cjernigan

iTrader: (13)

Running megasquirt in parallel like you would on your '99 anyway is emissions legal. It doesn't throw CEL codes to my knowledge and you retain control of everything except fuel and ignition(MS controls them) with the stock ECU controling idle, A/C, Fans all of that crap.

Joe Perez

iTrader: (8)

The AFM measures air by volume (ie: cfm), not mass. But fuel and air must be metered into the engine in a ratio determined by mass (weight) rather than volume. In order to make a fuel calculation on any engine, it's necessary to know the total mass of air being ingested, that is, the total number of air molecules.

Visualize in your mind a bottle with a removable cap. The bottle has a capacity of exactly one cubic foot. And imagine that we are on St. George Island, Alaska on a calm day in late January, standing on the shore of the Bearing Sea. So rather conveniently we are at exactly Standard Temperature and Pressure- 0°C at sea level.

Since there is very little to do on St. George Island, by this point in the conversation the two of us, along with Loki and Braineack, have consumed all of the beer out of the bottle, emptying it completely. So we place the cap back onto the bottle, thus sealing exactly one cubic foot of air inside. Under these conditions, the quantity of air inside the bottle is about 1.26 moles, weighing approximately 0.08 pounds. (one mole is 6.02 × 10^23 molecules of whatever substance we happen to be measuring)

Hypothetically speaking, if we wanted to burn a quantity of evenly atomized gasoline inside this bottle so as to achieve perfectly stoichiometric combustion, we would need to inject 0.00544 pounds of gasoline into the bottle- roughly 3.4cc.


Did I mention that the cap on this bottle has a small schrader valve attached to it? I'm not sure why the brewery put it there since they bottle-condition all of their beer rather than force-carbonating it, but it's very convenient for our purposes. So you attach an air compressor to the valve and raise the pressure inside the bottle to 14.7 PSI (relative).

We have now doubled the quantity of air inside the bottle. It contains 2.52 moles of air molecules, weighing 0.16 pounds, and yet its volume has not increased- it is still one cubic foot of air. Now, if we wanted to do the gasoline-burning experiment, we would need to double the amount of gasoline squirted into the bottle- 6.8cc this time, to achieve stoichiometric combustion. We would also need an ambulance, and most likely a defense attorney as well. The point however is that we knew we had to inject more fuel, because we knew that the total quantity of air inside the bottle had doubled as a result of the increased pressure.


There is a small barometric pressure sensor inside the 1.6 ECU (a MAP sensor, essentially) and the ECU uses this measurement along with the volume reading from the AFM to determine the actual quantity of air entering the engine. But it assumes that the pressure of the air flowing through the AFM is the same as the pressure that it's measuring down in the passenger footwell. A safe assumption in a naturally-aspirated engine, and on a turbocharged one as well, so long as the AFM is located on the atmospheric side of the compressor.

So let's say that we have recovered from our injuries, been found not guilty, and are driving our car around. The AFM determines that one cubic foot of air per second is passing through it. The ECU references its barometric sensor, determines that the ambient atmospheric pressure is 14.7 PSI, and thus knows that it must inject 3.4cc of fuel per second (in total among all cylinders) to achieve the correct AFR.

But if we are under boost, then that cubic foot of air is at a higher pressure than what the barometric sensor is reading. The fuel computation will be wrong because there is more air going into the engine and yet the flapper inside AFM has not moved. This is because the pressure on both sides of the flapper is equal. The ECU has no way of knowing that you've compressed the air going through the meter, and so it will not inject sufficient fuel. The resulting 30:1 AFR promptly causes the pistons to melt, forcing us to abandon the car on St. George Island, since we have no way of driving it back onto the ferry. When we get back home, we start a new thread on MT.net, regaling others with our tale of woe, and Philip promptly bans us all for extreme stupidity.

cjernigan

iTrader: (13)

Jesus Joe, you should write a turbo miata for dummies book with examples like that. Atleast it would be enjoyable for people to read.

mikeflys1

iTrader: (2)

lol i guess i did ask for that

imperial562

so the basic idea is that, Place the BOV as close to the throttle body as possible. right. all BOVs are carb legal, true?

