Forced Induction Efficiency Random Musings..

I started typing a question in to this thread here...

https://www.tundras.com/threads/magnuson-magnum-2650.128709

But ended up writing half a novel and decided to put it in it's own thread rather than create a wall of text nobody wanted in that thread.. So please feel free to critique what I've written. It was supposed to be question but as I thought through it, it turned in to to something else...

So, out curiosity (and I'm being serious here) how would one know that their SC is more efficient than another SC? I would think that ultimately the only way to tell would be fuel consumption. I mean, boost is measure of restriction, so you would need to be comparing Manifold Absolute Pressure and Airflow across two systems to really see a difference there. But we often cite power gains from forced induction as HP@PSI; the boost pressure is referenced and compared to atmospheric pressure to make a rough guesstimate at efficiency of a system to determine whether or not that boost is useful air or wasted pressure. To clarify that statement, reference atmosphere is 14.7 psi or 1 atmosphere, but we only refer to boost as pressure beyond 1 ATM. If a NA motor makes 400 HP with 0 psi of boost, than assuming similar efficiencies in the system, one would hope that at 14.7 psi of boost (that's one ADDITIONAL atmosphere of pressure) the motor should make 800 HP. At 29.4 psi it should make 1200 HP, so on and so forth. But that's not how reality works and I have typically seen (from magazine articles back when that was thing, posts, and dyno's of other vehicles that are not mine) that higher boost levels exaggerate inefficiencies. I'm supposing that those inefficiencies come in the form of heat and flow restrictions, but that's just a supposition. There's also the fact that the difference between No Boost/Mid Boost/Max Boost is quite high in a high boost system, which means the motor has to 'expend' more energy to close that gap and make more boost overall. Once that boost is made, efficiency can increase relative to no boost/mild boost - kind of like 6th gear benefits from all of the gears before it, but until your row throw the first few gears, 6th gear would be horribly inefficient to drive around in trying to get up to speed. Give the bolt on SC's for the tundra is a relatively low boost system; for comparison, I think the Hellcat and Hellcat redeye run around 11PSI and 14 PSI, respectively, and most turbo motors push 17-20+ PSI. I think some light duty diesels have run up to 40 psi from the factory.

Anyways, using that line of thinking, we can try to compare power gains boost for boost across different systems, then compare the added HP/psi to the NA HP/psi. Continuing with the 400 HP NA example above, that makes for 27.21HP / PSI of atmospheric pressure, so at a given boost level we should be able to guesstimate a HP level. The difference between mathematically derived HP and measured HP should accounts for inefficiencies in a system expressed as either a restriction to flow or a parasitic loss. If at 7.5 PSI of boost we are making 550 HP, but the math says we should be making (7.5*27.21) + 400 = 604 HP, that's a difference of 54 HP!! So the system is losing 54 HP worth of efficiency somewhere. In the case of a supercharger, we know that it takes HP to drive the supercharger itself; in this case, 54 HP. I suspect that under driving the SC with a smaller pulley will create more HP overall, but also increase the parasitic loss - as well as the percentage of parasitic loss. I would be love for someone line @snivilous or @reywcms to share some data a stock runs vs different pulley runs. I'm curious...

Furthermore, if the TVS 1900 makes 550 HP at 7.5 psi and the TVS 2650 makes the same 550 HP at 7 psi, one could say that the 2650 is somewhere close 6.5% more efficient than the TVS1900 judging by the numbers at face value. That small of a difference in boost pressure could be due to several different factors, but a 6.5% gain in efficiency appears relevant. Alternately, in the case of what I believe to be published numbers by Magnuson, we are only getting 13 more HP at the same 7 psi of boost (note that I'm using 7 psi for the tvs1900 and not 7.5 psi for this example only), which makes for a much more paltry ~2% gain in efficiency. But to note, at those boost levels, the system should be able to maintain a fair amount of efficiency.

However, overall efficiency is much lower than that. We were expecting to see an additional 204 HP on 7.5 psi and only saw an additional 150 HP. We lost a little over 26% of the power, so the supercharger is only 74% efficient at adding power compared to the stock HP/PSI. If we are able to make the same power at 7 PSI, that changes. We are expecting to see 27.21 hp/psi * 7 psi = 190.5 HP + 400 = 590.5 HP. That makes for a 40.5 HP difference, ~21% power loss, or a supercharger that is 79% efficient. That makes the TVS2650 5% more efficient than the TVS1900 at that power level. I'd be interested to see at what point that efficiency starts to diverge (ultimately, they will both fall like a rock when trying to create to much boost/RPM with the system) as well as how closely the follow each other both in trend and empirical value.

