Not sure if this should go in the tuning section or not, my apologies if missed posted. I pretty much only hang out in here.

I was looking through some of my dyno run datalog stuff recently and I had been thinking about trying a bigger MAF, for a couple of reasons, but I that is a whole other can of worms.

I know I am making less power than others have made with the MAF but I just want to know if it is currently a restriction. Can I tell this from the data?

RPM16 data and MAF flow:
(Summary: MAF flow pretty much flatlines ~4900 RPM. Peak power is ~5200 RPM. Never hit 255gm/sec.



RPM gm/Sec
1416 59
1384 58.7
1560 53.2
1546 64
1675 69.1
1669 69.7
1797 75
1752 77.3
1841 80
1946 84.1
2069 86
2132 88.1
2074 91.3
2100 93.4
2180 95.9
2409 97.1
2392 100.6
2421 103.7
2421 106.3
2715 109.1
2614 112.5
2587 114.8
2738 118.2
2934 121.9
2785 126.4
2934 130.6
3171 133.7
3202 137.9
3043 141.8
3101 146.4
3171 149.4
3223 152.8
3536 157
3498 159.9
3549 164.3
3498 168
3885 171.5
3654 175.8
3766 179.3
3752 181.7
3885 185.1
3885 189.1
3964 192.8
4130 195.6
4201 199.8
4274 203.5
4274 206.6
4274 210.8
4428 215.2
4408 218.8
4659 221.9
4659 225.1
4964 230.8
4818 235.8
4726 237.3
4940 240.5
5342 244
4940 246
4990 247
5229 246.6
5285 247
5313 247
5342 247
5342 247
5522 247
5401 247
5553 247
5461 247.9
5816 247
5649 247
6030 247
5682 247
6068 247
5957 247
6144 249.6
6425 247
5886 247
6068 247.9
6221 247.9
6030 250
6030 249.2
6221 251.9
6221 252.5
6342 250.2
6182 251.7
6383 253.8
6467 253


This is the best way (I think?) to tell, but I'm not sure. I know there are ways to tune the MAF that makes this insignificant. What I want to know is, what does this tell me about how much of a restriction I have upstream of the intake. (I don't have a MAP sensor.)

 

You need to hook up a vacuum line and see if it stays at zero throughout the rpm range or if you start drawing vacuum. Would be best if you could tap into the tract between the MAF and TB.

That said, I'm guessing probably no. What power level are you at?

 

About 390rwhp.

There is a dyno day coming up, maybe I can get a vac gauge on it then.

So this really doesn't tell me much then?

 

I was about the same hp as you and went to a 3.5" maf, retuned it on the street and didnt pick up any time at the track. Only matched my previous best. On a dyno, i may have been able to pull another few ponies but i'm not sure how much. I dont think it was really worth the effort other than the fact that a 3.5" CAI kit looks cool in the engine bay over the stock TPI snorkel.

Even tho it didnt hit 255, i'm pretty sure your maxing that MAF out. My car maxed it out around 4500 rpm I think and your close there, about 4900 rpm it hits 245-246 and flatlines at 247.

 

No, you're just hitting the flow limit of the device itself. All the tests I've seen done on a MAF they will flow a lot more air than the device can read. 255 gm/s translates to around 450 cfm or something like that and even with screens and fins the stock MAF flows well over 500 cfm. At 390RWHP you may want to check though, a 3" ID intake (about what the MAF is as well) might be too small and you may pick up a little power with something bigger.

How was your A/F ratio?

 

Orr, that is interesting. No change whatsoever? I've actually heard similar things, from Doc maybe? But yeah, right now I'd almost do it for cosmetics as it looks a little funny shrinking down to that tiny little black tube.

Madmax, yeah, this is descreened too. Supposed to be, what 638cfm? Or something around there? The dyno's wideband was not working the last time I ran.... so I don't know unfortunately exactly what the AFR was.

I do think the MAF is hitting the read limit and I know the tune can be altered to account for that. I'm pretty sure I can make more power through the stock MAF too. I'm just thinking if it looks better and it looses a bit of that restriction it might just make a bit more too. Plus there will be plenty of room left to grow for bigger CID/FI etc.

My only concern is: Going to a larger tube you loose resolution with the MAF right? Is there a negative drawback to that? Poor tune off idle/low airflow situations etc?

I'll try to do back to back dyno testing if I do end up building a larger MAF.

 

Yes, doc also went to a larger tube with the stock electronics. There are a few people who have swapped out to a different maf entirely. IMHO its all been a lot of work for little to no gain. As someone once said to me too much maze for too little cheese.

There are basically a fixed number of data points with the stock code. Right now, you have the 255 gm/s spread over the 6 MAF tables. Lets just simplify it and call that 42.5 gm/s per table. If you were to change the 255 gm/s to 2550 gm/s over those same tables, you now have 425 gm/s per table. If there is only one data point per table (there is not, but anyway) then you just allowed the MAF reading to change by as much as 425 gm/s before anything changes vs 42.5 gm/s. That is the 'lost' resolution people talk about. How much the amount the airflow across the MAF needs to change before the fueling is changed is the issue.

