Forced Induction Modeling in latest DynoSim5 Software

[DreadFox]

Hey everyone,

I've been lurking here at Speedtalk for more than awhile now, and have learned alot from the heads here. I posted this over in another forum with zero responses so far, and I've searched here the best I could but really couldn't find a definitive answer. Background: I've been using Performance Trends software since about 1998 to model various builds but all of those were either naturally aspirated or nitrous setups. This is my first turbo model. I'm modeling a fairly mild SBF 363 11:1 SCR, Weingartner ported RHS 215 heads, TFS R-series Intake with 75mm opening and 75mm TB, hydraulic roller cam 244/252@.050 114 LSA .375"/.382" at the lobe, T4 flange 7875 turbo on at least E85 with boost levels from 10-21 psi. This is a street build that will see some track grudge style racing from time to time. I'm trying to determine if the turbine side will choke down the engine too much with this camshaft so I've been running the camshaft iterator but I'm still unsure about the overall model. I know DynoSim will generate insane unrealistic data if you give it the opportunity...

Anybody here using the Performance Trends Modeling software to simulate their turbo build? Just curious as to how close the model was to your actual performance once the build was complete. And if it wasn't then why was it off, best guess? The numbers look a bit too good to be true on my build using the estimated ignition timing setting(BMEP in the 500psi range at lower RPM....that would destroy my junk for sure) I have two questions for anyone who has modeled their turbocharged setup.

Question 1 is what is the timing strategy you're using in the software? I unchecked the estimated box, and went with base timing at 12, and advance 2.5 degrees per 1000RPM up to 4000RPM. I still don't feel like this is an accurate timing curve being that my timing curve from my previous turbo build was not that simple. So I've resorted to no advance and just locking the timing out at an arbitrary 16 degrees for all my boost level models as I feel this would be the most conservative and realistic route. Even though the software says I have a chamber timing requirement of 28 degrees I feel that's being a bit ambitious even if I am on E85 but I've never had this new build up and running so maybe it would like 33 degrees at 21 psi...once. I don't know. I've heard E85 is not detonation limited so I guess I'll be able to throw max timing at it but 32-34 degrees at 21 psi seems scary to me being that I never ran more than 24 degrees at 12 psi on my previous setup but that was on spiked pump fuel.

Question 2 is what is the turbine size diameter asking for, the exducer or the inducer? The turbine size diameter section in the manual doesn't specify whether it's exducer or inducer on the turbine. Granted, it doesn't appear either makes any difference compared to the other. I got the same peak numbers at the same RPMs whether I used the exducer or inducer measurement. But it does seem to make a difference when running the cam timing iterator...

Any thoughts would be greatly appreciated.

[David Redszus]

I would seriously consider the use of Vannik's 4t engine simulation program. It provides a worthwhile insight to turbo performance.

A comparison to Performance Trends results would be most interesting.

[mk e]

I used to use DynoSim then switched to Dynomation5...which has DynoSim built in as the Filling/Emptying model but also has the more sophisticated wave model that DynoSim doesn't have. The timing is about right for NA engines on the ones I've done, but it doesn't adjust timing boost so you're on your own with that based on your experience.....not sure you need to pull it back like you are to get an accurate output estimate though, I never did. The couple boosted engines I've modeled then actually built were with blowers not turbos but but the FE (DynoSim) result was low not high.....but I wasn't pushing so much boost that just getting it to not detonate was a problem.... 9 psi with a roots straight in, and 24 psi with a screw type though an plenum mounted cooler, both on a little 3.0L V8 and I left timing at the NA 32 degrees in the sim, but they were 25 and 21 respectively in reality once I got them tuned IIRC. Recent (like in the past year or so) updates to dynomation and I'm guessing DynoSim have got the FE models WAY closer to what the WAVE model out is for a well tuned engine were it used consistently to be 15% low on my stuff which is all OVH, no idea on pushrod stuff.

I'm pretty sure the "turbine A/R" is for the turbine or exhaust side and that is spelled out pretty clearly int eh Dynomation manual so I assume it also applies to DynoSim

I've never been very happy with the cam iterator or the 10pt lode estimation.....using real lobe profiles makes a big difference in getting accurate results.

[DreadFox]

David, yeah I'd like to try Vannik's software but I bought into the Performance Trends stuff before I knew about Vannik. I don't build enough engines to really justify another $400 in software at this point. I'm just a hobbyist. If I can't get this to look like the real thing when everything is done then I'll consider an upgrade.

[DreadFox]

Thanks for your reply Mark. Okay, that all makes sense. I would start conservative anyway on the timing once the engine is together. I've looked at Dynomation as well, and I really wish I could justify a more robust software package. But the info you provided is good to know. Really I'm happy if it's on the conservative side in the model, I'd prefer that over overly optimistic numbers for sure. It's just the BMEPs in the 350-500psi range from 2000-4000 RPM made me question the camshaft selection as the turbo model example doesn't have numbers that high. The manual only mentions BMEP for nitrous but I know it applies to everything based on my engine math knowledge. Also if I let the software to an automatic camshaft, it results in 200-350psi BMEP over the previously mentioned RPM range. Yeah turbine A/R is straightforward, but the turbine diameter I was overthinking. I just measured the outside diameter of my wheel and plugged that in instead of using the given turbine exducer measurement. Either or didn't change much though.

