2.1 16v Tb Inlet Manifold/tract Length Change

[Sandy 189]

MOS contacted me a while back about the problems he was having getting his engine running smoothly. I'm always a bit apprehensive about trying to offer solutions with just some figures and subjective comments to work on and not being able to see the car, but I couldn't help thinking it might be worth trying a different inlet set up. the existing spec was:

87x88mm Mi16 Iron block

12.5:1 CR forged pistons

Wet sump

QEP Mi16 head with 3 angle seats, blended throats (but no visible porting on the inlet side)

Standard valve sizes

Catcam 4901551 cams with hydraulic lifters

4-2-1 Exhaust manifold and custom system

Jenvey/Longmans inlet manifold (single piece type)

45mm DCOE pattern KMS bodies

60mm KMS trumpets (overall tract length from valve to mouth approx 310mm)

KMS ECU with 340cc injectors

 

I had previously made up an inlet manifold with long runners and upside down DCOE flanges to allow much longer tracts and two sets of injectors (4 in the head and 4 in the bodies) for a 205 i'm doing for a friend as a long term project, but haven't got to the stage of fitting yet. I sold the original manifold and made another, but I don't think that one has been used yet.

Mounting the bodies upside down allows them to be much closer to the bonnet, improving the tract shape and aiming them down at the slam panel, so the trumpets can be changed without major space issues.

When MOS arrived we immediately put the car in the rollers (SRD again, helps to always use the same rollers!) and ran it up to take a set of base figures before any changes were made, so we could assess the real affect, i'm glad we did as it turned out. We got as far as 6000rpm, but after a cooling off to get the temps back to base, it wouldn't start, turned out to be the starter lead loose, but we carried on with taking off the inlet as time was pressing on.

Initial figures at the wheels:

2000 38

2500 55

3000 58

3500 74

4000 96

4500 103

5000 136

5500 152

6000 160

 

Once I'd relieved the manifold to clear the dowels etc, it was a straight fit and a case of extending the hoses and throttle cable (at great expense, had to get a new one from Pug sadly). We fitted the injectors in the bodies, worth pointing out that we could only run one set with the KMS, it sadly doesn't run staged twin sets like the DTA, more on that later. This injector position is quite far up the tract, but not dangerously so.

With 20mm Pipercross trumpets (just to trial) we had an overall tract length of approx 370mm, 60mm increase and what I generally aim for in my manifold designs.

Time was getting on at this point and we only had time to try it from 2-3500rpm at this stage before calling it a night. we already had these figures to take home:

At the wheels

2000 32

2500 65

3000 65

3500 81

As you can see, apart from 2k, a slight increase and I was encouraged, but nervous about what would happen beyond 3.5k!

 

Yesterday morning we checked the cam timing and found the exhaust cam was as near as spot on, but the inlet on 4.05mm lift at TDC, against a spec lift of 3.55mm, so slightly advanced. We ran up to 4-5k rpm and immediately saw a huge leap in power which we would not have believed if we hadn't seen it with our own eyes, nearly 40bhp increase at 4500rpm!! we tried laying the cam back to spec timing, but the power fell away, so we set it back as it was and mapped the engine up through.

The power just seemed to keep increasing and we had a job keeping it in the rollers (always a good sign!), the final figures are as follows:

At the wheels

4000 105

4500 141

5000 150

5500 178

6000 184

7000 200

So going by our usual expected flywheel to wheels losses (from experience of engines being run on the dyno and then in the rollers), peak power is approximately 230-235bhp at 7000rpm and peak torque a barely believable 197lbft@5500rpm!

It sounded so different as well, it's funny how engines sound right once they are working properly!

So here's a crude comparison graph:

 

 

We then went out on the road to trim and improve the map, it's the only way I find, to be sure that it's all doing what it should and it constantly amazes me how much part throttle needs to be trimmed. We soon found that at normal driving speeds there was a boggyness accompanied by a strong smell of fuel vapour, which seemed to indicate stand off. The quick easy solution would be to move the injectors inboard. We went back to the workshop to do it and fitted GTi6 injectors (265cc at 3bar). The 3bar reg in the GTi6 rail turned out to actually run at 3.2bar (275cc net), we'd been using the 340cc injectors at 2.5 bar, which means 310cc net. I added 15% to the map to compensate for the reduced injector size, but immediately took it back out again when we tried it and it transpired it wasn't needed! the GTi6 injectors (275cc net) in the head were giving the same mixture, for the same duration as 310cc in the bodies, over almost the whole map! An abject lesson in the importance of injector position and pretty much the same effect as I got in the Trio before. The only "fluffy" fuelling we seemed to be getting now was a small throttle openings over 2500rpm, where the duration was down to 2ms or less, which is down to injector control more than likely. the GTi6 injectors were running at about 85-90% duty at the top end of full throttle, so an almost perfect injector choice.

Twin injectors staged, would allow both positions to be used, improving resolution at low sites and mixing/spare capacity, at high sites. not an option though without an ECU change.

It was getting pretty late now and MOS and Ms MOS had the drive back to Manchester to follow, so we wrapped it up.

I'm really really pleased with how this turned out and it's raised some interesting questions to my mind about what the priorities should be if you would a chunky powered road going 16v XU!