Now lets look at the overall system efficiency. 550 hp / 604 HP = 91.1% efficient compared to stock. 550 HP / 590.5 HP = 93.1% efficient compared to stock.

Could we determine efficiency due to the constantly varying loads cycles of real world driving? Were the motor on a steady state generator tasked with a constant load, operating at a steady RPM, I think we could. And the easiest way, IMO, to measure that difference would be fuel consumption which should be steady over time. But that's not how anybody I know drives their vehicle. But I also don't know many people that drive their vehicle at redline or even at peak torque all of the time (yes, I do know a few...). So I think the actual efficiency difference would be very hard to determine without using the above model combined with somebody that drives essentially the same route of conditions, with one system at a time installed on the same vehicle. And, at that, fuel consumption would really be the best way to determine the difference.

Now to throw a wrench in to the mix, lets look at what the non-intercooled turbo setup that @TurboKits.com is working on can do. The test tundra made 310 HP to the wheels IIRC, and 409 HP on 3.5 psi. Stock is 21.1 HP/psi. 3.5 psi of boost should add 73.8 HP to make 383.8 hp... but they made 409 hp, an additional 91 HP... Hmm... That means the turbo charger was 123% efficient, or it increased the overall system efficiency by 6.7% to 106.7% compared to stock. If this is to be believed, I suspect it is due to the use of water-methanol injection to lower the IAT's below what a water-cooled intercooler is capable of alone. But I think there is something else going on.

Alternately, maybe I did my math incorrectly and was supposed to compare crank HP numbers instead of wheel HP numbers. So let's try that instead. If we assume that the motor made 381 HP to the crank and it made 310 HP to the wheels, we have a 71 HP in drivetrain losses. Assuming the 381 HP crank number, that gives us 27.21 HP/psi. At 3.5 psi, it should make an additional 95.2 HP or 476 HP at the crank. So lets subtract the drivetrain losses of 71 HP and we get 405 HP as an expected HP number. 405 hp.... 409 hp... 405 hp.... 409 hp.... I'd say that a 1% discrepancy is quite possibly within the margin of error. This would place the turbo charged system in the realm of matching efficiency of the stock motor. Considering we didn't have to use HP to turn a set of rotors or twin screws as we knew we would need to do with a SC, this would show a well built, efficient system for this boost/power level. This would also show how little parasitic loss there is for a turbo system operating at this low boost level. If that efficiency level holds true, at the same 7 psi of boost as the SC's above, it should be capable of right around 600 HP without any additional stress on the motor as a 550 HP supercharger system. Torque increase should be even better as you're not losing that torque turning the SC (remember, HP is just torque at RPM, so a SC robs torque EVERYWHERE, where a turbo adds torque in a specified range usually starting below peak torque of the NA motor).

But it's not all roses and unicorn farts. There are tradeoffs for efficiency - just ask all of the prius owners if they like driving around in an egg-shaped automobile... Ok that was a fake jab a family member who will never read this.. but, just the same, there are tradeoffs. The SC system is going to deliver predictable, linear power at ALL throttle positions while giving up efficiency compared to a stock or turbo'd system.

 

One must also realize that turbines make use of a force that is ordinarily wasted, a bonus to overall efficiency.

 

"I would be love for someone line @snivilous or @reywcms to share some data a stock runs vs different pulley runs."

I'll be your huckleberry...

here is stage 1 untuned (85mm) / stage 1 tuned(85mm) / stage 2 tuned (80mm + 650 inj's).

stock hp at wheels was 282hp.

 

this post lacked the usual @hagrid smarm I have come to love and respect. please do better.

 

Upgrades are pending.

 

I thought DAP’s back to back testing of the 1900 magnuson vs the 2650 Harrop in the same super hot day, same truck, and same tuner is the best representation of what better efficiency and charge cooling can do.


As for the broader theme of this post, most of which I lost concentration trying to read (not a dig, just highlighting my low attention span at this moment) I think this is a great video about forced induction methods in general despite it being 7 years old so some product lines have of course changed. https://m.youtube.com/watch?v=-UzDP....com/&source_ve_path=Mjg2NjY&feature=emb_logo

 

Now...can we ponder the issue of the Mag being upside down?

and the inherent inefficiency of a complex airflow compared to the straight shot of the harrop?

 

Further... this thread examines the mechanical means to achieve one goal: to increase the amount of oxygen molecules present within the zylinder.

What if we had a supply of pure oxygen handy? Could we not administer a steady yet measured volume of such that we achieve the same concentration as compressed air? The working fluid will still be present and in abundance.

They have those cryogenic LOX vessels now... a carboy? 50 litres including an evaporator?

