It seems that most MAF users accept that $6E is the way to go, so I am using that as my reference.

In my other thread, Grim Reaper said something to the effect of "Currently there isn't too much on the MAF system but it is growing...". That made me wonder why? Do the experts in DIY PROM mostly use SD or something else? I do sometimes run into questions that aren't answered by running searches. I could ask a million questions about the MAF system but I try not to annoy you guys too much. Anyway, just curious as to the reason of a lack of MAF info. Is the switch to SD, or something else, really worth it once you start modifying your engine?

I am curious is all. Thanks.

 

The reason there is more info on SD is because, when this board was first created; it was mostly SD owners who were interested in eprom burning...because the SD cars need an eprom change for virtually EVERY change, especially if you want it to perform optimally.

With MAF cars, unless you are HEAVILY modified, the MAF system does compensate to a fair extent. So there is not the same need to have your eprom changed as the SD system. But, if you want "optimal performance", then even MAF owners can benefit from modifying the eprom.

So, when the DIY Prom Board was in its "infancy", the majority of the guys were SD cars. The few MAF guys were the ones with heavily modified cars that were maxing their MAF (255 gm/sec limitation). And the MAF guys were trying to figure out the best way to tune around this limitation.

Today, the benefits of tuning is well known and even "mildly modified cars" are looking to modify their eproms...some just to increase their gas mileage. As most of the 3rd Gens were MAF cars (the SD cars are some of the lowest sales volumes for 3rd Gens), now there is more interest in tuning the MAF system.

 

Great reply. Thanks bud!

 

The was also another side to this, the '86 - '89 MAF systems are un-reliable. The stock MAF sensors frequently go out and are expensive to replace. The '165 ECM is known to be flaky. The first test for either is the tap-test. Tap on the MAF or the ECM, if the engine stumbles it is bad.

Now someone that is doing tuning and has a MAF go bad, many times they swapped to a SD system. Not only is it a cheaper solution then purchasing a MAF sensor, a DIY-er also has the skill to change it over.

RBob.

 

I've owned MAF cars since 1988 and have never replaced a bad MAF, somewhere around a half a million miles on family vehicles alone, plus friends vehicles who never had replaced mafs either. Not to mention I sold parts for 7 years and sold more MAP sensors than MAF sensors. I dont know what you're basing your statement on, but I have an entirely different experience. If you can back that up with some evidence...

 

I have had to replace my MAF only once... Is it possible that the new aftermarket stuff is a bit more reliable?

Edit: I should mention that when my MAF "failed" it was because my finger slipped and once of the wires snapped off inside.

 

I am probably going to regret replying to this subject am I do not want to start the old "MAP vs MAF" debate. So let's please keep this post "friendly and factual" as this topic almost always leads to hostility.

Madmax, the one reason why you probably sold more MAP than MAF sensors is because GM use far more MAP systems in ALL their model lines than MAF sensors in the 80s & early 90s. Specifically for 3rd Gen TPI cars, there were more MAF TPI cars sold than MAP TPI cars, But for GM's entire product line during the 1980s & early 1990s, GM sold more MAP cars than MAF cars. Thus, it makes entire sense that you should sell more MAP sensors than MAF sensors. And, if you add the number of guys converting to MAP from MAF...

As to RBob reply; we must admit that in the "early years" at DIY Prom we were quick to get people to convert to a MAP based system (whether their MAF was bad or not). Part of it, was because most of the guys at DIY Prom who where tuning their cars were SD guys than MAF guys. We knew more about SD than MAF - thus we were better able to help them tune their car if it was SD.

A MAF guy was out-of-luck, not because we didn't want to help them - we just didn't enough expertise amoung ourselfs and our members to help them...much like the guy with the "computer-carb" system still has limited help at DIY Prom

I was one of the few people at that time who was trying to help tune the MAF system. That is when I came up with the "tune via P/E" solution for the "maxed MAF" problem. Today it is common knowledge and it seem like we've been doing it forever. But in the early days, the lack of "MAF Tuning" was only because there were so few MAF guys who were doing any tuning at DIY Prom. Today, now that we have more MAF members who are tuning, we are able to give better advice for those who want to tune their MAF car (and not want to convert to MAP).

I find this is a "parallel" to the guys who wants to build their 305s and the only advice they get is "get a 350". I now feel that once the guy has decided to build his 305 (and has chosen NOT to get a 350), we should just tell him how to build a strong 305 and not debate the 305 vs 350 issue any further.

But, I do agree with RBob that if your part of a person's MAF system fails and becomes toast, converting to a MAP system may be a cost effective solution and should be looked at - especialy if the guy is already knowledgeable in tuning.