What, if any, has been done to the stock tune?

 

With the new large MAF, i had to scale the MAF tables up to get air fuel back to where it was just after my dyno tune with stock MAF descreened. Once it looked pretty close, the car ran the same as before in similar conditions. Was there more in it? Maybe. I tuned to the same air fuel ratio just weeks after running in similar weather and it showed same times/mph. I was hoping for 10-15whp but I dont think it would have showed up. Never did get it redyno'd. Never did mess with n/a tune again because I went with nitrous and sprayed the car to the 10's the rest of the year. I was gonna redyno it but cracked the flexplate so I wasnt gonna stress it.

It was surprising to see the stock MAF make that much power and show no gains with the larger tube. If any gains, i think it would have been minimal.

WIth the large tube i scaled the MAF up so much that I believe 255 max was now in the upper limit of table 3 or in table 4. I no longer have that bin file. It was lost on an old laptop that died on me. Although the resolution I think was lost there, i had no driveability problems

 

Youd do much better with an LT1/LS1 MAF. A transducer can be built for less than $30, and the $6E ECM can be made to register more than 255 grams of airflow by manipulating the injector constant. I ran an LS1 MAF with a transducer and a modified version of the $6E code, and it actually worked fairly well for what it was. Not super accurate, but it worked ok.

Again, the restriction is within the MAFs circuitry, and not the actual size of the MAF.

 

Thanks all.

Madmax, it is a custom dyno tune by a good tuner. However, for 400$ I'm now doing it on my own with half that going to moates for the right equipment. I already have datalogging cables etc. "Too much maze for too little cheese..." I like that... and to be honest that is exactly why I want to do it. The devil is in the details and I like the challenge and seeing the results. As for the loss of resolution...

Orr, good to know the loss of resolution wasn't a problem. As I see it, I value <255gm/sec resolution more than >255gm/sec resolution. Below 255 is where driveability comes into play etc and you can tune beyond the 255 with PE to get you around the 255 limit, as you well know. The way I see it, similar to you, is the >255gm/sec isn't really what I am after, that is merely a byproduct of changing the tubing from <3" to ~4". The main goal of this excercise being to reduce the restrictions preventing this motor from making the most it can.

Dimented, thanks. I'll look into the LS1s too. If your statement, "the restriction is within the MAFs circuitry, and not the actual size of the MAF" is completely true, this is all for not. Again, I'm ok with tuning (with a wideband on a dyno) beyond the 255gm/sec wall, however what I want to change and don't like is throwing a MAF that flows 50% of what my TB flows and based on theoretical flow numbers could be reducing the horsepower this engine would make, say if it was piped with 4" tubing.

Perhaps the easier thing to do would be go MAP... but I love the flexibility of MAF tuning!

Great input everyone, thanks for the comments thus far!

 

A 3.5" MAF can register up to 700 CFM of air. I think the late model 'vette ones flow even more. I dont think that it would be much of a restriction. Not as much has having the MAF flatline and having to fudge the tune to get teh AFRs your looking for.

 

MAF definitely is the best way to go on NA engines, and is much easier to tune. I love how its a set it and forget it deal.

Based on your data, you should be within the flowrange of pretty much all the V8 LS1 MAFs. You might also want to look into an aftermarket transducer that allows you to adjust the response. This will allow you to recover some of the lost resolution that you will incur with an 8-bit system.

Its possible to make the ECM legitimately register more than 255 grams per second. With coding, I was able to get the original $6E fueling code have just one MAF table that registered up to 512 grams/sec. It was still nowhere near as accurate as a late model PCM that can read the frequency directly, but it was an improvement on the bastardized MAF tables that the $6E has. Even still, you can make it so you at least have some control over the fueling with an engine that flows >255 grams/sec. That, and the Delphi MAFs are lightyears ahead of the junk Bosch ones.

 

IF you follow any of the LSx world, you can see they have throttle bodies up to 100+ mm. For a while there they didnt have MAF's that large and most guys just deleted the MAF and ran speed density. Now they make 100mm MAF sensors to go with the FAST 90, 92 and LS7R 100+ mm intakes. I believe there was a gain in power with the larger MAF over the 75-85mm ones. I'd have to look again to see how much was gained but I do not think it was a significant gain. These were on some very hot combinations that made big power to justify the need for a 100mm MAF. I'm thinking this would be nice to have in a 500whp + setup that most 400+ inch LSx motors make

 

I never played around with larger maf's or larger tubes with a maf inside, but I had the wells maf before I swapped to SD. Car ran 12.0x's and same mph with the maf and same times after switching to SD. The only power that I picked up if any was moving to a 4inch intake tube.

 

4" intake tube... yep. Word is ~2 feet of 4" tube will give you a bump in midrange torque. That is the goal, maximize where things are and minimize un-necessary losses.

