Bambam7

Just wondering if anyone has attemptd to run their AFM after the turbo, and what the results would be.
Not looking for any performance boost (Though it would get rid of the TID, increasing flow!), I'm just really tight on space there, and relocating the AFM (it's the smaller, S5 one) to the top intercooler pipe coming from the FMIC(it's cooler) would make the bay quite cleaner.


Any theories on how this would affect running?
A few things to think of:
- Temp sensor would read higher (even though there is one in the upper intake mani.. WHY???)
- Not *too* concerned with heat, the air charge isn't that much warmer after the FMIC.....
-Would the air metering change??? Since mounting it in front of the turbo is metering uncompressed air (volume, not mass), after the turbo it will be reading less air flow since it's already compressed won't it?
-I'm sure it's sealed... it could take 10-12 psi???

Any opinions would be great!

rotary>piston

I don't think it would work under boost very well, but I'm not sure.

Bigretardhead

many japanese tuners run their AFM's in the intercooler pipin, and I htink a few people on this forum. it should be no problem. I plan to do this with my S4 AFM when I get a new IC.

projekt

it doesn't offer any performance. if anything it'll just slow the air down some...

Bambam7

Yes... but the air itself will actually be traveling slower post turbo than pre turbo. The same MASS of air is going by the meter, but since it's compressed to 10 psi, the flow volume is less than if that same amount of air was metered before the turbo, and before compression.
That's the only thing I'm worried about, since it's not a MAF, but just a regular AFM. It'll be reading less air flow than before the turbo. But I suppose that can be tuned out by way of the S-AFC...

Wankel7

Isnt there a temp meter inside the afm? At least on the 89-91 cars.

JAmes

Evil Aviator

It would work, but you would need a piggyback fuel computer to make the engine run correctly. For all of the costs and trouble involved, it would be better just to get a low-end EMS like a Microtech that will run your car ultimately better. Really, the only advantage of relocating the AFM after the intercooler is that it will measure the air density more accurately, but the slow stock ECU would still run the engine, and would not have any idea of how to compensate for the AFM relocation, hence the requirement for a fuel computer.

Back in the olden days when the stock ECU was considered fast and several expensive whiz boxes were needed to tune the engine, this RX-7 performed quite well at Bonneville with a relocated AFM. The stat sheet shows the mods. I laugh every time that I look at what was required to perform the functions of a modern EMS. It's almost like one of those computers from the 1960's that took up a whole room and still had less capability than the calculators that you can now get at K-mart for $5, lol.
http://www.geocities.com/evilaviator/bonneville

If you REALLY want to increase flow and clean up the engine bay, then get a standalone EMS.

Bambam7

In that spec sheet, it stated "AFM relocated to inlet side ... AFM sping tension adjusted accordingly."
Hmm.
I do have the S-AFC.. compensation can be achieved.....
Maybe I'll give it a shot! I don't like my TID setup anyways. Looks like Home Depot.

RexRyder

iTrader: (10)

could u get some pics when ur done?

NZConvertible

The AFM's flap or cone measures air volume flow (L/s, cfm), not mass flow (kg/s, lb/s). The ECU must know the mass flow to work out how much fuel to inject. Since volume varies with temperature, if you measure the air temp at the same point that you measure it's volume flow, you can calculate mass flow. That's what the ECU does; it takes the two readings and calculates air mass flow from them.
The AFM's temp sensor will normally read the temp of the air entering the induction system (near ambient if you have a working cold air intake). The temp sensor at the intake manifold (NA) or TB elbow (turbo) measures the actual temp of the air entering the engine. In an NA this won't be too different from the first measured temp, but things like manifold heat soak could make quite a difference. Obviously a turbo requires this extra sensor because the turbo and intercooler dramatically change the air temp. This is the temp the ECU uses for functions like ignition retard and hot-start assist. Because were're talking about a dynamic system (not a static one) the pressure makes very little difference. The main difference will be the temp. The hotter air would cause a different volume flow reading becuase it's less dense, but it will also affect the temp sensor. Because the mass flow through the system is the same before or after the turbo (it's impossible to be otherwise unless you have a leak), then the end result will be the same. The ECU will calculate the same mass flow whether the AFM's before or after the turbo. It's not sealed, it's just like tube. Air goes in one end and comes out the other.

