Tandem Mi Engine Build

[taffycrook 2]

Interesting idea. I have done much of what you intend to do here. Without the inlet manifold swop, and custom exhaust manifolds. Slightly less CR on the 2.0 and std 2 row management. The 1.9 had a slight head skim, 3 angle seats and the cat cam inlet. Rest of the spec is as per the 2.0, in fact its the same gear.

Results were interesting, I will post after your tests but I bet we are pissing up the same leg, and telling everyone its raining.

[petert 588]

The correct procedure is to do a trial assembly, without liner o-rings or piston rings, after clean up, of the bottom end, before decking. Then measure exactly what needs to come off to achieve zero deck height at TDC (piston flush with liner). A skim of 0.4mm (XU9) and 1mm (XU10) will be on the money however if you're feeling slack. You can even push the piston above deck if you're really keen.

 

The liners should sit approx. 0.003" above the block, on both sides of the liner. The block often sinks half a thou. on one side, meaning the liners need to be machined unsquare to account for the difference.

 

Std. head gaskets are fine.

 

At the risk of erupting a huge debate, the difference in CR of 0.2:1 is not that significant. The important thing is getting the deck height the same on both engines. Thus, just the thickness of a head gasket between the piston and the head. You're ensuring the combustion chamber distance and burn rate is the same for both engines. You will need 98RON fuel for both engines.

 

My prediction is that the improvement in power/weight ratio will overcome the additional mass of the iron block.

 

You'll need to retain the 1.9L injectors but be prepared to add an adjustable fuel regulator if necessary. GTi6 injectors will be too big for the std. mapping.

Edited July 27, 2009 by petert

[leon 1.9 0]

EXACTLY!!! . I'm really only planning to build the engines. The cars are likely to be someone elses; but that's quite some time away, and may suit someone who has a need for an engine in 5-6months' time.

 

If I really was trying to 'settle' this argument, I'd do it the way you suggested. So I suppose this 'versus' thing is not entirely true from a strictly engine PoV. It's more of a 'how do two more or less identical 205s with these more or less identical engines perform?'

 

My current plan is to abuse my familiarity with Jeremy Clarkson's PA and get Lucinda to get him to let a few of us (naturally the car owners) go down with these cars to the TG test track. Then back to backs - purely for fun as it's always been - could be done where each driver drives each car. A set of averages would show a trend.

 

Justin (Kyepan)>

Hmmmmm. I am sure there is pure fact and logic in that, but allow me to ask something. This is assuming quite rightly that a certain rate of air flow is required, which at best case must match (or slightly exceed to give a margin of error) the potential max flow and air requirements of the Mi head, i.e. at theoretical full tilt @ 250cu ft per min. Is that the correct units and a correct statement?

 

If it is, that is a volume of air at a certain rate, then perhaps (read: I'm sure) there are calculations that could be done to see if even adding a 180 bend in ~72mm internal silicon hose would permit that rate of airflow. Yes it will slow air down travelling through it, but at worse case, with static air (i.e. a stationary car) does it still permit that potential maximum rate of air through? If flow rates approx equate to BHP, then perhaps we only need to allow 200 cu ft per min to leave that margin?

 

So I agree J this approach would slow air down, but would it be sufficiently restrictive to a rate of 250cuft per min through the ~72mm int dia AFM/silicon combo inlet?

 

I'm no engine builder, but I know a little about fluid dynamics. I'll give a few little bits of information worth reading, but I can't add any 'practical' knowledge to the discussion really.

 

As Petert stated earlier, cfm and temperature are both important to inlet performance, in fact they're important no matter what type of fluid system you're designing (and air is a fluid, before I lose you ).

 

However, your real aim is for the highest molecular quantity of oxygen into the engine. To achieve this, you need the highest volume of air available, at the higher DENSITY available.

 

Density is related to temperature, pressure and water vapour. Now we can't really do much about the water vapour content of the air introduced to the engine (apart from ensuring our inlet isn't scooping water into it, which isn't going to be too good for the engine anyway), so we have two factors we can influence:

-temperature, and

-pressure.

 

Take a look at this article, which was previously posted by someone in another thread on here:

Autospeed negative boost

 

Now, in that article, through fitting a different air duct to the front of the car and a few other minor mods, the guy dropped the pressure restriction by about 12.5" Wg, which is 32mB.

