Cart Engine RPM's

[ZoRG]

I had a look at Cosworths website and have notice that they used to run Cart engines with around 16000RPM's, but have now dropped it down to 12000RPM to make the engines last much longer.

My question is though, what kind of valve timings are they running to get a turbo charged engine to rev so high?

Cheers

 

[Colvindesign]

one thing that you should look at is square boring and oversquare boring. This is when the bore is nearly the same as or larger than the stroke. This, for example, is how a Suzuki 1000cc engine can rev up to 12,000 rpm, while having a piston speed 2% faster than a Honda S2000 engine at 8300 rpm. Since the Honda engine is nearly square, and the Zuk engine is about 1.23 bore to stroke.

This is one way high rpms are attained. And valve timing is usually a little different pfrom application to application, with larger overlap allowed for turbocharged examples.

 

[ZoRG]

I thought overlap is a nono in turbo applications due to reverse flow on the exhaust side?

I am busy building a 1600cc 4 cylinder engine, with a bore of 86mm and a stroke of 67.4, it is a factory turbo charged engine and I would like to play around and see how high rpm can be attained... This is a purely experimental engine and I dont really care much for reliability, sort of more of a hobby and a learning experience, I have facilities to put together a pneumatic valve system, and am busy trying to get a license for technology that allows the use of ceramic sleeves/liners, also planning on using carbon-carbon pistons and titanium or carbon-carbon valves, I have seen a F1 cam and it looks like a 360 degree cam, someone on a F1 forum confirmed that they are around there, the whole question is though:
does the same valve timings/durations work on turbo charged super high rpm motors than on NA super high rpm motors? I would have to get the cams custom made from billets, so I would like to have some guidence on where to start, I do not have a mountain of money so I can't afford to buy 10 cams ...

Any idea how one can work these things out?

Thanks
Jonathan.

 

[Azmio]

Zorg
Interesting! Not sure the real reason of why the cart engine rev is dropped from 16000rpm to 12000. One thing that I can think of is to stop the team from considering the use of pneumatic valve. This will increase the engine cost of the cart engine. The spring is just not reliable for that sort of rpms. At 16000rpm, I am not surprised if the valve floats and cause engine problem during the race. It aint fun to see lots of engine problems during race.
Anyway, back to your question, cam timing is preferred to be different for both NA and boosted engine. The reason is simple, NA needs a lot of valve overlap to use the exiting exhaust gas to pull the intake charge into the combustion chamber.
With the turbocharged engine, the airbox is pressurized much higher than the ram air can pressurized the airbox. With the same cam timing overlap, some of the intake charge mixture will be pulled into the exhaust port and cause higher fuel consumption and lower volumetric efficiency.

AO

 

[mister2]

Azmio - thats not entirely correct for turbocharging, in fact you seem to have described a supercharged application. turbo'ed engines have higher exhaust pressure and so during overlap tend to have reversion rather than blowthrough. but that doesnt help the original question

nick

 

[ZoRG]

Actually mister2 with a properly sized turbo, you get close to a 1:1 ratio if one can achieve this, then the standard NA performance cams should work just as well on a turbocharged engine than a NA engine since there is no difference in pressures anymore, the reason there is backpressure is due to factory tc applications using turbo's to small in an effort to reduce lag, and make a car more driveable, however when a larger turbo is fitted and one gets to the 1:1 ratio, where does this leave cam theory??

 

[patprimmer]

You still need less overlap in a turbo as the tuned length exhaust aids chamber scavaging in it's working range on a NA engine, but the turbo confuses this, so the gains from long duration near TDC on the overlap stroke are reduced, as not much is happening to scavange the chamber.

On a mechanical supercharger the same is true for a different reason. The boost pressure greatly improves chamber scavanging and not so much overlap is required to get a good result. More just wastes inlet charge which goes straight down the exhaust pipe

Regards
pat

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[Azmio]

Nick,
Thanks for your feedback. Anyway, I will still believe that NA engine needs more cam overlap if compared to supercharged or turbocharged engine. The reason is simple, with the pressurized airbox or plenum, the positive pressure from the plenum will cause the air to flow into the combustion chamber once the intake valves are opened. On the other hand, the NA engine has the airbox or plenum pressure much lesser than the charged engine, that's why we need large valve overlap at high rpm to make a full use of the scavenging effect from the exhaust gas.
A good example is the Mazda miller cycle engine, it works well with the supercharger. With the same cam timing as the Mazda miller cycle engine for NA application, the volumetric efficiency at high rpm is just not good. We have a single cylinder engine to test the effect of the cam profile and throttle less operation. Lots to learn from the research that we did.
As for the supercharger and turbocharger engine, I will categorize these two engine under 'boosted engine' category. The different is that for turbocharger, the exhaust backpressure is slightly higher than the supercharger engine. It requires slightly different cam timing for both turbocharger and supercharger engine. Still, the valve overlap for both turbocharger and supercharger engine are not as high as the NA engine.