Over the years, my attitude has change (or it's the heavy medication) to "help the guy tune or build" whatever they got; whether it is MAF, a 305, a V6 or even Ironduke. But only AFTER they have reviewed ALL their options and are making an "informed" decision. Even if that "informed decision" is not very logiclal. It is their car afterall.

 

It is difficult to compare failure rates based on repair parts sold to customers. The cost of the replacement MAF will preclude the common parts swap. Many of the replaced MAP sensors may very well of been good. It is just that they are cheap to replace.

GM moved to SD on the 3rd gen f-bodys just because of the MAF failure rate. Note that they went back to it with the LT1 with a newer & different MAF sensor.

RBob.

 

In the PROM tuning guide at the top of the forum, there is a section with links to MAF vs SD debates. There are plenty.. I have MAF and had to replace the sensor, it jsut randomly failed and I looked at switching to SD. I was and still am only a beginner, and I got the feeling from reading that MAF would be an easier place to start. A MAF reading is easier for me to understand than a MAP pressure reading. Im not yet picky enough to care which system is more advanced.


And I also think that in the 3rd gen MAF world, $6E is where you wanna be. $6E is newer and replaced $32 and $32b, therefore is better?? Ill assume it is. And youll find more topics about $6E and help, which that reason in itself is enough to atleast start with it.

 

I don't know if I would say $6E is "better" than the earlier MAF systems. Definitely "better defined & better understood". However, I know a few guys who "just had a couple of minor tweeks", and they preferred the driveability of their original system with the "minor tweaks" over the $6E.

Conversely, they were not willing to "work" with the $6E and just wanted a "plug & play" solution. Had they been willing to work with the $6E, maybe they would have changed their story.

When Bruce (Grumpy) was alive and we had a long chat on the phone, he confided that he felt that on a "new system" (ie a hot-rodder who wanted to drop a crate motor in and install EFI), that MAF would be the quickest to get "up & running" with good overall running.

As well, in "steady state" running such as driving down the freeway or constant WOT such as drag racing, that MAF was the easiest to obtain best results. But, in road-racing and overall driveability (where a person was willing to spend time tuning) MAP gave the best results.

I think that you need to determine what your usage will be and how much time you are willing to spend on tuning. For simplicity and ease of install, go MAF. For wide variation of use and if you are willing to spend some time, go SD.

 

Even the $6E is pretty primitive. GM did alot of things that didnt really make sense, such as the use of the scalars and multiple tables rather then use one large table and a better table lookup routine. I modified bits of the $6E code for use on my MAF TBI project, and I didn't really like it. The resolution was too low at idle. Also, the code is inflexible, and can only be used on V8's since the code calculates a duty cycle rather then the pulsewith needed for each cylinder. The other issue is the 255 g/sec limit. Even a super-mild 350 can exceed that at high RPMs. If I had to choose, Id probably go with the SD. The end result of both systems is the same, but the SD code has more room to grow built into it.

When the MAF is set up properly, however, it can work quite well. I currently have an LS1 MAF and PCM, and thats so accurate that the actual AFR in open loop is almost exactly what the desired AFR is. Its only off by a few percent at most. The other nice thing is that itll read up to 512 g/sec with high resolution. The only real difference between MAF and speed density is that the MAF doesnt require tuning when engine changes are made. So long as the intake remains the same, it calculates the proper fueling. Im lazy and impatient, so it was just what the doctor ordered.

 

I HATE it when someone takes my post and then breaks it apart. So please accept my apology as I do this ONLY for clarity since there are so many great discussion points in your post. I do not do this to pick apart your great post - but to highlight some great points you made.

Yes, it is primitive, especially when you compare it to today's newer PCMs. But, when you consider they were working with 8bit technology on micros invented at a time that Bill Gates was still looking for financiers for his company to develop PC DOS, you also have to give GM credit for a pretty advanced system (at the time).

And the reason they used multiple MAF Scalar Tables was for resolution in the lower rpm range (idle & cruise range) while the higher RPM range was still out of the performance range GMs most powerful engines at that time, and only needed in WOT (which was not under testing requirements for emissions even dyno IM240)


That, unfortunately is the mathematical limitation of the 8 bit processor, which was "high tech" when it was first released. Heck, the first microprocessor for "public use" was less than a decade old at the time.

But, I do wish GM had the foresight to use "double word" 16 bit math for it's calculations with the MAF sensor when they developed it, but again, the L98 was their most powerful engine at the time they developed it (and even that was an accident as GM origially was going to make the 305 their flagship engine. And the later LT-5 that was only released by GM in 1988 or 1989 (it's been so long) and used an different ECM & calibration.