 

Using this handy little calculator:

density-temp-pressure air calculator you can work out that the change is equivalent to a drop in inlet air temp of around 15 degC, which is quite an improvement. Ok, its not HUGE, but as you've said many a time, lots of small improvements add up.

 

Its probably true that these changes are very hard to measure on a dyno in a workshop or an engine dyno, so I'm willing to bow to those who know much more that feel the benefits are negligable.

 

Makes some interesting reading (well, if you're sad like me) anyways, but I guess its not all that relevant to the topic. The important thing for your tests is to make sure that both the inlets are the same. Another thing to remember is that you'll want to be checking as many variables as possible during future dyno tests. For example, you could easily have figures skewed by one engine running slightly lean under load or similar. I'm yabbering now anyway so I'll get back to work

[DrSarty 90 1 Cars]

I remember the Autospeed article and site very well. I seem to remember it was Justin (Kyepan) who linked to it because he spent so much time on it. I did too for a while reading everything.

 

I found your post very interesting so don't put yourself down.

 

I never related pressure to temperature although I should've really having A level physics and doing (but failing) A level chemistry, so I can see what you mean. Now the penny drops completely.

 

This is why, in Autospeed and in other airfeed location tests the goal is to find the highest air pressure entry point for the airfeed, such that any restrictions (negative boosts) do not effect the engine's performance.

 

I seem to remember there being another Autospeed test where they searched on several cars for the ideal location for an airfeed. I seem to recall the best places commonly were at the bottoms of both the front valance and the windscreen. Quite understandibly these locations aren't always practical and hence what you end up with is a compromise.

 

This could very well be where the misleading notion that a cone filter is suddenly going to unleash hidden horsepower from your car came from. Arguably it has nothing to do with the cheap paper filter as either the manufacturers claim or people perceive, but it's a combination of doing away with negative boost sites hidden within the standard airbox and associated trunking AND the positioning of the cone filter in a better (perhaps heat shielded) place in the engine bay.

 

So as you said (and said that I said ), if you can avoid restrictions in the air intake (whilst retaining good filtering to avoid engine damage), keep air pressure up and temps down, this will all add up with all of the other little bits and perhaps net 10 extra horses, which is apparently noticable.

 

I have heard that for each 6degC (or it could be 4degC ?) inlet temp drop you can achieve you gain 1 BHP. So your example of pressure drop reduction causing a reduction in temp of 15degC is nearly 3BHP in the hat. Now just imagine that the triple cut valve seats gives you 4BHP, the 4-2-1 exhaust 6-7BHP and the new cam on higher CR finds you 8BHP, then you can see that we may have upped a 160BHP Mi16 to 182BHP. Every little helps.

 

So I'm taking airfeed very seriously, which is again why I valued Justin's comments, and yours too Leon and gone and read the whole Autospeed site all over again.

[GLPoomobile 958]

Now just imagine that the triple cut valve seats gives you 4BHP, the 4-2-1 exhaust 6-7BHP and the new cam on higher CR finds you 8BHP, then you can see that we may have upped a 160BHP Mi16 to 182BHP. Every little helps.

 

Careful now. It's that line of thinking that the MaxPower boys use to make their little 1.1 Corsas put out 250bhp

[welshpug 1,661]

I know its likely to be down to cost, but why stick a nice big GTi6 throttle body and manifold on there, only to stick a restriction in the tract, aka an AFM?

 

Especially when the 2.0 Mi16 has a perfectly capable Motronic MP3.2 Map based coilpack equipped management system, far nicer running than anything with an AFM in my opinion (I put together a 205 with this combination of 2.0 RFY engine and management and RFS intake manifold last year) though I don't think its ever been on the RR, it worked very well.

 

 

(RFY = 405 2.0 Mi16, 306 S16, ZX 16v, RFS = ZX Dakar, Xsara VTS, 306 gti6/rallye)

[Batfink 201]

paper filters when new flow very well but effectively get blocked up when dirty. 'Performance' airfilters flow virtually the same when dirty as the filters job is to hold oil in the airflow to catch the dirt in the air as it goes past.

[DrSarty 90 1 Cars]

Careful now. It's that line of thinking that the MaxPower boys use to make their little 1.1 Corsas put out 250bhp

 

You very funny guy! I thought those figures were quite reasonable.

