falcon

iTrader: (13)

Sorry I didn't blow anything off I must hav ejust missed it.

falcon

iTrader: (13)

Yeah I would have a boost guage on the turbo compressor and one at the manifold.

So Jason you think it would be better to run the turbo into the Rotrex?

Rara

falcon, I'd suggest running the numbers for what you're thinking. Plot your lug lines on the compressor maps using the individual pressure ratios each for the blower and the turbo (is there a comp map available for the rotrex?) and if you aren't bypassing the turbo compressor, regardless of the order (turbo -> rotrex, or rotrex-> turbo) the entire mass flow going into the engine must be considered on both compressor maps. Don't forget to consider the speed lines for the rotor group. This is going to put you well outside the choke line on anything but a turbo that could handle the same performance without the rotrex present.

The only way doing this is going to be any better than a properly sized single turbo alone, is if you run it sequential and not as a 2-stage. Run a smaller turbo, and bypass it to keep the speed and choke flow down. Doing it different than everyone else can be cool, just make sure its somehow at least better than what everyone else is doing too.

JasonC SBB

falcon,

Firstly, what are you trying to achieve, over a rotrex-only, or a turbo-only setup?

falcon

iTrader: (13)

More torque with very fast spool and good top end while not taxing either blower to 20PSI. Compounding two blowers = much better efficiency (from what I understand). And to be different.



Hell I already Rotrex'd a built 1.6L after everyone told me it was a waste of time. I'm all for going against the grain.

JasonC SBB

1) I doubt daisy chaining compressors will net you better overall efficiency

2) "more torque" - at what RPM?

triple88a

iTrader: (2)

I've yet to see a diesel miata. Toss a turbo cummins in there

Whatever you decide dont forget pixs.

sixshooter

iTrader: (12)

I disagree with the fundamentals of your physics and engineering assumptions. A compressor map is based on one atmosphere of ambient pressure. That will not hold true when the Rotrex begins making positive pressure.

The turbo will not overspin for two very good reasons. The dense air the compressor is trying to compress will offer plenty of resistance and the large wastegate will preclude excessive pressure in the exhaust over-driving the turbine. Additionally, it's been done before so the theory has been proven in practice.

See series turbocharging. Here's an engine that uses series turbocharging.

sjmarcy

I'd also consider the tractor pull motor scene. Those crazies compound their stuff all the time. I've heard of them running over 100 psi.

triple88a

iTrader: (2)

Actually there wont be dense air on the compressor side until it hits the supercharger therefore a lot less resistance slowing the turbo. To make it worse, the supercharger is pulling more and more than the turbo lowering the resistance further. The only savior i see is the bigass wastegate.

falcon

iTrader: (13)

There will be a BIG *** wastegate.

I got some info from Abe on possible manifold set ups. It's definatley not going to be an issue from a packaging standpoint. Just need to figure out the proper turbo and which way to feed the air. I've seen it done both ways now. But more often than not the little turbo gets it's air fromt he big turbo (in my case it would be the Rotrex)

falcon

iTrader: (13)

Lol, totally just found a GT2252 on clist locally for dirt cheap. Hmm... now I just need to make sure it will be somewhat close to what I need. I think it will...

triple88a

iTrader: (2)

I've only seen it go big turbo > little turbo > TB.... Never little turbo > big turbo... One could only wonder what will happen when there is a belt.

Rara

Feel free to disagree, it just means you're wrong.

You completely misunderstand the point of correcting a map to standard temperature and pressure. They are done this way so that you have a common point of reference for any way in which you want to use the map. You just have to make sure your engine input assumptions are also corrected back to std temperature and pressure, regardless of what they actually are. This is also at least partially why the Y axis is listed as a pressure ratio, rather than a compressor outlet pressure, so that it can be used when P1C is not ambient.

Did you even read what I wrote? The turbo will overspin if the wastegate is left closed enough to pass the full mass flow through the compressor (instead of bypassing the small turbo compressor at high flow rates as I suggested). If the wastegate is opened enough to prevent overspeed, then the turbo compressor will become a significant restriction to the supercharger at the same high mass flow rates.
And we aren't talking about theory here, we're talking about practical sizing of the system and how to make it work well. He could slap damn near anything he wanted together and it would function to some extent. But if he wants it to work well, some careful thought into the design will go a long way. It's one thing to be different with something that works, its completely different if you're "different" thing sucks compared to the normal way.
Further, you don't see proper series turbocharging on automotive gasoline engines, because gas engines typically can't run at extremely high pressure ratios due to practical considerations around detonation. These pressure ratios can virtually always be achieved with single stage setups (either single, or multiple turbo in parallel or sequential). The diesel applications like you linked below, and mentioned as used in tractor racing type stuff, are capable of running significantly higher pressure ratios, but the mass flow doesn't increase as much. These are prime applications for a series turbo setup, and not even on the same planet compared to the OP's application.

Yes, I'm quite aware of it; a number of my co-workers were responsible for the design of most of that turbo setup. But I only work on gasoline stuff.

JasonC SBB

I have long suspected that this may be true about compressor maps. However, why would they use mass flow rate as one axis, instead of volume flow rate? I tried doing a search to answer this question, to no avail.

JasonC SBB

However, I did find these 2 interesting threads on twincharging - turbo feeds PD s/c:

http://www.eng-tips.com/viewthread.c...=267889&page=1
http://www.eng-tips.com/viewthread.cfm?qid=174690

wayne_curr

iTrader: (8)

Lol, finally your car is going to be quiet!

From Jason's second article:

sixshooter

iTrader: (12)

Compressor maps show a gas volume measurement across the x-axis either expressed as square meters per second or cubic feet per minute:





If the x-axis is scaled in pounds per minute it is called "corrected air flow."



Corrected for what? Standard temperature and pressure (and humidity, etc.). Why? Because though we are listing it in the graph with "pounds" in it's name we are still measuring volume. Air by weight has a specific volume at standard temperature and pressure and that is what is being expressed in the x-axis.


Now take a cubic foot of air and compress it into a space exactly half its volume. Now it takes up half the space it did but still contains all of the same delicious little molecules. Since it is half its original size it fits more neatly through small openings than it did before. In fact, it can fit through openings half as large as it could before with the same number of little molecules. It has the same mass, but is more dense and fits through tight openings more easily- not unlike a supermodel going through an airliner over wing emergency exit versus Chris Farley trying to do it. So the same mass of air can easily fit through a much smaller compressor housing than it could normally.

I don't know why this seems so mysterious to some people.

Yes, this is commonly done with engines where the builders are looking for substantial pressures above ambient where achieving the compression of gasses in one step is inefficient or impractical. But Falcon's aim is specific and this method is sound in design to target both of his goals of a broadened power curve and it being a fun piece of DIY engineering. It certainly sounds like something fun to tinker with.

JasonC SBB

Said supermodel would have to weigh as much as Chris Farley.

sixshooter

iTrader: (12)

You are saying supermodels aren't dense? C'mon...