Very big bore throttle body?

[Alyman]

I've been thinking about ways to squeeze more power out of an M20, and I have a couple of questions I hoping that someone more knowledageable might be able to answer.

Mr Moore often says that a key factor in increasing engine power is increasing an engine's efficiency as an air pump.

My car currently has a six branch and a nice exhaust, but as nothing has been done on the inlet side, there must be improvements to be made, no?

I was looking at an M30 TB the other day, and despite the obvious problems (different flange/pipes/direction of opening) I was wondering if it would be a worthwhile project making it fit onto a 2.5 inlet manifold? Would I have to swap to the M30 airbox too?

Along with changing to slightly larger injectors (2.5 M50 maybe?) and a rolling road remap would this be a good way to increase the amount of air and fuel going into the engine.. or would the larger TB cause problems by slowing the velocity of the air heading into the inlet (something I read on another forum for a different car)?

Or should I just buy an ebay resistor/vortex sprial kit/magnets and be done with it

Ta

[Speedtouch]

Just whacking a huge throttle body or carb on a normally aspirated engine won't necessarily improve the performance, in fact, quite often the opposite, as torque is usually adversely affected due to the lower air speed through a large bore inlet (venturi effect). This tends to give poor low-speed running characteristics, meaning you have all or nothing power delivery, and flat as a fart low-down throttle response!

However, some have had good results from doing this - you may be interested in this article http://www.davelength.net/car/bigsix.html

[Alyman]

I actually read that article while doing my reseach on this.

The guy in the article doesn't change his TB though, which I thought was a bit strange. I've also been reading a few threads where that very article has been discussed, and a few interesting points have been made - here's one of them:

http://forums.bimmerforums.com/forum/sh ... p?t=983918

a brief summary:

- the guy in the article has effectively given himself a bigger flap/door inside his airbox. But seeing as he is using the M20 AFM circuit board, then as far as the ECU is concerned the flap is still the same size as an M20 one. Lets say for arguments sake that it flows 20% more air. This means that when the flap is open X degrees then the ECU thinks that X amount of air is coming through, when in fact it's X+20%. Clearly, this would lead to a lean mixture. Larger injectors, flowing more fuel seems to be the logical answer, and is what this guy did at the same time as the AFM swap.

Some say that this will not solve the underlying problem of the ECU always having the wrong air flow data.

Also, I think I'm right in saying that US cars have a lambda sensor in the exhaust which allows the ECU to have a bash at fixing a lean/rich mixture, but UK cars don't.

The more I read about it, this seems like the type of thing to leave for people who have aftermarket tunable engine management with o2 sensors and the like!

[GeoffBob]

The more I read about it, this seems like the type of thing to leave for people who have aftermarket tunable engine management with o2 sensors and the like!

I'm inclined to agree, if only because an aftermarket ECU enables you to keep tabs of all parmameters that could otherwise be deviating from ideal.

The underlying problem, to which you originally referred, is one of improving what is termed the "volumetric effciency" or VE of an engine. Any modification that allows an engine to either inhale or exhale easier is essentially an improvement to the engines VE. There are many ways to improve the VE of an engine, including replacing standard cast-iron log manifolds with tubular headers, removing the catalytic converter and replacing the exhaust system with free-flow components, changing the cams, porting the head, enlarging the valves, modifying the inlet manifold, and removing obstructions from the induction tract. The modification that you listed above falls into this last catagory and essentially involves (at least in principle) reducing the obstruction presented by the air-flap by fitting one that is larger.

While I can see the advantage of removing the air-flap altogether in favour of a more modern air-flow measurement method, I cannot see any advantage to an increase in the surface area of the flap. The engine will in fact induct more or less the same mass of air as before, but (due to the larger surface area of the flap) will now open the flap less than before. One way to compensate for the now smaller signal returned to the ECU from the flap position sensor (since the flap is open less) is to fit larger injectors.

I'm sorry to say this has to be one of the worst ways I have heard of to improve the VE of an engine, and it will cost you a set of larger injectors to boot (the extra flow capacity of which you will never use, or see any benefit).

Just to be clear, the old flap type AFM's were sized by the manufacturers according to the cubic capacity of the engine. Increasing the size of the flap doesn't cause the engine to induct more air, it just results in the flap opening less which (using the same unaltered ECU) will require larger injectors to compensate for the reduced signal from the flap position sensor.

One possible modification to a flap type AFM is to reduce the tension on the spring that holds the flap closed. This truly does reduce the restriction presented by the flap to the inducted air, BUT it is necessary to remap the ECU or recalibrate the flap position sensor (to stop the engine from running rich). It is also necessary to make sure the flap doesn't open wide before the engine has reached maximum air-flow at wide open throttle.