Loki047

iTrader: (6)

Joe, remember what happens at St. George Island, Alaska stays at St. George Island, Alaska. good explanation.

magnamx-5

iTrader: (7)

Damn good explanation Joe. Damn didn't kurt run a blow through maf :gay: no wonder his **** didn't do so hot.

Joe Perez

iTrader: (8)

No. In fact I don't think that any BOV is CARB-legal, at least not in the same way that, say, an exhaust header is.

You can, for instance, purchase an exhaust header from Jackson Racing (or whoever they are now) and it will be CARB approved. The header itself, as a standalone item, has been certified by CARB. This is possible for two reasons: First, because an exhaust header can only be installed one way- short of cutting and welding there is no room for error on the part of the installer. And second, this header is specific to one particular vehicle and engine configuration- it only fits on Mazda Miatas with 1.6 liter engines, sold from 1990 to 1993.

It's therefore very easy to certify that one particular configuration, without worrying that someone will try to install the header on a Mazda 323 or a Ford Probe- it just won't fit. Some headers, primarily for American muscle cars, may fit a wide range of vehicles. The CARB paperwork for these will state precisely which combinations of make, model, year and engine trim the header is approved for. A header which is CARB approved for Ford Mustangs with 5 liter engines built between 1986 and 1993 might physically be able to fit onto a 1994 Mustang, but this would be an illegal configuration.

A BOV is a bit different. It's a generic part which will fit just about anything, and can be installed in just about any configuration. Right off the bat that makes it damn near impossible to certify.

In order to get a BOV to be certified, it needs to be part of a kit.

For example, all of the turbo kits that BEGi sells include a particular bypass valve. It looks like a Bosch unit, but I'm not positive. Some of these turbo kits are CARB-certified. So installing the particular BOV that comes with a BEGI kit, along with the rest of the kit, on the specific Miata configuration that it is approved for, is legal. The certification for the kit includes the BOV. In fact, technically the kit would be illegal without the BOV, since it was present during the certification.

Another example would be if Flyin' Miata chose to pursue CARB certification for its 05-29121 Mazdaspeed Miata upgrade kit. (They'd probably have to delete the MBC, but ignore that for now.) In that case, the 05-90120 GFB recirculating BOV would be legal to install on Mazdaspeed Miatas if and only if the rest of the parts in the upgrade kit are also installed.

In either scenario however, the BOV is not blanket-legal. If you were to install only the GFB BOV, even on a Mazdaspeed, it would be illegal because the the rest of the kit is not there. And if you were to install the BEGi BOV onto a Miata that does not also have the whole rest of the BEGI turbo kit on it, it would be illegal.

The California ARB publishes a wonderful database which lists every single CARB EO# in current use, and is searchable. Once you find the number you're looking for, you can download a PDF containing the complete text of the EO. Here is the database: http://www.arb.ca.gov/msprog/aftermk...es/amquery.php

For example, I selected "Turbocharger system / modification" and "Sort by Manufacturer", then located Greddy. That gave me D-397 and a few others. Clicking D-397 gives me a .PDF file containing the complete certification for their '90-'93 Miata kit. http://arb.ca.gov/msprog/aftermkt/devices/eo/D-397.pdf

Notice that it specifically states the conditions which must be satisfied in order to be compliant. You must install all of the parts on the list (Mitsu turbo, manifold, downpipe, FPR, filter, etc), you must tune the car according to the specs in Greddy's documentation, and the car must not have ABS. If you were to install the system and then add a device which is not on the list (intercooler, BOV, etc) or install it on a car with ABS, then you would technically be in violation of the order. Now, it would be technically possible for someone like Stripes to put together a package consisting of an intercooler, some pipes, and a BOV and submit this package for testing. In that case, the ARB might issue it a certification declaring it to be legal when used on a '90-'93 Miata without ABS, and in conjunction with the Greddy turbo system specified in D-397. This seems unlikely to occur however as ARB certification is costly and time-consuming, serving as a de-facto barrier to trade for those businesses or individuals operating in very low-volume or low-margin areas.