Ultimately, I'm trying to figure out if this camshaft is right for this build which is why I had a question about BMEP and boost.

[mk e]

You have high compression and high boost....that means high BMEP. My 24psi 3.0 was only 8.6CR but opening the file I see 355psi BMEP and it was just fine. Bumping the compression to 11 I see 390 BMEP and that number doesn't scare me...but the 11:1 CR does because that means high combustion temperature and it's mainly temperature that causes detonation not BMEP.

Nitrous works very differently from boost. Boost puts the cylinder under pressure and you get more power directly from the increased pressure. Nitrous does some cooling which puts more air in the cylinder but mainly it increases the % O2 in the air and that makes the combustion temperature go up, which causes more pressure and you get more power....but it's that temperature increase that is the concern because temperature melts stuff and causes detonation. Dynomation/DynoSim don't spit out combustion temperatures, but BMEP on a non boosted engine will give you a good idea ofr where you are and that is why the manual says watch BEMP on nitrous applications....but that doesn't really apply to boosted application because several things could be causing the high BMEP and you may or may not have a high combustion temperature issue. I don't even look at BMEP but do calculate inlet air temp.....because temperature causes detonation and the hotter it starts the hotter it gets during compression.

For the cams keep an eye on the BSFC. That will give you a sense of how much fuel is being pumped out the exhaust and how hot your exhaust is likely to be. If 2 cams make similar hp, go with the one that give better BSFC and if you have know good and bad examples to low at to get a sense for where the limit is even better. The better indicator is probably charge loss to exhaust flow but that is not available in DynoSim I don't think so BSFC will have to do.

[DreadFox]

That makes sense with you explaining it that way. I should just ignore the BMEP on boosted applications and low RPM, and it's what I was hoping was the case. I've considered reducing the compression ratio but I've seen excellent results on 10.5-11.5:1 turbo engines on E85 or methanol on high boost granted these are full-on drag setups. I didn't want to drop the compression ratio too low where the car isn't responsive until it's well into boost. It won't ever see more than 10-15 psi on the street, and will only see pressures above 15 psi at the dragstrip and even then it wouldn't be until well into the run. The car is a basically a full weight Fox body hatch, and is on a 17" rim with a 27" tall drag radial. It does have adjustable upper and lower control arms, 10 way shocks, and an airbag but I foresee a fair amount of wheelspin being an issue without any electronics available. But it will be using the MS3Pro EMS so I have electronic boost control and traction control at my disposal. If those tools don't reign it in then I'll be moving to a 15" rim and 275 radial. I'm trying to keep it sleeper style as much as possible. The previous engine was a SBF 347 with a 10.2:1 SCR and I ran 12 psi with pump fuel and torco accelerator. I never had an issue with detonation but I did start out very conservative on the tune.

I haven't calculated inlet air temp so I'll have to do that. I'll do that tonight.

That's a good point on the camshafts, and I have a new direction to go in now to evaluate this one. I picked it up spur of the moment as it was a good deal, and came out of a car that was making 947 whp albeit it was a supercharged 377. So I'm not tied to it. I might save it for another build depending on how other known profiles respond in this model setup. I've noticed here that many people aren't happy with the cam iterator function in DynoSim so I'll stick with known profiles. Thanks a bunch Mark, I appreciate it.

[SWR]

I would seriously consider the use of Vannik's 4t engine simulation program. It provides a worthwhile insight to turbo performance.

A comparison to Performance Trends results would be most interesting.

Even with all the boxes ticked in EA Pro you will find that any decent build - I.e. less backpressure than boost or anything close to 1:1 ratio - will show too little hp/slow spool at low rpm using as close to correct actual turbine specs you can, then a steeply climbing "rush of power" up to the torque peak and from there on it's not that bad. Using the Compressor Mapping feature correctly helps with the low end accuracy. I added nearly every turbo example file the program has, including every turbo in the then-current 2009/2010 Garrett catalogs.

Any turbo sim in EA Pro using just the "general specs" will have a lot of the scenario shown above, 80's style F1 power delivery. I know, I helped take the turbo part of EA Pro from where it was - unable to simulate say, a 2JZ Supra running anything over a 600 hp turbo - to being just a few hp off known builds using all the bells and whistles of the program. You have to do that, or you'll be off by a factor of GIGO.

Does overestimate the Knock Index on alcohol fuels a bit, while at the same time slightly underestimate the power possible, at least on 4-valve stuff. But on moderate "normal" builds - which I almost never do, so really haven't got much input - it should be fairly close.