Orr, I agree. As I see it, this car is a money pit so I'll probably end up doing a 42_ cid SHP build next, so I'd hope to be in that range. Then again if I went LS1 I could avoid all of this... haha

Dimented, I like that idea. I'll have to play around with the tables some. Do any of the LSx MAFs use a blade/chip style sensor or are they still the wire sensor like our stock ones?

 

AFAIK, GM currently uses probe and tube type MAFs using the hot wire type of sensor. One or two wire(s) heated, with one wire at the ambient temp. The output is the current needed to maintain a constant temperature differential between the heated wire and sensing element.

Edit: Actually, it looks like some of the newer motors like the LS7 may use the blade type setup. Those are also frequency based, and would allow you to make your own MAF tube. Delphi.com is a good place to look for more info.

 

If you don't want to blaze your own trail, an elegant solution to this problem already exists in the form of an analog slot-style sensor, housing and matching tune from Blowerworks.

http://blowerworks.net/

Be aware that I'm posting this on behalf of Blowerworks.

This is a simpler and more reliable approach than what has already been discussed, since there is no need for a translator device unlike a frequency based MAF conversion.

The ecm's bin file can be tuned to accomodate any reasonable housing diameter if a larger than standard housing is desired.

 

What would be the flow capacity on a setup like that? Would a 3.5 or 4" tube MAF flow enough for a 1000hp turbo motor?

 

In a 4" tube, the sensor itself could measure up to approximately 1300 gm/sec within a 5 volt signal output range. This airflow quantity should support up to 1100 hp or so.

It might be a bit restrictive if used as a draw-through sensor, but should be fine for a blow-through set-up for these power levels.

 

Max for that MAF is 500 g/sec. About the limit of many of the larger GM MAFs. Its spendy, but it would likely work. Also a possibility is an analog ford MAF. To adapt it would require the methods I discussed above, less the transducer. The advantage of a transducer is the ability to alter the response, which can help preserve resolution if applied correctly.

 

Yeah just curious as it would be a blow through setup. Just wondering if the ecm could actually handle fueling when trying to read a 1300 gm/sec signal in just 0-5 volts. Not sure how this is accomplished in the bin file, i guess you use the 6 tables and just have a different resolution scale? I would be worried if the sensor and ecm are fast enough to accurately make adjustments at those large scales.

Would more than likely make tuning abit quicker and easier for a boosted motor, but in then end I dont have a problem with MAP, it just takes a better base tune when starting from scratch and that can be somewhat difficult and time consuming.

 

A scaling of 5 gm/sec per bit would just about cover the expected 1300 gm/sec range. 1275 gm/sec ought to be close enough.

The concern would be at the bottom of the range where 30 gm/sec would be represented by only about 1.29 volts.

The normal driveability range would only have about 8 voltage points to work with requiring quite a bit of interpolation between values. It could be made to work, but may be a little rough.

Edit: subsequent work on a 4" calibration has permitted up to 17 tuning points between 0 and 30 gm/sec. This will provide plenty of adjustability for fine tuning the normal driving/idle/cruise airflow range, even with a high flow capable sensor.

Of course this represents an extreme case, but it is still worthy of discussion.

Reconsidering the resolution aspects: Even at only 1.28 volts output at the aforementioned 30 gm/sec range for cruise there are still 64 a/d values in play, so the actual available signal resolution is approx. 0.5 gm/sec per bit in this airflow region.

With the recent calibration changes, these 64 possible A/D values are then multiplied and will be interpolated between the 17 points in the MAF table and filtered to result in the airflow signal in gm/sec.

In reality, even at this extreme there should be plenty of working resolution and adjustability for smooth running provided a relatively smooth input signal.

Getting 90+ lb injectors to idle well at 0.4 ms or below may be an issue with the 165 ecm.

 

Especially with larger injectors the pulse width difference could really swing air fuel ratios so depending on what injectors you use, might be a chance for driveability issues?

Not to get off topic with it, but it sounds like the large tube MAF that blowerworks makes with modifed bin could help with a stout n/a motor

 

Thanks tequilaboy! I've been discussing this issue with you over on CF in PM's actually.

So, I should be fine in my current state as a draw through if I went to a ~4" tube? Or am I going to have some issues until it gets blown or ++ CID?

 

As for the drivability thing, I can say that it may be an issue at idle. My first MAF setup used the modified $6E code running in a TBI ECM (with TBI). I had my current vortec MAF hooked thru a home-made converter box. At low engine speeds and idle, there where only a few hundred mV of ouput. Bear in mind that there are only 256 steps from 0-5V. At idle, a couple hundered millivolts would equate to an A/D output of 15-30 steps. The net result was that the fueling moved in a step like fasion. If you watched the AFRs on a wideband, they would jump from one value to the next rather than smothly vary around the commanded AFR. It does result in a chunkier idle, and could result in some issues if you have a cam.