 

That's called nitrous though I'd imagine LOX would work but probably suffer the same atomization issues as a rocket engine has, where as nitrous is much easier to work with in that regard.

 

Lots good info out there from Gale banks and this dude Richard:

https://www.youtube.com/live/AvJ1nLRYteE?feature=share

 

Thanks for the replies to ramblings at 3 am. I'll pick through them as I can.

 

That's awesome! Any chance you would have the correlating boost pressures to go with this runs? PSI at peak power is what I'm after.

 

Thanks for providing that info. DAP's results are along the lines of what I was thinking, but I think his conclusion is skewed. He appears to be running the 80 mm pulley on the Harrop for all of his runs whereas I though the 85mm was stock (as is noted in @nobodyintexas Dyno charts). He doesn't note what pulley he is running on the Maggie, nor does he have equivalent boost runs to the Harrop - just a higher boost run. Expecting to see less efficiency from high PSI on the smaller blower, it might be at the point where it's creating more heat being way out of it's efficiency band and is killing power. For the sake of data, I'd like to see an actual boost for boost comparison. But I also didn't fork out the moneys for the parts, install, R&D and dyno runs. So I can't complain.

Either way, it does illustrate the ability of the Harrop to handle the heat better and, in turn, make better better power on a hot day.

 

it was right @ 6lbs on the 85mm

just north of 8psi on the 80mm. I believe Alex said it was 8.4psi max.

 

Actually, each Harrop run is a different pulley. The magnuson had the currently smallest available pulley at 2.30” installed. Harrop run 1 at 5psi was the 85mm, run 2 at 7psi was the 80mm, and run 3 at 8.8 psi was the 75mm. He was getting all of the baseline data for Harrop on each of the pulleys during this testing as shown in this chart distributed by Harrop.

 

Sadly I don't have any good data to add, but I will sprinkle in what I have for shits and giggles.

First being that from my testing boost and pulley size correlate pretty well to a linear change. Take that for what you want, but increased rpm on the 1.9L doesn't show a curve in either direction for max boost.



A stock truck we know makes around 300whp/14.7psi = ~20.4whp/psi, and my truck the one time I dyno'd it was 490whp/24.7psi = 19.8whp/psi which is actually extremely correlated as well (I thought it wasn't until punching the numbers). And the 300whp and the 24.7psi are rough estimates, plus my engine has nearly 300k on it, so being within 3% is actually kind of insane for maintaining specific performance.

I sadly do not have dyno data from the lower boost pulley, nor do I have any data for a stock truck in general. But I do have Dragy timed runs of 0-60mph:

19.2psia = 5.58s (2.3" pulley)
22.7psia = 5.33s (2.3" pulley)
22.7psia = 5.26s (2.0" pulley)
25.3psia = 4.58s (2.0" pulley)

What is interesting with those numbers is running the smaller pulley at higher elevation resulted in nearly identical timing as the larger pulley at lower elevation (with the same overall boost amount) despite a 15% change in blower rpm. However (and this is really stretching the analysis), the change in timing vs change in psi went from -0.25s @ +3.5psia to -0.75s @ +2.6psia. I think that's really reaching, but one could maybe suggest the power under the curve not being a linear increase despite peak horsepower POSSIBLY being a linear increase.

All of this is a bit of a stretch though and I wouldn't even call any of it scientific, but fun to throw the numbers around. It would be cool since I have four different size pulleys available to me, to do dyno runs and plot the data as the boost/effective displacement goes up. I do think it's very interesting that at the extreme of my smallest pulley, the whp/psia matches well within margin of error.

 

Word count looks good.

The whole premise of the post could be condensed to PR x NA HP = Expected/Predicted HP #s vs Dyno Actual HP #s given a PSI to estimate what you are calling "overall system efficiency" if that is what you are after.

Most of us without meth or on unmodded engines are running fat AFRs on the big end to keep the pistons and rings happy which could account for some of the additional unexpected losses you are seeing in your calculations.

 

I’d love to know the gains if we didn’t have to run these things at 10.0 AFR with boost.

 

Intolerance!

what if I lean out my afr and lean out myself?

Maybe fold in the mirrors?

it all matters.

 

Just looking at ideal Stoich and no other numbers or factors it is a loss of up to ~30% potential. Real gains would be smaller as we couldn't run that without risk. Big numbers if the engine could take it which we know it can't.

 

You set a high standard as the king of concise communication.

 

<hat tip>

truth be told…I’m was/ am prepping for an upcoming board meeting and cruising Tundras.com simultaneously

hence all my activity today

I dread poo-bah meetings.

my financial reviews are just as concise!

 

havest thou ever stood up before the board and declared, "regarding our financials, we are truly fuckered"?