 

And what do you guys think of the link below?

 

I found it on the Autospeed site, and which may address the high load leaning issue with the fuel which I may have, as the Mi injectors reach peak. On a related article it explains how variable FPRs are useful IF you have a mappable ECU. With a standard ECU it says the fuelling across the lower part of the rev range will be rich.

 

These means I can focus on just increasing the fuel pressure in the rev range where it's needed so that the mixture doesn't lean out. It seems doable for £100 which sounds reasonable, and a bit of an added challenge too.

 

http://www.autospeed.com.au/cms/A_2619/article.html

[DrSarty 90 1 Cars]

I know its likely to be down to cost, but why stick a nice big GTi6 throttle body and manifold on there, only to stick a restriction in the tract, aka an AFM?

 

The GTI6 manifold is better designed. The AFM is roughly the same size as the AFM (72mm IIRC) used on the Audi in the Autospeed article above, which is a 5 cyl turbo charged S4, so I'm not sure it's much of a restriction to airflow. It has no gauzes or meshes in it either to create negative pressure drops, and Peter already gets 180BHP out of one.

 

Kev>

As for paper filters again I'm afraid I don't agree. It's shown in the same article with the guy measuring pressure drops before and after it, and the conclusion is that they are not a restriction in the sense that you mean. I really, really do believe what I said earlier about why aftermarket cone filters are shown to be effective; that it's because it's NOT just a replacement for the badly viewed paper filter, but because by having it in a better place with heat shielding, and perhaps removing other negative pressure drops by changing the factory design, these other factors improve flow. I.E. it's not through the removal of a restriction added by the standard filter, and this guy's tests prove that.

 

Would someone kindly measure the dimensions both ends of an Mi AFM just for fun please?

Edited July 27, 2009 by DrSarty

[AndyJ 1]

Would someone kindly measure the dimensions both ends of an Mi AFM just for fun please?

 

Manifold side is 68mm dia, filter side is 40mm square (with the square plate on the end), looks like the actual casting itself is about 51mm square if you take the plastic plate off.

[monty69 6]

Manifold side is 68mm dia, filter side is 40mm square (with the square plate on the end), looks like the actual casting itself is about 51mm square if you take the plastic plate off.

Drsarty i admire your enthusiasm for everything 205. Its great to see projects like this and prevent us from learning the hard way. If i ever find the time and funds i will do a similar project of a s16 acav with larger throttle body vs gti6 intake, showing just how much better the s16 manifold is for a road car while not affecting top end power.

[Batfink 201]

The GTI6 manifold is better designed. The AFM is roughly the same size as the AFM (72mm IIRC) used on the Audi in the Autospeed article above, which is a 5 cyl turbo charged S4, so I'm not sure it's much of a restriction to airflow. It has no gauzes or meshes in it either to create negative pressure drops, and Peter already gets 180BHP out of one.

 

Kev>

As for paper filters again I'm afraid I don't agree. It's shown in the same article with the guy measuring pressure drops before and after it, and the conclusion is that they are not a restriction in the sense that you mean. I really, really do believe what I said earlier about why aftermarket cone filters are shown to be effective; that it's because it's NOT just a replacement for the badly viewed paper filter, but because by having it in a better place with heat shielding, and perhaps removing other negative pressure drops by changing the factory design, these other factors improve flow. I.E. it's not through the removal of a restriction added by the standard filter, and this guy's tests prove that.

 

Would someone kindly measure the dimensions both ends of an Mi AFM just for fun please?

 

I dont understand what you are disagreeing with. Did they compare a clean paper filter and a used one, then against another performance panel filter?

I'm not saying anything about improving the flow and position, just how the filters work. I agree that filter location is important and i'm not questioning the article results of modifying the induction.

 

Edit: I've just read the article where they measure airflow in the airbox. It doesnt actually tell you anything other than there was no restriction from that airfilter. To make it a valid point you need to know the condition. A new paper airfilter is good at flowing air, but that will decrease with use, otherwise you could just bang it out to remove the big bits of dirt at every service.

Edited July 27, 2009 by Batfink

[DrSarty 90 1 Cars]

Manifold side is 68mm dia, filter side is 40mm square (with the square plate on the end), looks like the actual casting itself is about 51mm square if you take the plastic plate off.

 

Thank you Andy.

