EFI Help Needed in The Seattle Area

[jbscheff]

My 1973 911 has a 3.0L with a custom PMO EFI system and I need an expert to dial in the tune. It’s running way too rich. I need someone in the Seattle/Tacoma area who can make it right. Any suggestions?

Al at X-Faktory did the install and initial tune and then I took it in for dyno testing and fine tuning at Carb Connections. The car is certainly driveable and goes like hell, but fuel economy is not quite what I’d expect ( I know - a 911 hot rod isn’t supposed to be a fuel miser) and I occasionally get a backfire under certain conditions. I’ve also been told that driving behind my car smells like the 70’s. Looking at the AFR table there’s not a single value over 14.0. The engine specifics are:
3.0 with SSI’s
PMO ITBs
Mega squirt 2
MSD ignition
Timing is combo dizzy and ECU

Posting on PCA PNWR and OPR Facebook pages and here for maximum visibility.

Thanks, Jason

[Showdown]

Jason,

It sounds like your car has a basic “get it running” tune which is not really a tune meant for driving.

I can’t help you with a local tuner but read this thread, you can in fact get yourself a respectable tune yourself. It won’t be dyno perfect but it’ll be 90-some-odd percent there.

You’ll learn a lot about the car and the efi system in the process

https://forums.pelicanparts.com/porsche-911-technical-forum/1134997-efi-tuning-how-process-description.html


Sent from my iPhone using Tapatalk

[mb911]

Do you have a laptop in the car? Perform autotune? That feature works really well to smooth out everything. I did mine last summer and it was really pretty straightforward and easy. Could you have a bad CLT or air temp sensor or have not calibrated them properly? That would cause it to get pretty darn rich.

[jbscheff]

Thanks, I have read that post and several others - it’s true that there is a lot of information on this forum. I’ve run my car with the laptop hooked up for data logging but I’m reluctant to use the auto-tune. As I understand things, you’ve got to have a decent AFR table to start with. I don’t think I’ve got that. This weekend when I get a chance I’ll post images of my tables and the 3-d graphs.

I’ve had this system for 3 years or so, and it’s been on the dyno 3 times in addition to the road tuning that was done initially. As I wrote before, It runs pretty well and makes good power. But it shouldn’t be so rich when cruising and I should not have to worry about fuel burning in the exhaust. It could also use some improvement at idle. I need to get this thing dialed in and I need some help to do it.

[evan9eleven]

I assume that is power at the wheels, if so you've got good horses in any case.

Three times on the dyno is not just a get-it-running tune. Even the first time should have given you a good table for steady-state running, that is to say cruising. Acceleration enrichment really requires tuning manually, on the road. Drive, stop, tune, repeat. But that is a different animal. Are your backfires in the intake or exhaust?

Anyway, it was cruising you asked about. Best is to get a helper and your laptop, and get cruising. Note the RPM and throttle opening % as well as AFR at the speeds where you typically cruise. The corresponding cells at these points can potentially be leaned out. IIRC MS2 also has something called EGO control, with a good VE table in place you can turn this on and MS will make small corrections on the fly (not permanent, just real-time tweaks.)

If you do decide to try auto-tuning, just remember to save your current tune so you can revert back to where you started in case things don't go as you'd like.

Having said all that: I'm running MS2 and Al's retro-cool Triumph ITBs on an otherwise stock 3.0 with SSIs. Dyno'd it twice because I made mechanical changes to the throttle linkage after the first time. Then I spent two summers tweaking the accel enrichment to get it really streetable. Accel enrichment with ITBs is tricky, and it is easy to go very rich at low RPMs. In my experience, however, it seems these engines like a bit of a rich tune, and that is definitely what I have too. Behind my car it is 70's all the way... I average about 13l/100kms = 18 MPG. I'm going with headers and cats on the next engine and a different ECU, but that is a story for another day.

[mb911]

So I run carbs as TBs and inject into my MFI ports so a little different than what you have. I run ms2 and had a good afr target to start with. I use auto tune quite often when setting it up. I saved each tune in case I wanted to change back . My car runs really well and is not super stinky at cruise and way less than when I was running carbs. I would verify your getting correct sensor values as I had a CLT sensor go last summer and changed dramatically

[jbscheff]

It's got great power. My chief complaints are:

It runs so rich at cruise drivers behind me have told me they can smell it. "Nostalgic" they called it. But it also means it's certainly not as efficient as it could be and that's one of the advantages I should be seeing from my EFI. My mileage should be better.