Joe Perez

iTrader: (8)

A blow-through MAF sensor might, in theory, work.

Heated-wire MAF sensors do in fact respond to air mass directly. The operating principle is that the air passing through the meter cools the sensing wire, which in construction is similar to a lightbulb filiment and across which a constant voltage is applied, causing it to become hot. As the wire cools its resistance drops and the current flow through it increases, which in turn causes the wire to become hotter, increasing its resistance and restricting the current flow once again. Thank you, Dr. Augustus Matthiessen. This current flow is measured and converted to a voltage which is the output of the sensor.

What is convenient is that the more air you pass over the sensor, the more the wire is cooled. This is true not just for increasing the volume of air (higher CFM) for a given density (PSI) but also for increasing the density of the air for a given volume. Heated-wire MAFs do not rely upon external barometric compensation to work. By knowing how much current is passing through the wire, the ECU knows roughly how much air is passing across it, trying to cool it down.


Let's say that you happen to be in a spacecraft orbiting somewhere in the lower thermosphere, which the part of the atmosphere just before you get into the exosphere and then "outer space." At this level there is still some detectable air, albeit very little. You go for a space walk, taking with you a large fan, a multimeter, a battery, and a MAF sensor. You immediately die because a space suit was not on that list.

So Loki, being somewhat more level-headed about such things and remembering to don his spacesuit first, goes outside and collects the objects that you dropped when your torso exploded. After cleaning off all of your expelled bodily fluids he sets up a test whereby he is able to pump 1,000,000 CFM of air through the MAF sensor while observing its output. Despite the fact that a tremendous volume of air is being passed through the sensor, each cubic foot of that air contains virtually no actual matter (it's mostly just empty space) and so the reading from the MAF sensor is very small. In fact, it's exactly the same as if a small mouse were breathing past the sensor down on terra firma- roughly the same mass of air is passing through it.


The trick however is that MAF sensors do rely upon external temperature compensation. For a given molar mass of air, the amount of thermal energy that it removes from the wire as it passes by varies with the temperature of the air. All else being equal, the rate of conductive heat transfer between two objects is a function of the ΔT (the difference in temperature) between them. Blowing very cold air across the sensor will cool the wire by a greater amount than blowing the same mass of very hot air across it.

The system is designed to compensate for this. There is a temperature sensor which determines how hot the air flowing through the intake tract is, and a correction factor is applied to the MAF reading based upon this. The hotter the air gets, the more positive bias is given to the MAF's reading. So the question all boils down to how quickly the temperature sensor can respond. Under normal circumstances it is only subjected to very gradual changes in temperature- the change in ambient temperature as spring gives way to summer, the under-hood temperature rising as the engine warms up, day yielding to night at the Bonneville salt flats, and so on. The air coming out of a turbocharger however varies quite rapidly in temperature. As you step on the pedal and the compressor starts making boost, the air temperature will rapidly shoot up. Then as you release the pedal and the compressor idles, the temperature will rapidly drop. Putting the MAF sensor after the intercooler would mitigate the severity of this effect somewhat, but it would not be entirely eliminated.


The other thing to watch out for is that MAF sensors require laminar airflow across them to produce an accurate reading. The air coming out of the compressor is likely to be highly turbulent, and will thus screw with the sensor somewhat.


In "The Muppet Movie" (a classic if ever there was one) Gonzo tells his new traveling companions "I'm going to Bombay, India to become a movie star" to which Fozzie replies "You don't go to Bombay to become a movie star, you go where we're going- Hollywood!" Gonzo, somewhat put off by this reasoning, responds "Sure, if you want to do it the easy way."

I guess what it all boils down to is why would you want to place the sensor downstream of the turbo? It offers no real advantages and presents signifigant challenges.

Loki047

iTrader: (6)

This has got to be the best reading on an internet forum ever.