 

I remember the plastic thing as I'm sure a new one with 4 holes in it comes with the K&N cone filter I used to have, which the rubber base of the filter (with great difficulty) pushes onto.

 

That Autospeed site recommends no less than 3inches (75mm) round internal diameter of air feed ducting (or equivalent area if not round) for more than 200BHP.

 

This 40mm square thing seems tiny for 160BHP, although it must work, especially if people can get a genuine 180BHP through it. The 68mm side seems just fine.

 

I believe I would remove that plastic piece to get at the 51mm AFM body casting as that is at least closer to the 75mm recommended for 200BHP plus. The 40mm square size is only the equivalent of a 45mm round duct. At least the 51mm equates to a nearly 58mm round duct, which is only 17mm under what I would expect rather than a whopping 30mm.

 

What I would be looking for is no 'negative boost' sites in the ducting/AFM combo, which would induce pressure drops. To that end, the underside of the heat shield and flow director around the cone filter would have a cold, high pressure air feed added in from behind the front valance. This would counteract any pressure drops and potentially lower the temp further of the intake air charge. This is all beneficial to better cylinder filling. Yummy!

Edited July 27, 2009 by DrSarty

[DrSarty 90 1 Cars]

Drsarty i admire your enthusiasm for everything 205. Its great to see projects like this and prevent us from learning the hard way. If i ever find the time and funds i will do a similar project of a s16 acav with larger throttle body vs gti6 intake, showing just how much better the s16 manifold is for a road car while not affecting top end power.

 

Cool Mr Monty.

 

Go and have a good look through the Autospeed site. You/I could get lost in there for days!!

 

It mentions inlet plenums and runners here.

 

It goes to the extent of showing basic calcs (and even software available) for determining inlet runner length and plenum volume. It seems in variable inlet systems like ACAV you can change either, by varying sizes or having two sizes to choose from. Fascinating.

 

I would say that the ACAV system is probably part of the whole design incorporating the ECU and the map too, and compression ratio and valve timing is all linked. That's my guess rather than fact.

 

When it works it must be excellent and I've heard good reports, as effectively you have two sets of runners and/or two plenums, to get the engine's power curve much more even, i.e. getting rid of any dip(s). Should pull well at all revs, which in turn could mean that non-ACAV systems are in fact a compromise, and the mapper must choose to go one way or the other, or compromise somewhere in the middle. I think the R32 and maybe R36 also has variable inlet tract technology and it's there for a reason to make a very driveable car with power access across the whole rev range.

 

Nice stuff. The only downside I hear is reliability. If it goes poof, surely the performance suffers (?).

 

This is why I gather some say the GTI6 inlet is a good design. It has long runners (good for torque/torque spread), a large plenum and TB and flared runner entries.

 

Very interesting article above as it sort of explains why high power i.e. high BHP usually comes from using shorter runners. It's to do with pulse timing.

Edited July 27, 2009 by DrSarty

[shalmaneser 26]

Wow!

 

This looks great and i look forward to seeing it develop!

 

I'd have to say though it seems a shame to use an AFM with these engines - It's always struck me how much of a restriction those big old things must be!

 

Either way this should be facinating to watch! good luck and take plenty of pics!

[GLPoomobile 958]

I believe I would remove that plastic piece to get at the 51mm AFM body casting as that is at least closer to the 75mm recommended for 200BHP plus. The 40mm square size is only the equivalent of a 45mm round duct. At least the 51mm equates to a nearly 58mm round duct, which is only 17mm under what I would expect rather than a whopping 30mm.

 

I don't have an AFM in front of me to compare, but I think you've misinterpreted the info you've been given. The plastic piece (top hat adaptor) doesn't form any restriction, it's matched in size to the square inlet of the AFM. So removing it won't give you a bigger hole. Or are you suggesting that you could machine the AFM body to make the inlet bigger? Not sure that would be a workable idea

 

It's also worth mentioning, although the outlet side of the AFM is bigger than the inlet side, it's bigger because it's sort of flared out from the AFM body. The actual passage through the AFM is still much smaller than the outlet.

 

I wouldn't get tied up in all of that though. As you said, if it's been proven for 180 odd bhp then it's going to do the job.

[DrSarty 90 1 Cars]

Manifold side is 68mm dia, filter side is 40mm square (with the square plate on the end), looks like the actual casting itself is about 51mm square if you take the plastic plate off.