AO

 

[mister2]

Azmio,

I dont pretend to have very much experience in this area, I was simply going off a interesting technical document I read from the NACA repository which investigated the effect of increasing overlap (and duration) on a single cylinder engine. This made the point that supercharged engines (ones with Pintake/Pexhaust > 1) make significantly more power with increasing overlap (and duration) due to the fresh charge clearing the cylinder. If I may ask, who do you work for, as you have made some interesting posts?

Zorg,
True, once you reach 1:1 then you can go back to NA-sort of cams, but I am having a little trouble believing that you could keep this ratio high into the RPM range and still make useable power/boost. From what I have seen (in my limited experience, Im the first to admit!) that going to a larger turbine will still result in choked flow, just at a higher RPM than the original. Im basing this on a flow/pressure graph that I saw on the Garrett catalogue of a GT series turbine. Of course, I could be mistaken!

cheers

Nick

 

[Azmio]

Nick,
I have no doubt that the increase in valve overlap would help in increasing the volumetric efficiency of the supercharged engine at high rpm. However, the question was whether a boosted engine needs valve overlap as much as the NA. My answer would be no.
I guess the confusion can be cleared once we make a comparison between NA engine and a supercharged engine. Of course the different in cam timing requirement is clear when we compared between one and another.
As for the supercharged engine, we still need to check the peak revolution of the compressor as well as the peak rpm of the engine. Then only we can get the proper cam timing for it. There are lots of other factors to consider.
Next, to answer your question, I currently work for Yamaha automotive engine division in Japan. My group did a lot of works for Toyota, Ford, Volvo, Fiat, Renault, etc. Before this, I used to work for Sauber Petronas Engineering in Switzerland.

AO

 

[ZoRG]

Ok, but still, what kind of overlap are we looking at at high rpm's 12000 or so? I understand the less overlap needed etc, however I also understand that fairly large durations are needed for these ultra high rpm cars, when they were runnning 16000rpm's surely there must have been quite a bit of overlap?

 

[beaverjones]

"Ok, but still, what kind of overlap are we looking at at high rpm's 12000 or so?"


possibly none-overlap is a single item to think of in an engine package.efficent ports may need very little duration(overlap for that matter) to flow enough to support a given engine size well past 12000rpm.

 

[Retracnic]

The valve timing overlap required to sustain a given RPM in a turbocharged engine is dependent on several factors. Chief among these is the pressure differential between the exhaust and intake manifolds. If we look at a typical OEM setup, we're talking about exhaust manifold pressures in the neighborhood of 2 to 2.5 times greater than intake inlet pressure. Typically, large overlap periods are not advisable when such great pressure differentials are present. At high RPMs the exhaust gas reverts back into the intake track causing severe charge dilution. Eventually this charge dilution will bring a halt to your tachometer's forward progress.

Yet in terms of motorsports, we would not be dealing with undersized and restrictive turbochargers. Using larger turbine housings, correct A/R ratio, and fine tuning of trims we can realize a pressure differential approaching (and sometimes exceeding) 1:1. This allows a lot more freedom in the selection of the camshaft's duration and overlap.

Theoretically we can use the same cam selection for a turbo vehicle with a 1:1 pressure differential as we can with an NA motor. Yet cams for turbo cars tend to be much milder than their non-turbo counterparts for one practical reason… you'd actually like to spool the turbo before the race was over. Long duration/overlap cams for high RPM operation tend not to produce sufficient exhaust energy at low RPMs to spool the turbo in a timely fashion. Especially when we're talking very large turbines.

So my advice for anyone who wants a high revving turbo motor is to reduce the pressure differential as much as possible with correct turbo selection. Most everything else will fall into place.

Regards,
Bryan Carter

 

[Azmio]

Zorg,
It is hard to say how much valve overlap is needed for your particular application. It depends on a lot of factors. One thing I can suggest is for you to use an engine performance simulation software like GT power or Wave. From there you can get the best theoretical cam timing after quite a long iterations. Later, you need to test the engine with that 'best' cam timing on dyno. Once you have verified that cam timing on dyno, you still need to do some tuning on other parameters.
Anyway, it is still difficult to simply get that magic number for the valve overlap. Differences in exhaust backpressure, exhaust runner lengths, trumpet length, number of cylinders, bore and stroke ratio, number of cylinders, airbox pressure variation, fuel injection timing, engine speed, intake and exhaust ports geometry, ignition timing, fuel injectors position relative to the intake valves, valve numbers, combustion chamber geometry, fuel type, ambient temperature, humidity, etc. will ensure that the most optimum valve overlap is different from one engine to another.
That's why we have to spend months to get the best output from a newly designed race engine. Somehow I understand that you need to start somewhere. If you dont have any performance simulator, just do your best to pick a starting point and iterate the timing, it will take a long time but I am sure you will get there.

AO

 

[dgallup]

The reason the engine speed has changed is CART went from having 3 engine manufactures competing against each other to a spec engine formula where everybody runs the Cosworth engines. Then to reduce cost Cosworth reduced engine RPM's and increased boost pressure. Thus the engines make about the same power but don't need such expensive pistons, rods and valve trains and are very reliable.