Author Topic: Still at it...MAF conversion V5.0 (previously V3.0) (Read 12896 times)

sumyungguy · « **Reply #15 on:** September 10, 2007, 03:17:28 PM »

I am extremely interested in this.. I have the parts lying around for a low boost turbo set up and I would love to get rid of that AFM. You need a dyno? Come out to Livermore and I can get you a hook up on dyno time.

Im also wondering what MAF your are using? I can get toyota MAF's, IAT sensors and an intake pipe with the flanges to but them on fairly easy.

bmwman91 · « **Reply #16 on:** September 10, 2007, 05:34:18 PM »

You have a hookup at Edge Motorworks by chance? That's where I went last time. I would really like to save some $$$ since I just want to do a couple runs to compare the AFM and MAF. Should take all of 5 minutes.

I WOULD like to use a different MAF honestly. This one is pricey and not terribly easy to source (but I got it free, so whee). Now that I have redone all of my computations and built a very robust conversion "toolbox", it would not be much work to use a different one. My #1 concern is ease of fitment to the car. I will NOT be having tooling made for injection molded parts to fit one. That is way beyond my financial abilities based upon the market for this. The one I am using now happens to fit very well, and hence why I stuck with it.

I am taking friday off to dyno the car and go out of town for the weekend. What is your schedule like? I would be interested in looking at the Toyota MAF's. As far as IAT's, I have sourced one that is nearly identical to the M42's stocker between -20C and +120C...that should about cover it. Below 20C (which is not unheard of), there is no problem either thanks to the conversion algorithm, but then again most people do not go out joyriding in that weather!

Another tangent:
Honestly, I think I might plan on making this a "track only" OEM replacement deal. I would LIKE to be able to sell the converter for about the cost of an AFM rebuild. The pricey parts are:
1) the MAF (for now anyway)
2) hardware to make it fit the airbox (which you have to modify a little)
3) the harness (if you make me build you one...plans will be publicly available).

Ideally, I would like to do this somewhat Megasquirt style, but I am not planning on releasing my program code or sensor data at this point. I would like to sell a programmed controller and some PCB's...you just buy some components from Digikey and solder it up. This saves me time, and you a lot of $. As far as airbox fitment, it would cost more than anyone wants to pay to have me make them adapters right now. Now, if someone can find a source of 2.750" (or 70mm) OD mild steel, thin wall tubing, that might change. So far though, it does not seem to exist. I was actually surprised that there is no exhaust tubing of that size anywhere. Still though, you CAN make it fit the airbox for about $10 with some 2.75" OD acrylic tubing from TAP, epoxy, or hot glue. Paint it up, and you are good to go. It sounds bad, but honestly, it works just fine, and with some effort it looks clean too. I just machined a steel adapter plate and epoxied the tube into that so I could screw it on and painted it up. Now you see why I want some 2.75" steel tubing...the MIG welder is my preferred adhesive!

silverblades181 · « **Reply #17 on:** September 10, 2007, 06:09:01 PM »

Well you're not catching me at a good time financially but I am very interested. Here in Quebec we do see below -20C, but it is kinda rare and you're right, I'm not out much on those days. Keep us posted. Do you have any pictures of the fitment in the engine bay?

Will this work with my MarkD 93 chip? (I would think so since you're converting the data...)

DZick · « **Reply #18 on:** September 10, 2007, 08:04:17 PM »

Quote from: bmwman91;33473

Now, if someone can find a source of 2.750" (or 70mm) OD mild steel, thin wall tubing, that might change.

My Father owns a Saw Sharpening company and a ton of his customers are in the steel business. Ill ask him to check this out for you. we have at least 10+ places around here i deliver to that cuts tubing ill ask when im out.

sumyungguy · « **Reply #19 on:** September 10, 2007, 09:59:40 PM »

I have a hook up at Modacar/Forcedfed. I can get Lotus Elise MAF's and this arm...

Shitty pic..

bmwman91 · « **Reply #20 on:** September 11, 2007, 09:29:30 AM »

How much can you get the MAFs for consistently?

The tough part of fitting the MAF is getting it to work with a square hole. The current one actually fits into the intake boot perfectly, but when used with the airbox's square hole & dinky velocity stack, it has some issues around 3000RPM. No matter what, a longer, round velocity stack is needed. Other sensors might behave a little differently, who knows.

sumyungguy · « **Reply #21 on:** September 11, 2007, 02:43:25 PM »

I see.. You still want to keep the air box for CARB reasons? I know there are AEM intakes that have the MAF flange on them and they would be CARB legal, all of their intakes are. AEM short ram intakes are under $100 most of the time too.