 

I'm not quite sure Steve how I could've misinterpreted the above that Andy J wrote. I'm also not sure where you thought I might ever machine or open out the body of the AFM from either; I only said 'get at'. Either more swahili from me I think or people continuing to interpret what I write rather than take it for exactly what is written.

 

Perhaps you'll double check Andy J's measurements for us, because you say the 'top hat' thing isn't a restriction whereas Andy says it's only a 40mm square hole when the other side's 68mm square. That strikes me as quite a size difference and hence why if I didn't use it, I'll somehow try and plumb into the 51mm AFM entry. In fact if you or anyone else could post some pictures of both ends of an Mi AFM including Mr top hat it would be most useful.

 

I completely agree with the below though, because of the reason stated and also that larger displacement, more powerful turbo'd and N/A engines use AFMs still which aren't much bigger.

I wouldn't get tied up in all of that though. As you said, if it's been proven for 180 odd bhp then it's going to do the job.

Seeing some pics and having the measurements double checked would help me clear up things in my mind though, even if I am agreeing with you that it's nothing to get tied up about. It's more a case of understanding it. I too was led to believe that an AFM is a restriction. Of course anything in the induction line is. Even the interior surface of the intake line creates friction and impedes flow; it's unavoidable.

 

But my point since we started discussing this, and is exactly what I was saying to Justin, is how small, long or curved does it have to be before those restrictions have a noticable impact on the performance capability of an engine? And again, new cars still use them, and I cannot believe they would continue to use something that would overly affect the performance of their cars. Improvements can be made for sure (removing the 'negative boosts'), but they can't be totally restrictive or they wouldn't be used.

 

So someone posting some pics and confirming the entry dimensions of both ends of an Mi AFM would be greatly appreciated.

 

Here's my idea for the inlet, including a heat shroud and air flow director (the yellow bit) for around the AFM and into the filter. The ally or stainless shroud will be highly reflective on the outside and matt black on the inside. The front grille location is already good for a supply of higher pressure air and therefore there may be a cold air feed duct with flared funnel into the underside of the shroud from low down behind the front valance. This should ensure not only cool air but higher pressured air, which will not only reduce intake charge temps, but make the charge denser and pressured enough to overcome any 'negative boosts' imposed by the AFM and curved inlet into the TB. (He hopes)

 

Please excuse the crappy PowerPoint drawing. I hope you get the idea.

 

Edited July 28, 2009 by DrSarty

[kyepan 291]

keep looking at auto speed rich, the answers you seek to induction are there.

 

Especially how standard air boxes using panel filters (clean new ones) can be made to flow with virtually no restriction, and very very good cleaning. Go back to the negative boost revisited articles.

 

Secondly I still think your siting of the afm and filter is wrong with needless bends and twists, your best high pressure source is behind the passenger (fog) driving lamp, go up from there into a suitable air box procured from another car, and then straight through the afm into the throttle.

 

Give yourself a cold, direct air feed that is not sitting on top of the radiator (metal object that radiates heat) or in front of the inlet manifold(metal object that radiates heat) ..

 

Saying that I've actually purchased a manometer and some pipe to test pressure areas on the front of the pug, but with one thing and another, i've not got round to actually testing it. Once again autospeed have done this, and just below the front bumper, is where they recommend.

 

The inlet and exhaust for the air box can be modified (bell mouthed pipes) autospeed style to help air flow, and a large panel that will give low resistance to flow.

 

No need to move the radiator at all, or any of the complex pipe re routing etc etc.

 

Keep it simple stupid... If only i listened to my own advice.

 

Instead of placing two un needed bends in your system, place one, and use it as the filter at the same time.. therefore the disruption caused by the filter and the bigger space for the air to turn will lead to less restriction.

 

the other good thing to do is get rid of the awful ribbed affair that masquerading as a pipe, smoothness is better, unless it's there to produce turbulence ahead of the throttle for better acceleration (like the stepped hull of a speed boat) when the throttle is opened...

 

J

Edited July 28, 2009 by kyepan

[GLPoomobile 958]

Perhaps you'll double check Andy J's measurements for us, because you say the 'top hat' thing isn't a restriction whereas Andy says it's only a 40mm square hole when the other side's 68mm square. That strikes me as quite a size difference and hence why if I didn't use it, I'll somehow try and plumb into the 51mm AFM entry. In fact if you or anyone else could post some pictures of both ends of an Mi AFM including Mr top hat it would be most useful.