I occasionally get backfires. They usually happen after I take my foot off the gas after accelerating. Since it's a rear engine car, I've got no idea if it's from the intake or exhaust, but I'm assuming exhaust? I haven't seen anything on the inside of the air intake to indicate it's coming back up through the stacks.

It could idle better. And sometimes when starting out lightly from a stop it seems to struggle.

Here are the tables and graphs for my current tune.







The spark advance is a combination of the distributer and the spark table

[Showdown]

With ITB‘s you spend almost all of your time in the 80% plus load range of that table where you’re AFRs are 13 or 12… no wonder you’re rich- you have the fuel map set that way.

There are two ways to self tune a car: using auto tune, and doing it manually. With auto tune you need an AFR table that is dialed in, and yours isn’t. Doing it manually you just have to adjust the fuel load until you get an AFR you want. Go out with somebody and have them drive while you sit with the laptop and adjust the fuel on the fly until you achieve an AFR reading that you’re satisfied with. Or log a Drive and then review the logs, make some adjustments, and go drive again. There are no shortcuts and it’s not terribly complicated, it just takes some time.

When you are on the highway, just cruising and not under load, lean the car out as much as you can before it starts backfiring. When you’re on hard throttle add more fuel until it starts bogging down. Play around with these and you will eventually find that tuning the car by fuel will give you really good results. Then you can take it to a Dino and calibrate it that last 5%


Sent from my iPhone using Tapatalk

[brighton911]

Jason, are you running a wideband O2 sensor? That really is the only way to know where you are at with A/F or lamba (better). MS2 can take the feed from the O2 and adjust mixture on the fly. Also, how much advance does the distributor contribute ? The advance shown on the table alone is not enough, figure on around 30 degrees total advance.

[gilbertjames654]

Originally Posted by jbscheff

It's got great power. My chief complaints are:

It runs so rich at cruise drivers behind me have told me they can smell it. "Nostalgic" they called it. But it also means it's certainly not as efficient as it could be and that's one of the advantages I should be seeing from my EFI. My mileage should be better.

I occasionally get backfires. They usually happen after I take my foot off the gas after accelerating. Since it's a rear engine car, I've got no idea if it's from the intake or exhaust, but I'm assuming exhaust? I haven't seen anything on the inside of the air intake to indicate it's coming back up through the stacks.

It could idle better. And sometimes when starting out lightly from a stop it seems to struggle.

Here are the tables and graphs for my current tune.







The spark advance is a combination of the distributer and the spark table

Great works

[jbscheff]

I am running a wide band O2 sensor from 14PointSeven. The engine is a 1981 and there’s been no mods to the distributor, other than disconnecting the vacuum lines, so no vacuum advance. I had found these curves on another post. The rpm is dizzy shaft, so x2 for engine rpm.

[mb911]

Julian is right. JP helped me and I think Julian as well to come up with a solid afr target. Mine will pop from time to time depends on how warm the engine is and can also be a symptom of a bad exhaust gasket as well.

[Joesmallwood]

Curious, and maybe I missed it, what are you using for load - TPS, MAP, or?

[brighton911]

So Jason, what A/F or lamba are you showing at cruise ? Is EGO correction enabled in MS ?

[rwest]

Originally Posted by Joesmallwood

Curious, and maybe I missed it, what are you using for load - TPS, MAP, or?

His graphs show TPS

[rwest]

Jason,

I had bad results using autone, but it seems like just about everyone else praises it, so not sure what I was doing wrong.

Here’s what I did: first off make sure you turn off EGO correction otherwise the computer will be trying to make temporary corrections while you are trying to set constant corrections; do a “save as” to your current tune and do this often so you can return to one if your tuning goes south- label them for easy recall; find some empty roads and start data logging trying to adjust your engine speeds and loads to cover the most VE cells; go over your logs and then do slight adjustments to the cells that are rich or lean- I only go one or two digits up or down at a time. Rinse and repeat.

Once I was more comfortable with my process, I would drive and when I could safely (I have country roads fairly close by) take quick looks at the AFR gauge on the laptop and pull over several times in a session and make minor corrections as needed and then see how It ran- of course if you make changes during your drive, don’t make changes from the data log afterwards as you may have already fixed it.

Best of luck, it is a long process.

I would also suggest checking with Al; since he sold you the system not to mention a great guy, he should be able to look at your logs and give you advice.

Rutager

[Joesmallwood]

Originally Posted by rwest

His graphs show TPS

Yeah, I saw that, but I vaguely recall there being some inconsistencies with how the load graphs display what was actually being used. ITB mode, TPS, MAP, TPS w/MAP blend, etc, so I wanted to be sure.

