posted on July 10, 2012 03:11
External Combustion Rocket Zero-lag Teaser
by Eric Hsu
On the other side of the world is a lush green island with some of the brightest minds and most creative people in the world. They gave the world the Beatles, the universal joint, the Rolling Stones, Cosworth, the Smiths, Aston Martin, the wind tunnel, James Bond, bangers and mash, stainless steel, Depeche Mode, black pudding and of course, fish and chips. England is the little island that's given the world quite a bit. Now (or actualy several years ago in WRC), they give us external combustion anti-lag technology. Well they haven't quite given it to the world yet, but here it is in the first working street driven Subaru STi that belongs to my buddy Ben who's the best engine calibration engineer that I know.
Ben promises more will come soon. I was supposed to make it over to the UK late last year and then once again in May, but life got in the way last year and the Team America GT-R got in the way in May. If you want more in depth coverage on MotoIQ, be sure to let us know in the comments section below so MotoIQ can pay for my trip to the UK. Ben says he'll take me to Nurburgring in the Spec-C (I'm bringing ear plugs) for a super bad ass series of stories. But to answer some questions I'm sure there will be:
- The car is a 2007 Impreza STI Spec C.
- The knob over to the right of the steering wheel is his "cal pot" or calibration potentiometer. Pectel ECUs provide up to 4 switchable maps.
- That's custom display in Caltool, Pectel's calibration software, on an Acer tablet PC.
- When the "zero lag" is activated, the Pectel SQ6M ECU cracks open the throttle to start the zero lag process. That's why the revs go up.
- That's the turbo speed at 130,000+ RPM. The reason why it shows 0.0 at the beginning is because most turbo speed sensors do not read below 10,000 RPM.
- The turbo is an IHI RX-6 from one of the later generation of Prodrive WRC rally Imprezas.
- How does Ben know how to do this stuff? He's a genius that's why! Not only did he develop the latter years of the Prodrive WRC engines (and the Rocket itself), but he was the ex-senior calibration engineer of Cosworth F1. Now he's at another large F1 engine builder. I don't think I'm supposed to really say, but it has a silver star in the logo.
Enjoy, and don't forget to comment! Ben wants to know what you, the intellegent readers of MotoIQ, have to say. Also, let him know that you want to see the Rocket on the Team America GT-R!!!
EDIT: Ben sent me a picture to put up. Elmo gets props for one of the cleanest Subaru engine bays ever. Click to enlarge:
Tuesday, July 10, 2012 5:41 AM
First of all, I want to say that is one of the most awesome-sounding motors I've ever heard, and I usually hate the sound of boxers.
Secondly, I am very curious as to how this works. I've always wondered just how anti-lag systems operate.
Tuesday, July 10, 2012 6:41 AM
There's some pictures of the outside of the thingies on the WRC spec stuff at http://subaruwrcspares.com/11.html In basic principle, there's a valve and plumbing to bypass air around the engine to the rocket, which I suspect looks something like a jet engine combustor inside, albeit modified to be able to pass exhaust gas down the middle without much restriction. From there... engine supplies fuel in the form of rich exhaust, turbo supplies air, voila, jet engine.
Of course, that's just the conceptual form. I rather imagine that doing all this without overspeeding the turbo, while blending the transition from it being propelled by the ALS system to being propelled by the engine smoothly, and all the other details makes it significantly more complicated.
Tuesday, July 10, 2012 7:03 AM
We certainly want to know more and see more pictures. An article about mapping an F1 engine would be cool too :)
Tuesday, July 10, 2012 7:57 AM
would love to know more. im sure there is much more going on when its activated than just the revs building up, because that alone isnt gonna build so much turbo speed.
Tuesday, July 10, 2012 8:19 AM
Watch the car's tach compared with the turbo tach. The first blip up to 5kish RPM is getting the turbo spooled (with the assistance of whatever) but after that, the RPM is dropped back to 2k RPMish without the turbo RPMs dropping. The engine isn't being the primary source of hot gas to spin the turbo.
Tuesday, July 10, 2012 8:38 AM
http://forums.nasioc.com/forums/showthread.php?t=2064792 A couple pics in this thread; you can see a lot better how the air is allowed to (when the valve is open) bypass the engine and go to the 'rocket' thing right ahead of the turbo. Looking at it, I half suspect the valve is something on the order of a stepper (or something) driven butterfly or barrel valve, albeit built to withstand a lot of heat.
