Cannot recommend FiTech F.I.conversion.

After being hounded to swap from carb to EFI for years and having the old Q-Jet go bad I gave in and did as bid.
Seems to have been a mistake, after several days of fighting issues the FiTech is coming back off and a carb going back on.

The FiTech 30003 is getting sent back in for testing, if they claim there are no issues with it then I will know for certain that it is just expensive scrap.

If they do find/admit to an issue and make repair I will give it another shot.
Seems fairly nice when it works, but only when it works!

Keeps dying and losing the tune at idle, then I have to reload the base tune just to get it started again.
I've done all sorts of "Tricks" to protect the signal wire and ensure good power but nothing makes any difference.

Now it is being increasingly difficult to get the hand held to communicate with the ECM, something is clearly failed or failing.

Carbs Rule!

Engineering...

Sometimes the simplest solution is the best solution.

Yeah, FI Tech isn't highly recommended here.  A bit of research would have done you good.  Factory OBD2 is good stuff.  Before that?  You might as well install an opensource-ish product like MegaSquirt or similar.

Open Source style supported products are slightly better as you can generally find the documentation and users who are willing to lend assistance.  Sometimes you have to filter the wheat from the chaf, but generally the theory is the same no matter what you're trying to do.  Tell the computer to provide X amount of fuel at this point on the table.  Now that point on the table could be generated by RPM and Throttle input or RPM and Vacuum signal, it depends on your particular setup, but generally that's how they are supposed to work.  The ECU will then use your setting as a base and modify it slightly based on O2 levels in the exhaust, coolant and air intake temps, etc.

On the plus side, you can find people who've created solutions to their issues that could provide useful to others.  Like those poor bastards with rotary engines or those that use low impedence injectors or want to use MegaSquirt to control their electronic automatic transmissions, etc.

MegaSquirt has quite a bit of OpenSource style support and places like DIYAutoTune.com to provide support for products they've sold.

Holley seems to at least have the customer service piece due to how large they are and the fact that they sell and support a very LARGE range of products.  Their low end/early TBI models aren't well received though, so you have to be closer to their upper end products to get decent results it seems.

Tech support has been pretty good, they got me past the worst early hurdles.
Tech even determined that the rubber seal for the MAP sensor was mangled and sent me replacements.
I've done the shielded signal (Tach) wire and carefully ensuring I have full 12v power at all times.
But no matter what else I do the unit keeps losing the base tune and killing the engine at idle.

There is a pretty active FB group that are mostly into the high end versions but have provided some clues.

Tune may be easily modded, but I cannot do that while driving by myself.
And of course the tune is lost anyway when the unit glitches.
Now the hand held seems to be losing its ability to communicate with the ECM in real time.
I've printed out the RMA sheet to send it in to be tested.
Only another $85.00 and postage. 

i helped install a FiTech kit on a c3 corvette not too long after it was released, and it worked very well.  Absolutely no issues.  Easy to set up and damn that throttle response was just IN SANE.  Made that pig of a 350 seem decent.  I was properly impressed.  Motor was stock other then long tubes and stock-type duals I did, along with an square-bore Edelbrock intake.

We didnt have to do no tomfollery or nothing.  That was 5 years ago and the owner has never had an issue.

However it seems the later FiTechs and Holley Snipers have been having some occasional issues beyond install errors.  I am on pages for both and Id say out of 10 issues...maybe 2 or 3 are actual unit issues.

My neighbor who has GM muscle cars bought a FiTech system a few years ago (I only know this because it was delivered to my house by mistake...).  I'll ask him if they ever got it running... 

There's a thing here, and I don't want to appear arrogant about it. Engineering and testing and QA is hard. It takes a lot of intelligent people to continually produce competitive and consistent products.

I spoke to the people at DUI about my rotor which didn't fit. Apparently the guy I talked to was -the only- staff. It was a short conversation.

I interviewed at Richmond Gear, and there were essentially four people on staff. Four, for all gears, cars, transmissions, boat transmissions, shifters, etc. Just something like the heat treatment design and testing on different gears in different locations is way harder (pun!) than it would appear, you can't just wing it and expect it to work. We can manage in our garages, with very small sample size and the expectation that some things would have to be redone, but when you ship tens of thousands of the things to go into whatever the customer has, it's a lot harder.

That's just not enough eyes on something intended to be spread far and wide. Something like fuel injection conversion on a common V8 has to cover an incredibly wide variety of conditions and engine combinations. I do not see a way that you can do this without either lots of experience or lots of resources.

I guess I am saying that I am not surprised.

I really love the concept...

I'm wondering if the issue with FiTech is that it's not enough carb, but not enough EFI...