I'm glad you understand this isn't a power-improving mod. A lot of people here think that because they've seen it done on 400hp Japanese FC's, there must be huge gains from it. The truth is it gets difficult to mount the AFM before the turbo once big turbos on custom manifolds and FMIC's are involved, as you noted. It also eliminates the problem of overfuelling casued by open-vented BOV's.

TonyTurboII

For all the trouble, you might as well get a stand alont and get rid of it completely.

Henrik

NZconvertible, great explanation and although you're right that the mass flow is the same you're forgetting that the AFM reading will be dependant on both mass and flow rate. Under boost, the flow rate changes (lower than at atm) for the same mass and the AFM will read correspondingly lean. Bambam7's got it right.

Henrik
87TII

NZConvertible

I personally think it will have very little effect on the mass flow rate the ECU eventually calculates, but even if it does it won't be much and will be easily compensated for by the S-AFC.

Bambam7

Hardly any trouble at all.. especially considering what I've done with my N/A anyways converting it to turbo! (not TII!!!)

My upper IC pipe is right out in the open, it's almost too easy.

Bambam7

NZ- We often seem to have differing opinions! Oh well, I like debating with someone who actually knows what they are talking about, and has an in-depth knowledge of not only rx-7's, but the priciples and engineering behind many of the functions... There are too many guys on here which voice opinons, (which is good) but usually just from heresay, don't know what they are talking about... and usually end up perpetuating incorrect theories.

I find it interesting that you would emphasize density DECREASE from the heat as more of a factor than the increase from compression....
I'm no engineer, but with compression at 10 PSI, I believe that equates to close to double (well, 168%) atmospheric pressure, (0 psi of boost actually being 14.7psi -atm.press- of course) and therefore the inverse of that as volume decrease...
I think that a volume decrease of almost 40% is much more significant than the volume increase from heat.
I'll admit I don't know exactly what delta volume would result from any given increase of temperature, but I don't think that a 50 degree F temp increase would give that much of a volume increase....

Anyways... I knew that the AFM will read leaner, I can easily compensate for that, but I'm just wondering if there is any kind of ratio or curve that will change, which no amount of tuning can account for. Doubt it, but just want to be sure.

NZConvertible

Damn, where's a psychrometric chart when you need one!
Maybe you're right. Since volume changes with the square root of pressure change, you might be looking at an air density increase of ~30% at 10psi boost. I'm not 100% sure whether the denser air will have more of an "impact" on the flap's movement or not.
If this does cause the AFM to read different, you may actually have trouble compentating accurately with the S-AFC. This variation only occurs as boost increases, but the S-AFC can't compensate for manifold pressure, only rpm and throttle position.
I think this'll definitely be a case of "suck it and see"!

j9fd3s

iTrader: (3)

well couldnt you hook the safc up to the boost sensor?

mike

NZConvertible

The S-AFC has no boost sensor input.

TonyTurboII

I gotcha. Buy some cheap pipe and try it out. Sounds like a better way than the TID.

Henrik

Here's the physics behind my reasoning. The force exerted by a stream of particles on a vane is:
F = A*rho*v^2 where A is cross-sectional area, rho is density and v is the velocity of the stream.

We all agree that its a closed system, ie the mass flow rate is constant regardless of pressure. To make the math easy, lets say we run 1bar boost and ignore the effects of heating right now. rho doubles and v halves, A remains the same (constant mass flow through the same dia meter remember) but since F is proportional to v^2, F is down by half, so the flapper does not open as much and the AFM reads lean.

Of course, temp is going to go up so rho won't exactly double and v won't exactly halve so the actual amount its running lean will be hard to compute. The other issue will be the temp sensor in the AFM, the ECU uses this to compute rho and its pretty much maxed by 90deg C (if you're driving in 90deg C ambient you've got other problems to worry about!) - You'de probably need to place this pre-turbo as well.

I've never tried this, but on my paper it looks correct and I would certainly verify before attempting the mod.

Henrik
87TII

[YMMIJ]

i've seen it done on many turbo cars, never a seven though.
I think it's done here mainly for the pigeon "coo-coo-coo" noise, or flutter as most of us call it.

Aaron Cake

iTrader: (1)

Use the throttle input.