Agreed. The code in the 7730 does have more room within the accessable memory area to add additional code or tables. So, if a person is interested in "bit twiddling" and wants to get inside the code, they will be far happier working with the 7730.


I can attest to this. When my buddy made his first round of modifications to his 1989 Iroc (ZZ4 with ported Aluminum 113 heads, Miniram and a Crane hydraulic roller with 222/228 @ .050 duration IIRC), we spent most of our time tuning the MAF Scalar tables, spark tables and a bit into the P/E (it only maxed the MAF at the highest part of the RPMs). But we did need to do a fair amount of tuning at that time to his $6E.

Later, when he did a rebuild and modified his engine further (383, kept the Miniram but swapped the heads for AFR 195 Competion Port with bigger 2.05 intakes and a solid roller), my initial modification to his bin was strictly for the larger Injectors, a bump up in his idle speed and the P/E - but nothing else.

Amazingly, this ran like a charm. I couldn't believe how those few changes were only necessary. We continued to fine tune his top end (as the motor makes power beyond 7,000 rpm) for P/E with the WB O2 sensor but that was about it. Our only tuning problem (which SD doesn't solve either) is we cannot compensate beyond 6,400 rpm. Luckily, my buddy has decided that he doesn't want to rev his motor beyond 6,800 rpm - so high rpm tuning isn't a limiting factor at this time.

But, you should see motorcycle rider's faces when they race my buddy AND they see him right beside them at 100 mph (instead of their mirrors).

 

Quite true. I've just seen a lot of 'slipping fingers' and mistreatment of the sensor rather than true failures. I replaced one, once, on a car that had been in an accident and had cracked the housing. For a while we had the thing taped together with duct tape and it was fine, just not the best thing. Also if you do a search here, you will find a lot of posts like "I got code 33" or 34, or 36... and followed with "and I replaced the MAF and still get the code". So there's also a lot of misdiagnosis going on. I've found a lot more wiring problems and sometimes bad relays than anything.

4thgen is entirely different, better MAF, better ECM. I might agree that the failure rate of the 85 with MAF module was very high, it was an extremely problematic pile of electronic junk.

 

Also, if you're doing a TPI or similar swap and have access to either it might be the time to go with the SD instead unless you had some reason to run a TPI-era MAF.

There's only one advantage I see to the $32/32b, thats the unused pulse width table. I messed with it for quite some time before going to the 6e, and it solved quite a few quirks I just couldn't get rid of completely with the 32b. Sure, its not perfect, but its not all that bad. I should let someone else chime in, but I did a new combo (heads, cam, intake, exhaust, and the old one was tuned), did 10 minutes of datalogging, made some adjustments after some brute force breakdown and analysis of data, and stole some timing from Traxion (that I slightly adjusted for my combo and code), ran some math for my PE (and checked with Trax's program), graphed it, said yea ok, and went to the dyno. I had the flattest AFR of the day, and right where I wanted it. Only a Superram car made more HP than I did, with SD and through a 5-speed rather than MAF and slushbox. So... 6E is plenty capable or at least I find it to be, and it seems to be easier to get 'right' because there's a pretty wide 'right' range with it. SD is picky in that regard. I have a couple very minor issues right now I'm not happy with that just dont seem to want to go away, that I have a feeling would go away with SD, mostly because of methodology. And I wouldnt have to worry about spiking the MAF limit as soon as I hit the throttle. But, it works. Makes good power, drives good 99.9% of the time, and gets pretty good mileage as well. I have little reason to swap. Maybe if/when my now 14 year old descreened finless MAF dies I'll grab my SD stuff off the shelf and put it in, but until then I think I will leave it alone.

 

Interesting conversation.

I'm an SD guy, mostly because most of my vehicles that I have modified, rather all of them have been SD, honestly just getting into the PROM game.

I like some of the ideas of MAF, like the measurment of actual airflow (as close as a heated wire/screen can), instead of calculation based on load or vacuum.

On the flip side, there isn't much support for it and has been kind of a black art in some circles to get a MAF to run right, probably due to lack of support. There's also this idea that a MAF causes a large restriction in the intake stream. If you use a large MAF for a built combo, or turbo application, the velocity at lower RPM can from what I've read cause improper readings, needing lots of tuning, to get close.

I always found it intersting that back a few years ago, while GM guys were swapping thier MAF systems to SD, Ford guys were doing the opposite, swapping from SD to MAF, go forward a few years and the swapping gets reversed, Ford guys going to SD and GM guys going to MAF, or maybe just more of a diverse application of each system?

For my V6 I think I'll be sticking with SD, since it's supported for my application, but I'll have to keep MAF in mind for other projects.