What I was after here is the scaling. With MAP you want to bias your rows in the 80~95 percent region, meaning use more rows in this area due to the resolution. Everything happens with ITBs in this small region. An example would be to use rows of 10~15 from lowest MAP up to around 80 and then use rows of 3~5 from 80 to 95.

The opposite is true for TPS. You want the scaling to bias to low TPS values, e.g. use increments of 2~3 at low TPS values up to 15~20% TPS and then 5~10 up to about 40~45%TPS. Beyond 45~40% TPS it would mostly be considered WOT so the rows can be spaced out quite a bit. You can see this transition in the posted VE table where the slope is higher. Honestly, it's not as dramatic as some ITB tables that I have seen so that was another reason why I was questioning the load method. At lower TPS the table jumps more than it does at higher TPS values.

The point is that scaling is important for drivability.

Originally Posted by rwest

Jason,

I had bad results using autone, but it seems like just about everyone else praises it, so not sure what I was doing wrong.

Here’s what I did: first off make sure you turn off EGO correction otherwise the computer will be trying to make temporary corrections while you are trying to set constant corrections; do a “save as” to your current tune and do this often so you can return to one if your tuning goes south- label them for easy recall; find some empty roads and start data logging trying to adjust your engine speeds and loads to cover the most VE cells; go over your logs and then do slight adjustments to the cells that are rich or lean- I only go one or two digits up or down at a time. Rinse and repeat.

Once I was more comfortable with my process, I would drive and when I could safely (I have country roads fairly close by) take quick looks at the AFR gauge on the laptop and pull over several times in a session and make minor corrections as needed and then see how It ran- of course if you make changes during your drive, don’t make changes from the data log afterwards as you may have already fixed it.

Best of luck, it is a long process.

I would also suggest checking with Al; since he sold you the system not to mention a great guy, he should be able to look at your logs and give you advice.

Rutager

Autotune worked wonderful for me when I was using Megasquirt. A couple of things: try to autotune on consistent temperature days and set the IAT curve at zero on the day that you autotune. Otherwise, you'll be chasing autotune every time the outside/IAT temperature swings a bit. Once you Autotune, use the IAT curve to adjust the impact of IAT/outside air.

Secondly, Jamie mentions a technique for road tuning using autotune. Essentially, use your brake pedal to emulate a dyno. Hold your RPM at a consistent value, say 2000 RPM, and slowly press the pedal from low TPS all the way to 100% TPS to catch all the cells in that column. You have to be careful not to overheat your brakes so give them time to cool off after each column, but this method works really well to cover every cell. And yes, you have to do this to get the higher RPM ranges, as well. 6k, 6.5k, and 7k can be nerve-racking, but it's the only real way to get them all covered with enough hits in each cell to make it accurate.

[Showdown]

To echo what Joe said:

I went as far as enabling a second table for my ITBs. Table 1 went from 30% load to 80% and table 2 from 80-100. This was all using speed density.

This way I could get granular resolution above 80% this kind of negating the wot effect ITBs have…

As far as road tuning, I can cover the whole fuel map 3-4 times in about an hour: each run should be short and sweet- no dragging on the brakes.

Do it enough and the changes start to be minimal.

But, BUT, you MUST have an AFR table that’s spot-on and I’ve found that to be an unnecessary extra step particularly when each engine will have its own idiosyncracies. Simply adjusting fuel based on actual AFR readings yields better results for me.


Sent from my iPhone using Tapatalk

[dannichols1474]

Hi Jason,

You asked for Seattle area dyno ECU calibration technician recommendations. I don't have that, but the calibration tech I used here in SF, CA bay area has a lot of connections, so you could try connecting with him to see if he can help you or can point you to someone in your area. His name is Gary Bains (Bain Tuning in Fremont, CA), email: bainstuned@gmail.com, phone: 707-514-0364, url: bainstuning.com or Bains Tuning on Facebook.

Gary tuned and tested my ITB EFI setup before and after I installed Al's option #2 kit, then again 4 more times after I rebuilt my 2.7 into a 2.8 and did engine development with 2 different cams and 3 different sets of exhaust headers.

[jbscheff]

I'm pretty DIY-capable with my car (I put it together from the bare tub, including the wiring), but when it comes to major work on the engine or transmission, I'd rather go with a Pro. That being said, when the weather clears up, I will try to get some data to confirm the cruise region on the table. Then I'll make some tweaks to the AFR in those cells and then let auto-tune make suggestions for changes to the fuel load. If I can lean out the cruise region without inducing any other bad behavior that might be a win.

Dan, I will also reach out to Gary to see if he's got some suggestions for experts in my area.