Guys who are good with carbs are generally befuddled by anything electronic. Guys who are good at EFI would find the FiTech much too rudimentary and limiting (and they would not like the lack of control with the self-tuning mode)

That makes me wonder if it's a good product for a guy like me... it understand the basics of carburetors (but certainly can't tune one), and I get the theory of EFI (but am totally lost when it comes time to dial them in).
 

Then I get scared off by internet comments (from guys on both sides of the aisle). 
 

Unsure...

If you can tune a carb, you can certainly do rudimentary tuning with EFI.

EFI gives you a lot more thinks to tweak (or twerk).

I was helping a guy on a C10 forum sort out his FiTech, and I worked with his tune and the software, but the issue with his was something nobody could find.  It would just reset itself in operation, and lose sense of all reality.  It would leave him stranded, or run like crap or run like a top.  Very tough to find.  No - grounds were all good; clean, new, proper and confirmed.  He swapped to a Holley EFI unitwhen one went on sale, and never looked back.

In reply to SVreX (Forum Supporter) :

I can handle both and I've had maddening experiences with both.

Had a Holley on a '71 Corvette that made me pull my hair out trying to get it jetted right, done lots of other carburetors with no problems.

Had a Chevrolet RamJet 350 in a '34 Ford that we simply could not get tuned. Even the guru at the dealership couldn't get it to work right because GM in their infinite wisdom used a Mercruiser ECU that wouldn't talk to anything but Mercruiser software properly. The fix was ditching the Mercruiser bits and plugging in a Holley computer and harness. This seemed to be a common thing with the RamJet 350 from the googling I was doing.

Sometimes (in the GM crate motor case) there simply wasn't enough testing done before something was dumped on the public.

Everything that Tuna said times 1000. 

I have worked with OEM as part of their quality chain. It blows my mind how much goes into design verification and production quality. NO WAY any small- shop start up is going to come close. 

I went around the block with the Fitech for a long time. I really wanted it to work if for  no other reason than I had 2k unvested and I am a stubborn bastard who does not want to quit or spend the $$$ for a new Holley.

The Holley is working fine thank you.

Things I learned:

1-EFI installation is not for the non educated in EFI technology. The "self learning" mentioned in the marketing is more about the buyer than the TBI.

2-I will never install another intake system of any kind without doing a smoke test of the intake and the exhaust. I need to KNOW that those two are good. 

3-In a perfect world where the drivetrain and the product are in working condition, the thing is a piece of cake to install. It should run just fine with the initial wizard that ferrets out your volumetric efficiency and engine volume.  For about 75% of buyers that seems to be the case, so there are a lot of happy customers out there who will gladly call you a moron if you don't get lucky like they did. Few have a clue why it works, it just does, If this meets Fitech's business model ( it very well could) then the more power to them. I like to see small companies succeed.

4-If new to EFI, you are going to go on the forums and try to button-tune your way out of the issues. You wont have a clue about data logs and how to make any sense of them. But you will try to learn and you will spend the next 6 months trying to be systematic about it. But the logs that fitech gives you are just columns of excel data with no analysis function so good luck.

5-The TBI programing will try to compensate for issues with the rest of the system, Be it vacuum or exhaust or timing, the Fitech will try to adjust to meed whatever targets you have set. This can send you down a lot of rabbit holes especially if you are Forum button-tuning.

6-I should take the fitech off the shelf and check the MAP sensor grommet. I never did get around to that. There is an internal ground that is on the same level of rumored QC concerns.

7-Exhaust leaks: Sorry, but headers are famous for leaks. And we are all aware of the variability of homemade exhaust systems based on what we do here. Bolting on a fuel system that requires 100% exhaust system integrity is a risk. 

8-Chances are you are not going to take the car to a qualified tuner with a dyno who can look at the data and verify that it is running optimally. Without doing that, chances of getting the full potential out of the TBI are slim to none. 

In the end, I believe that the root cause of all my fitech issues was a faulty HEI style distributor, a bad AOD gearbox, dual plane manifold and a small exhaust leak. Pretty sure that if I go back to the Fitech  now, it will be fine. ( subject to checking the MAP grommet). But I wont, because when it is all said and done, I can think of few failure modes with a carb that will require getting towed home.  With the Fitech I am aware of a multitude of failure modes that will leave me stranded. 

I could probably getting a pretty decent tune on this if it would just stop being erratic and losing the tune every time it has to idle for more than a minute.
Off to a parts store now for fuel line to put the Demon back on until FiTech gets back to me with their results.

Just in case anyone stumbles upon this thread in future - FiTech datalogs can be read easily in (free) MegaLogViewer.

In reply to SkinnyG (Forum Supporter) :

Could you expand on that? Recall that Fitech and Sniper are marketed to non-EFI trained buyers. What is "Mega Log" and what can it do for the user? How is it used?

Pete

MegaLogViewer is a free datalog viewer used alot with the Megasquirt crowd.

