around that whole resolution issue I mention in the long post.

Brew - I agree on the "making it more complicated" part. I think GM went with the VVE equations in order to save time and computing power within the ECM. Have you got the upgraded software for HPTuners that allows you to tune the VE like the old Gen3 stuff and then convert it to the VVE equations? I think it came out about 3-6 months ago.

[BR3W CITY]

Yea we have that. When we were doing the Escalade 3.0 wasn't out yet, and we had to use 2.25 (beta)....which was less than reliable. The only downside to the conversion-tool is that its much harder to beat the math. What I mean is; when you start playing with things like hugely modified stoich values, you can't just whip out the calculator quick and do an equivalence. Not to mention the fact that the ECM's running VVE are MORE powerful...why are they trying to save it on process load? I honestly don't know.

The Esc was the first boosted DI Esc in the country, and it was probably one of the hardest vehicles I have ever seen tuned. There were no HPT files (had to literally call them and have them issue a credit # for the 2015's), Procharger didn't acknowledge that the unit could work on the truck. The 8spd is a nightmare on a Dyno.

[davepl]

I'm going to try replacing the MAF - even though I think its a low probability that it was bad out of the box, for $30 on a $7000 setup (or whatever) I've got to try it.

I mean if its not the intake tract and the MAF is powered, what else could it even be?

[BR3W CITY]

The harness potentially, but again the fact that you were seeing voltage on it at all means its not cut/disconnected. A pinched wire would be a freak thing, or the ever-ominous "bad ground somewhere"....but thats going out of my way to chase less likely stuff.

[davepl]

Hmmm... it's a long shot, but the manual says there are three ground eyelets and I only remember hooking up two. I'll have to dig through the harness and see if I can spot a third.

[AnotherWs6]

Changing the MAF is something I would have done long ago. Don't assume your new, sensitive electronic part works while playing with everything else in the world and beating your head against a wall. Make SURE that your new, sensitive electronic part works and THEN beat your head against a wall.