[e301988325i]

Here's my thoughts, having completed it successfully, I am running a lambda. . .

i) It's not the flap that causes the restriction, it's the maximum Cross Sectional Area (CSA) of the AFM.

ii) The davelength article - his car has a lambda sensor, this is critical to the conversion working well. Uk spec E30 's don't have a sensor, but do have the loom for one. ie, buy a sensor, a lambda heater circuit relay for next to the fuel pump relay, and get a boss welded on. £100ish.

iii) The M30 AFM is 50% taller and therefore 50% greater CSA than the M20 AFM. If you get an M30 AFM from a 1986 3.5L M30, part number ending 027, the resistance runs to 50% more than an M20 AFM. ie, 50% more CSA, 50% more resistance = the same ammount of air, the same resistance between these AFM's.

iv) At full throttle the BMW ECU ignores the AFM output and uses a 'stored' fuelling map. This is where the greater CSA comes into play. However the car is getting more air but no more petrol because the values are just 'stored', the bigger 19lb M30 vs 17lb M20 injectors come into play here. The M30 injectors are of the better 4-hole vs M20 1-hole design and atomise the fuel more finely leading to greater efficiency.

v) AFM's are picked to suit capacity, but the 320i and 325i share the same AFM. . . and injectors.

vi) I found it neccessary to reduce the tension of my M30 AFM by 2 notches of the black plastic gear wheel.

On a seperate note for discussion, the 2.5 M50 has an 80mm throttle body, yet a 2.5 M20 has a 60mm TB. The M50 has an 80% bigger CSA!!!!!!!

[DaveD]

As uk e30s did come with lambda sensors .. at least the late model ones did....i think that all you need is the sensor and a boss and the ecu will do the rest .....

[GeoffBob]

Here's my thoughts, having completed it successfully, I am running a lambda. . .

Fair comment from someone who's tried this mod. Out of interest though, is the diameter of the standard AFM smaller than, larger than, or same as that of the air/induction pipes that run to and from it?

I'm afraid I'm of the opinion that any perceived benefit from this mod is as a result of your point vi) where you reduced the tension on the spring. Unless the AFM was smaller than the pipework to start of with, then the only mechanism that will increase the air flow (as you state occurs) is as a result in the reduced tension on the spring which allows the air flap to swing wider open, thus presenting less of an obstruction.

BTW, not arguing, just discussing/debating.

Regards

[AlpineAde]

I run a near standard engine and have always thought of it as a big old airbox that burns stuff. I base my mods and what I plan to do on that principle.

At the moment I run a BBTB, a ported inlet manifold and Miller MAF conversion. I dynoed before and after. After the MAF conversion, BBTB and ported inlet manifold, things are smoother and with more power from earlier on the results are interesting: between about 4300-5800rpm there are instances where power is greater by as much as 7-8 Kw with a lower figure of 2-3 kw or so. Around 4-5 Kw seems to be the median. This would explain, along with the increase in torque (though not as much as described by Miller) why the car really seems to start to boogie beyond 4000rpm. If anything, running on the standard Miller map is compromised by the BBTB and inlet manifold. Put the car on a rolling road and tune and results would have been better.

As yet, on the exhaust side, I have done nothing aside from a Supersprint cat-back. I do have a 6 branch going on and I'm replacing my cat with a hi-flow one but that is yet to come. Come next week I'm having a top end rebuild done that involves headwork, a cam, bigger injectors and after that my Miller WAR chip will be fitted. The car will then be dynoed again. And again when I organize the rest of my exhaust.

Overall, after the MAF, BBTB and inlet manifold I gained next to nothing in terms of peak power. A few horsepower. I did, however, gain massively as to where the power came in and how long it held. As I said, between 4300-5800rpm the gains are simply great. Peak power is developed and held 1000rpm earlier than when previously tested.

[e301988325i]

Fair comment from someone who's tried this mod. Out of interest though, is the diameter of the standard AFM smaller than, larger than, or same as that of the air/induction pipes that run to and from it?

I'm afraid I'm of the opinion that any perceived benefit from this mod is as a result of your point vi) where you reduced the tension on the spring. Unless the AFM was smaller than the pipework to start of with, then the only mechanism that will increase the air flow (as you state occurs) is as a result in the reduced tension on the spring which allows the air flap to swing wider open, thus presenting less of an obstruction.

BTW, not arguing, just discussing/debating.

Regards

The AFM's have a hole which is a rectanglular in profile and this rectangle fits inside the diameter of the piping, and therefore is of a smaller CSA, whether it's smaller than the TB is another question though. In fact, the M30 AFM rectangle also fits inside the diameter of the piping!!!

I might not have stated before, but the M30 AFM is of exactly the same design as the M20 one, just the flap and corresponding hole are 50% taller.

It's difficult to explain, and took me a while to get my head around, but bearing in mind it made no sense to restrict airflow at the design stage, the flap itself presents little resistance to the air flowing through the AFM, the problem is that of CSA, at high intake volumes.