[Sam 4]

Bloody great results. It seems people get various results with longer induction but thats probably due to other reasons, who knows?

 

You're tempting me to "downgrade" to N/A all the time

[eeyore 0]

Bloody great results. It seems people get various results with longer induction but thats probably due to other reasons, who knows?

 

You're tempting me to "downgrade" to N/A all the time

Its not a down grade Sam!

[Sandy 189]

The only variable other than the manifold, was the mapping, but who knows, the power really did seem to leap out of nowhere.

[welshpug 1,626]

very interesting read, you dont by chance know how long the inlet tract is on the gti6?

 

I was wondering what exactly makes such a difference between the ACAV and gti6 inlets, as the injector positioning is near identical.

[Sandy 189]

I don't know, but I would expect the ideal length to be much different for a plenum compared to TBs and cams, tract shape and so on all add variables.

[Dj_mini 7]

Great result agaian sandy! Did you get any pics of the set up in the car? Do you think there is more power to be had by having diffrent length trumpets or is that as good as its going to get?

[Sandy 189]

Camera phone only, apologies for the quality! Fuel rail in this pic is temporary to hold the dummy injectors in, it was still using the outer injector location at this point:

 

I was aiming for 360-370mm, so it worked out great like that. We could have tried longer trumpets again, but the results as it were were close to the maximum likely BMEP and the slam panel would have become an issue. Mark (MOS) still has the challenge of getting an air filter back on it now. It's going to need the slam panel cut and reinforced (he doesn't want to use pins) and the back plate re-cut lower and upside down for bonnet clearance. i really think it's worth it for the gains though!

[Henry 1.9GTi 36]

nice, interesting read, good work.

[James_m 0]

12.5:1 CR!!!

Nice result. Must be an animal on the road.

[Bonzai 2]

Ive got the original manifold. Some nice results. I wish i could afford to take my car off the road and play with my setup more, but as its my only mode of transport i cant really see it happening, and i think James_R has been eying up the manifold for his own project!

 

Heres the only pic i have of a retro fitted manifold on one of James's previous engines

 

http://img.photobucket.com/albums/v223/205...le/DSC00442.jpg

[B1ack_Mi16 67]

Always nice to see good results from hi spec engines

 

One question, did you have a problem about low rpm hi load pinking?

My 2.3 with "only" 11.6:1 compression seem to have pinking issues at wot around 1500-2000rpm even with ignition retarded all the way to 0 degrees.

[Sandy 189]

It was quite boggy that low down and you shouldn't really be trying to use full throttle at those revs in a fairly highly tuned engine really. We had to back it off around 2-2500 3/4 throttle to make it safe, but above that it was fine.

[pug_ham 241 3 Cars]

the final figures are as follows:

At the wheels

4000 105

4500 141

Crikey, that is a major hike in power over 500rpm & in the mid range so it must be an beast to drive, especially as it continues to climb at a good rate on up to 7k.

 

Brilliant results yet again Sandy, top work. I bet Mos is well pleased.

 

Graham.

[mos 1]

he is!

[Mandic 0]

This inlet tract length is still a bit of a mistery to me.

 

One day one sees good results with short tract and the other day with longer one.

 

Cheers

 

Ziga

[Sandy 189]

It's not a definite science IMO, too many variables involved, you can only try it. But I do think that too many people bolt on short tracts as it comes off the shelf, assuming it'll work well enough and possibly deny themselves the full potential of their investment as a result.

[Mandic 0]

Agree there are many variables, but in this case, where lack of top end power was assumed, I wouldnt think that longer runners can solve the situation, at least partly.

 

In previous post I was referring to computer simulations and other inlet tract calculations, where each give different results. For example Engine Analyzer Pro gives better top end torque with shorter tract, where when You go and calculate Helmholtz resonances and other stuff You always seem to get weird results when You put calculated stuff in EAP.

 

I guess it's more like trial and error stuff.

 

Cheers

 

Ziga

[Daz_C 8 1 Cars]

Excellent Result ! I bet the drive home was a blast

 

You've just made me more pissed that I haven't got time to get mine sorted at the moment.

Just need to make a loom and your manifold will be coming too good use Sandy

[Sandy 189]

Good stuff daz, look forward to seeing how it works out.

[James_R 3]

This confirms to me to yoink this manifold off Ben, I'm having a 2.1 GTi-6 varient engine built at the moment.

 

Results will be in Feb once I've got it all in the car.

 

Sandy I've got a set fo GTi-6 injectors and Saab Reds I've got the oppertunity to run 2 banks, would you suggest the smaller injectors at the head end and the larger in the bodies? I'm guessing so from what you have said above?

 

Thanks

 

James

[Sandy 189]

With twin injectors run the same size in both locations and I'd expect around 180cc in each bank to work best for you. Alfa orange injectors are plentiful on ebay, about the right size and cheap. GTi6 ones are in demand and you'll probably make a profit in the process!

I recommend using automatic switching on duty cycle, because that relates best to the air volume/speed and in my experience makes the switching point undetectable. If you use two sizes, the switching point will be obvious!

[James_R 3]

Brilliant thanks Sandy think I've got two sets of opel red top yellow's knocking about I'll give them a try.

[Sandy 189]

MoS has asked me for a torque comparison, so here it is:

 

[danpug 1]

That thing must be grunty as hell.