I loaded a FiTech datalog into it (*.csv), and it showed me a time graph with all the inputs I can select available on a graph.

This is the FiTech datalog I was sent:

I was also able to download and install FiTech EFI Pro Cal to check out the fellow's tune and make suggestions & changes, but the issues he was having were very likely all hardware.  He's running Holley now, and it's never driven better.  Dude is not a tuner guy at all.

ShawnG said:

Sometimes (in the GM crate motor case) there simply wasn't enough testing done before something was dumped on the public.

I can tell you that the "factory" tunes in some of the LS variant crate motors are...a little rough. Definitely not as sorted as the actual factory ones.

I was just digging through the GM CAN documentation and came across this. Made me laugh because good lord.

Vehicle Mass Nominal is received by Platform. This signal is used to translate between vehicle acceleration and axle torque values. Refer to GMW8769, PPEI Cruise Control Subsystem for more details on the conversion between acceleration and axle torque.

In reply to RichardSIA :

FiTech systems are a popular system among GM muscle car owners (I know this only because I accidentally received one).  He should be able to tell you if it ever runs... 

One of the projects I have to try to get done this weekend is yanking the FiTech off and putting a Carb back on.

If I still have similar issues then I will suspect the HEI distributor.
But with the way the FiTech runs well one hour, or barely at all another, I really doubt it is a distributor issue.

I have the form to send it back to be looked over and am as certain as I can be that at least one sensor is bad. Not sure if it is air intake temp, barometric pressure, or ??? but something is out of whack.

Local tune shops refuse to touch FiTech, "Too many problems".
I'm guessing their tunes do not "Hold" and customers get PO'ed.

I do hope to be able to refit the F.I. once it's been repaired.
Looking for the self calibration for altitude when I cross the Sierra's.
If FiTech techs still insist there is nothing wrong with it then it's expensive trash and I will get rid of it.

I chose Holley over FITech because of their tech support, and did a Holley Sniper swap onto my '89 Chevy stepside with a cammed 350. When I was reading comments and reviews, many of the negative reviews were people fighting with a carburetor and thinking the EFI would solve everything. If you can't get your carb to run right, don't even start on an EFI system.

1) No vacuum leaks!

2) No exhaust leaks! I had to torque my headers a little too tight, and I still get a little noise out of the passenger side, but the system abides.

3) I broke the connector that goes through the firewall to connect the handheld to the throttle body. I emailed Holley's tech support asking how to get a new plug, but they never answered. I was able to fix it with a dinky pair of needle-nosed pliers. For reference, the display on a Holley Sniper uses a 4-terminal CAN connector, specifically a Mizu-P25, P/N 52117-0411 with terminals P/N 50039-8100. The terminals are 22AWG.

4) The cheap throttle position sensor Holley uses is junk. Mine crapped out after a couple weeks and I swapped it for a 32 year old original sensor from a factory TBI - it's the same part as the round-plug TPS. No problems in the year since.

5) After reading last month's article about dialing in your ride, the Holley Sniper will never be 'dialed-in' where I can toss the keys to someone without preface. The system always takes a couple seconds to boot up. If you skip that crucial step, it takes extended cranking to get it started. If you let it boot, then start, it busts off instantly. Also, the throttle tip-in a super sensitive. My truck always tries to blow the rear tires off, but there's a $14 throttle extension that's supposed to help. Personally, I think it should come with the kit.

The self-learning does work, but probably better on the most common engines. On the first drive mine bucked and snorted and had all the usual problems, but a year later with regular driving it's sorted itself out nicely. This last week I had to make some longer across-town drives and my gas mileage jumped into double digits. That said, when I ditch the locked small-cap HEI and buy a matching Holley distributor I'll take it to a local tuner.

All that said, the ECU 'forgetting the tune' sounds bad.

2) No exhaust leaks! I had to torque my headers a little too tight, and I still get a little noise out of the passenger side, but the system abides.

I'm confused-why would an exhaust leak cause an issue with tuning an EFI system?

In reply to CrustyRedXpress :

Probably inaccurate O2 sensor readings

AngryCorvair (Forum Supporter) said:

In reply to CrustyRedXpress :

Probably inaccurate O2 sensor readings

Quoted for truth.  This is slathered all over the FiTech instructions.  It invalidates the sensor readings the ECU uses to self-tune.  
I firmly believe that FiTech could have done a better job with EMI/RFI shielding of their ECU especially on Ford products with the distributor up front.  I had a lot of issues that I now believe are resolved but it was painful for quite a while.  

In reply to AngryCorvair (Forum Supporter) :

But if the exhaust gas is under positive pressure and escaping to the atmosphere, how would that change the 02 readings?

Is there some part of the combustion cycle where air could get sucked into the exhaust system before the 02 sensor?