Moving the plastic wheel was neccessary to set the idle CO2. Once all the adjustment has been used on the AFM screw, the next step is to adjust this wheel, assuming the rest of the AFM is in good order!!

IMO the M20 b25 was restricted by BMW, perhaps because it exceeded their targets, because the brakes are already beyond their capabilities of handling 170bhp, tyre cost/safety in terms of limiting MAX speed (the 80's remeber), limiting insurance costs, not outdoing the M3, making sure the engine would do a million miles. . . I mean why doeesn't a 2.5L BMW hit 60 in 2nd gear, I mean what is 1st for, towing overladen ar-tics up hill?? My opinion is, because it makes the 0-60 longer and reduces insurance costs and also gives the M3 a bigger advantage.

In order of air flow the restrictions are, AFM, TB, cam and exhaust manifold.

Discuss!!!!

[AlpineAde]

Well, I cast my vote.

I threw the AFM in the bin. I just don't like them. Stupid flappy barn-door things.

I also threw the TB and standard inlet manifold in the bin. (simple bolt-ons; easy to do and relatively cheap)

I'm about to throw the cam in the bin.

Then the log manifold goes.

Ideally the cam and manifold would go at the same time.

[e301988325i]

For my manifold to be fitted, my steering linkage had to be removed, the rear part of the manifold had to go on first with the studs removed from the head for the 3 rear most cylinders, and then screwed back into place throught he manifold flange, then the front half went on rather more easily!!! Soooo glad I paid someone else to deal with the stress!!!!!

[e301988325i]

for info the M20 AFM is 38 x 48mm = 1824sq/mm
for info the M30 AFM is 38 x 72mm = 2735sq/mm
M20 60mm TB = 2350sq/mm - spindle shaft is actually 8mm dia.
M20 64mm TB = 2700sq/mm - 15% bigger in area

I hadn't worked that out before, makes interesting reading.

edit - improved accuracy based on actual 8mm TB spindle shaft, and notice how the big bore M20 TB is so close to the M30 AFM.

[AlpineAde]

The BBTB I run is 4mm oversize. I've got pics if interested.

[e301988325i]

2700sq/mm then, + 15%, do show any pics. . . and any of the manifold internals and MAF in place??

[GeoffBob]

[e301988325i]

The M20 b25 AFM is 25% smaller than the M20 b25 60mm TB, itself a well known restriction.

My old renault 19 16v used only a MAP, variable TPS, Lambda and temp sensors, meaning nothing in the inlet tract. That would be even better than MAF, IMO.

My friend has a similar vintage F.O.R.D. with motronic, the ECU is running with an AFM, but the ECU has unused connections for a MAP sensor, the idea being bin the AFM, attach MAP and hopefully off you go. . .



Cheap MAF idea, needs mapping to suit, but no mention of the CSA.

http://forums.bimmerforums.com/forum/sh ... st16291350

[AlpineAde]

Around 2900sq/mm then, + 15%, do show any pics. . . and any of the manifold internals and MAF in place??

I'm on it, mate. I'll load some up now.

[AlpineAde]

Inlet Manifold:

Standard



Detail



You can see where the standard inlet manifold has some interior lazy metal just sitting there, doing nothing and just plain getting in the way. Bah! Away with it!

Ported!



Sweet!

Throttle Body:

Standard



BBTB (all freshly rebuilt & 4mm oversize)

[AlpineAde]

MAF in situ:

[e301988325i]

Very nice, the MAF looks so tidy, OEM even. I'm going to have to bore a spare TB out!!!!

[AlpineAde]

I threw the AFM in the bin. I just don't like them. Stupid flappy barn-door things.

Fully agreed, best way to minimise an obstruction is to throw it out altogether. Problem is, what do you replace it with if you want to keep your motronic? Is there an easy conversion to a hotwire type MAF meter that will connect to the motronic without too much bodging? Although a fully aftermarket ECU is a great idea and offers the freedom to tweak just about anything and everything. Although with an aftermarket ECU I’d probably just fit a MAP sensor and implement a speed-density system.

That's why I went with the Miller unit. Very simple. Comes with a chip designed to see the MAF and is optimized for it.

I have now bought the WAR chip which allows communication with certain parameters of the Motronic system: fuel, ignition timing, etc., etc. Hook your laptop up and away you go! A few guys in the States have had the Miller base maps tuned for their cars in a matter of an hour or two.

If you don't want to go a complete standalone or don't have the savvy for a DIY like MS, then I reckon a solution like this ~ with software that is maturing at a wonderful rate! ~ is definitely worth a look.

[AlpineAde]

Very nice, the MAF looks so tidy, OEM even. I'm going to have to bore a spare TB out!!!!

The MAF is perhaps one the last billet ones they did. I picked it up for a song as part of the Miller clear-out. They are now onto their 3rd generation unit.