Let me get back to you about the MAF's.

bmwman91 · « **Reply #22 on:** September 12, 2007, 09:36:13 AM »

Nah, I am not to worried about the SMOG thing. Though, if I was selling these and people hacked up their only airbox, AND got that one guy who actually cared at inspection time they would be upset I suppose. So yeah, I am sticking with the OEM airbox partially due to considerations about SMOG. Also, I am using it because there is NOTHING else that will work as well as or better than the stock one on an internally stock M42. It is a hard choice though...the AEM kit probably costs a bit more than any retrofitting hardware would for the stock box.

bmwman91 · « **Reply #23 on:** September 19, 2007, 10:59:05 PM »

I have made some good changes to the mapping based upon my dyno time and it seems to run a little stronger at WOT. The problem is that this causes bad mileage...the M42's intake system behaves differently at WOT than it does at smaller throttle openings. This means that the MAF converter would have to have 2 different maps: one for WOT performance with little consideration for economy, and one for daily driving with my pre-dyno map to boost the fuel economy and help minimize environmental damage.

So, there are a couple options I will mention, as well as the one I am going with.
a) Install on/off switch on the throttle linkage to let the converter know when WOT conditions exist.
b) Pull a signal straight from the TPS and use that for the WOT/normal breakpoint.

Pro's/Con's of each
a+) Simple, on and off
a+) No electrical tolerances to deal with
a-) Hard to physically fit
a-) Potential to interfere with linkage
a-) Not many tuning options

b+) No extra parts needed externally
b+) Potential for numerous tuning options
b-) Tolerances in TPS modules are large
b-) Requires user-interface to create 0% - 100% throttle set points

I will be using option B. In the end it will add the least additional cost to the converter overall (maybe ~$3 and some hole-drilling in the case). i will use some LED lights as indicators to walk the user through calibration. Basically, press the button to tell it you want to do TPS set-points. Press it at 0% throttle, then press it again with the pedal floored. Simple, eh?

From here I am planning to have the maps swap over at 80% for the moment. Above this amount, the resonance at the sensor begins to interfere with the readings and lean things out. Nobody is REALLY worried about maximizing power at anything below 100% anyway, and it is not like the car is a slug without the corrections. Ask sumyungguy for his impression of the MAF conversion before any changes.

So there is the current update. I am prepping the assembly code for its new microcontroller home (need more IO pins). I am damn sick of soldering stuff though...oh well, my M42 owns me.

ose30 · « **Reply #24 on:** September 20, 2007, 01:54:42 AM »

I suppose your choice is right, i would also prefer option B.

Quote

I will be using option B. In the end it will add the least additional cost to the converter overall (maybe ~$3 and some hole-drilling in the case). i will use some LED lights as indicators to walk the user through calibration. Basically, press the button to tell it you want to do TPS set-points. Press it at 0% throttle, then press it again with the pedal floored. Simple, eh?

Some factory made piggypacks used similar way to set up TPS, for example AFM Link has it.

bmwman91 · « **Reply #25 on:** September 20, 2007, 01:43:25 PM »

Aaaaah the AFM Link. If those were still on the market i would have bought one a long time ago. I heard that they worked very very well.

The other reason I need a throttle-switch is because of the Motronic's limited adaptability. If it is running rich in normal driving conditions (closed-loop), it will adjust the long-term fuel trim, and thus lean out my WOT operation (open-loop) a bit anyway. i want this thing running perfect in daily driving so the WOT instances remain untouched.

OKTAY · « **Reply #26 on:** November 03, 2007, 09:29:11 AM »

Any updates? What happened to your plans of production?

bmwman91 · « **Reply #27 on:** November 19, 2007, 01:27:42 PM »

A random update...

I thought I had nailed a solution. Oops. Still at the drawing board after some testing, but with a little more direction. More to come.

bmwman91 · « **Reply #28 on:** November 28, 2007, 01:25:45 PM »

Update again:

I made a new datalogger that captures data 25% faster on 50% more channels. In reality, it is not all that bad-ass compared to the first one though (first one was 800 samples/sec x 4 signals = 3200 samples/sec. Current one does 1000 samples/sec x 6 channels = 6000 samples/sec).

Anywho, I am now armed with new data for idle, street, highway, and WOT conditions. The issues involved in getting the MAF conversion working "properly" at WOT are as apparent as ever. Solving them without adding components will be tough, but I really do not have anything better to do in my spare time.

Sheesh, if only we had 6 cylinders this resonance would hardly be an issue! Stupid lack of intake valve overlap!

nickmpower · « **Reply #29 on:** December 13, 2007, 02:22:34 AM »

anything new?