 

Dude, the plastic adaptor isn't the restriction! It's the actual mouth of the AFM! That's what I'm on about, and why I'm confused about what you are suggesting. The plastic adaptor is matched (dimensions wise) to the mouth of the AFM. Taking it off doesn't give you a bigger opening. At least that's the case with all the plastic adaptors I've seen.

 

Although I'm now starting to wonder what plastic adaptor Andy has if it's only 40mm square . Looking at a ruler, that is far too small, and 50odd mm sounds more likely.

 

EDIT: Here's a picture with dimensions of the Green Filters top hat adaptor

 

http://www.green-filter.co.uk/images/Top%2...01V%20Large.jpg

 

I'm pretty sure the K&N style adaptors are more or less the same dimensions as the Green ones, and I'm pretty sure the K&N ones are more or less the same as the OE ones.

Edited July 28, 2009 by GLPoomobile

[GLPoomobile 958]

Saying that I've actually purchased a manometer

 

A manometer?

 

Are you having doubts since buying the MX5?

[kyepan 291]

A manometer?

 

Are you having doubts since buying the MX5?

 

no... it's awesome.. wish i'd bought it years ago, it's so much more composed than the pug, and goes sideways. And, i'm not ashamed to say that i did stop last night after parking it up and exclaim out loud that it was a sexy little car... it looks so low and minimalist, like a slipper. Silver slipper of man love. The gay grey battleship of bum love.. It rocks like a 70's rock star in a one piece power suit playing a power balad, it's the gay grey hom rocket from the gods.

 

Be proud.

 

Irony is I have a good mate who is gay, and will be taking him for a spin with the top down when he gets back from his holiday. He's got a big hair cut, that flaps about like poodles ears... it will be hom's-o-clock.

 

anyway... if there is anyone gay on this forum i hope they do drive mx-5's and laugh at anyone who takes the pisss, as everyone else is missing out IMHO.

 

May fit the 24mm arb back in the pug this weekend.

 

Manometer measures relative gas pressure...he he, and between You and sarty, there is enough hot air on this forum to tip it off the scale.

 

speaking of which, rich, shorter sentences please... you're long rambling ones are a PITA to read. Protective bumper stickers... Cotton wool and all that.. you know.. the usual kyepan posts The readability quotient of your posts is el-gasho atm..

 

nice punctuation though, keep up the good work.

 

J

Edited July 28, 2009 by kyepan

[DrSarty 90 1 Cars]

Poor capitalisation Justin. See me!!

 

 

No: thanks; both of you.

[Batfink 201]

Justin - maybe you shouldnt be reading this at work... then you'd have more time

 

saying that...i need to do some work....

 

K.

[DrSarty 90 1 Cars]

saying that...i need to do some work....

 

And the rest! Stop mincing and get your car running!

 

J> Ref your post concerning placement. I covered everything you said; I guess you didn't notice because my sentences were too long...for you. The air flow will be supported by a high-pressure, low mounted air feed. The heat from the rad and plenum will be shielded by the shield.

 

The bends I wholeheartedly agree with. And once I'm standing in front of these cars looking at the real world space available I'll make a plan accordingly.

 

What I was saying was that the suggested bends for this layout will be large, smooth and a vast improvement over even the standard 405 Mi16 inlet tract.

 

And this makes 160BHP, so I'm not entirely convinced that they would be such an issue. Happy to be proven wrong as ever.

[GLPoomobile 958]

There is an alternative, along Justin's lines.

 

Mount the AFM and filter as per normal for an Mi transplant, have the washer reservoir relocated so you've got a decent space in the inner wing, then cut and flare a hole in the bottom of the inner wing under the filter. You can then duct your air feed in through the hole from a high pressure scoop in the valance. And if you want o go the extra mile, construct an enclosure around the filter in the inner wing area to keep it high pressure and protect it from heat soak.

 

But I would like to point out that I think all of this is possibly a bit pointless. Perhaps it's going the extra mile for the sake of it. I just think if these engines are capable of producing the power you are expecting with the usual AFM arrangement, then is there any need to go to all this extra work?