What really excites me about the Miller gear is the amount of interaction you can now get. Essentially tune for YOUR car and YOUR mods. No need for a generic chip.

[e301988325i]

I've been interested in the W.A.R. but I'd not heard any feedback about it. It could be really good for me, LPG map, petrol economy map, petrol power map. I've got a few things to do before it, but it's possibly on the list. . .

[AlpineAde]

I'll see if I can find a link that reviews it for you. Hang on....

EDIT: Have a read of the last two pages of this thread:

http://forums.eurocca.net/showthread.ph ... WAR&page=6

Good guy and a nice car, too.

[eta]

I have an M20B27 wih 885 head, matching inlet manifold and motronic 1.3. It runs the stock exhaust manifold. Beacuse it was an easy swap I replaced the TB with BBTB from alpina527. It makes nore noise now and it feels like it develops more mid range torque. I am getting it dynoed again soon so I will actually know what improvement there is. I think it has been worth it.

I also want to do the M30 AFM conversion as I have no intension of ditching motronic 1.3. However I have one question. How do fit the TB to AFM pipe to the M30 AFM, surlely it require a larger diameter one? Perhaps not. With this answer I will be buying an M30 AFM and fitting it with a new chip to get round the fueling issues(fitting an O2 sensor and larger injectors is fiddly). I could also take to circuit motors to get it mapped with unichip, that should get arround all the fueling issues. I also take it the air box will require some modding to get the new AFM to fit.

[march109]

is alpina527 on the e36 forum? I think he did mine.

[eta]

I don't know but his work is good.

What about the AFM to TB intake pipe. does it need to be changed to fit the M30 AFM or does it fit the larger AFM. If it does then upgrade is simple.

[e301988325i]

Yes the standard pipe fits the M30 AFM just, after being warmed in boiling water

[eta]

In May my M20B27 powered manual e28 put down 159.6 hp and 186ft.lb of torque.

spec
Stock eta bottom end (10.2:1 pistons)
885 head
eta throttle body
083 AFM (the wrong one but it all works, idle is good and strong)
stock airbox and filter
Motronic 1.3 (173 ECU with A-tech chip)

So I fit a big TB 325i TB from alpina 527. I just had it dynoed today at Surrey RR and 141.8 hp 175ft.lb! That a big loss. How can this be.

The only other changes are API SE CC spec oil 20W50 in place of the 10W40 oil I used to use (I thought I would give it a try as the manual specifies API SE spec oil) and NGK BP6ES spark plugs as the BP5ES where out of stock.

How could I ahve lost 10 ft.lb of torque across the entire rev range and 18 hp! Could it be the dyno or do I have a hidden problem?

Also I recently puchased a 027 M30 AFM. I was thinking about fitting this but now I have 18 hp to find but how on earth do you get an M30 AFM to mount on the 325i air box. there is no where to secure it. The E28 525e and E30 325i air boxes are the same.

I would not mind doing this build in a lighter E30 one day but how on earth do you get it right.

[AlpineAde]

Same dyno? Same tyre pressures? Corrections for temperature? Everything as similar as can possibly be between runs? This would be the first place to start. You might have ended up going to a low reading dyno. It does happen.

[eta]

I have posted plots in another thread "now I am confused". Same dyno, same operator Alpineade. Temperature was 8 degrees lower today than may and tyre pressures are good. As the run in May where corrected for temperature and pressure I assume today's are as well.

[AlpineAde]

Hmmmm, it is possible that your ecu is starting to get a little confused. It could be the installation of the BBTB itself. Have you double checked the integrity of the unit and of your installation?

It could also be that a totally unrelated problem has sprung up between the runs and is affecting power.

[e301988325i]

My best guess Eta is something that is often overlooked. Stale petrol.

Evo magazine dyno tested nearly every brand an spec of petrol, 95, 97, 99, in a mk5 2L turbo golf and a BMW M5. Their main conclusion was that using the busiest petrol station was the best answer.

Many premium 'optimax' type fuels, don't sell well, sit in the underground tank for ages and go off!!!

The M5 has a built in torque limiter and made only a tiny ammount more power on £5/L race fuel, yes £5/L, than supermarket 95RON!!!

[eta]

I use Tesco 99 only Tesco pumps are always pretty busy. A fuel tank last me less than a week about 430 miles of driving. So no chance of it going stale.

I cannot find anything wrong with the TB install either. There is an extra port on the TB that is not used in my car that I had to put a bolt in then some tape. It remains sealed.

I am going to have to have it looked at.

[e301988325i]

There point was that premium products sell more slowly and then go stale in the tank underground.

Why are you using 99? is your car tuned for it? The standard BMW motronic ECU has no way of knowing it should advance the ignition timing to take advantage of the higher octane rating. With your compression ratio, 95 would be fine.