Tuesday, July 10, 2012 9:56 AM
here is a thread on how it works
Tuesday, July 10, 2012 11:05 AM
Love it! I have been developing a vehicle in China (don’t hate it’s a good vehicle) and appreciate the cool video, thank you. I could watch it 100 times, I may be close ;-) CalPot 3 (0, 1, 2, 3) ~5000 RPM, TPS unknown WOT maybe. Then Ben switches to CalPot 2 ~2000 RPM, TPS unknown but what is very cool is that the Turbo RPM stays constant with Tip-In. The histogram chart tells a good story. Then CalPot 0 ALS Off, don’t want to start melting things. I am very interested in MAP, EGT and TPS but I understand, you can’t give it all away. PS Nice work with the Team America GT-R, looks awesome.
Tuesday, July 10, 2012 12:16 PM
Ah, ok it's what I thought it was. Funny thing is refuse trucks actually use a similar combustor to pass emissions though it's used to add heat to the DPF and not to spool a turbo.
I'm curious how similar the mapping strategies are for control function of the devices.
Tuesday, July 10, 2012 12:32 PM
Eric, you're lucky to be friends with Ben. I wish I could pick his brain for a few days:) It would be awesome to see this used on the R32.
Micah, most cars (in the last 4-5 years) have similar systems as well for cold start cat light off.
Tuesday, July 10, 2012 12:38 PM
I think the cost involved would be pretty large for the specific valve, turbo and exhaust involved with this system. The components for an S11-S14 rocket are a probably a minimum of $15,000. The IHI are NLA and a lot of existing ones are out rebuilds.
Tuesday, July 10, 2012 1:41 PM
BTW, that WRC car is using the same Murray clamps Project G20 Racecar is using!
Tuesday, July 10, 2012 1:53 PM
Augh, comment lost. Trying again.
I still want to replicate something like this, in between all of my other projects. I mean, the RX6 is a nice piece but I don't see any reason it would have to be that... the valve can't be that complicated (aside from getting it to work reliably, flow well enough, and actuate fast enough to prevent turbo overspeed) and I'm somewhat sure I know what the rocket is for and think I may have narrowed it down to about half a dozen guesses as to general configurations internally. I mean, then it's just tuning! How hard could it be?
(I realize the answer is probably somewhere in the range from "very hard" to "not doable with my resources" but I enjoy tinkering, and it's easier having the head start of knowing it's possible)
Tuesday, July 10, 2012 2:18 PM
I'd love to see some more in depth articles on the and some F1 toys/tech... 'Ring coverage is always a good thing... maybe you can meet up with the guy from my second favorite blog (http://bridgetogantry.com/) and do something crazy...
Tuesday, July 10, 2012 2:42 PM
Bill, I mean they actually have a combustor and inject charged air into a fat mixture to light off diesel in the exhaust. The combustor does it's just combusting and thus we get heat.
What cars are using this now? I was completely unaware OEM's were using similar units for cars (are you meaning diesel or gas?).
I agree, quite costly to produce something like this that will actually last in a performance application.
Tuesday, July 10, 2012 3:13 PM
So, my questions are as follows:
1) Do Pectel ECU's come with that "cal pot"? I want to run an SQ6, and obviously would like to use a switch that's designed to work with it for map switching. Either way, I've looked at just about every single accessible Cosworth product page, and said switch/potentiometer doesn't appear to be available on its own.
2) Can you run such an anti-lag strategy on a standard SQ6, or does it need to be the "M" version? I'm aware the SQ6M probably has some features not originally activated on the standard SQ6, but my street/track Impreza 2 door probably doesn't need most of the more motorsport-oriented features of the SQ6M (for example, I don't need to run waterproof, military-grade connectors, and I don't even want traction control).
3) With the right equipment, could an alternative turbo be used, or is the RX-6 pretty much the only choice? I'd love to see such a device used with a single scroll EFR turbo, though I suspect said turbo wouldn't necessarily be capable of handling the heat generated by basically having a secondary combustion chamber literally right before the turbine; at least not while expecting any sort of longevity, though I certainly wouldn't be using anti-lag terribly often on the streets if I had such a facility.
4) With access to the right materials, could a small shop or highly proficient enthusiast build a "rocket" into their own exhaust manifold designs that works roughly as well as the manifold that they'd have to pay THOUSANDS to buy from certain companies that sell used WRC parts? It's all well and good that Ben is a genius (and yes, he is), but some of us "regular joes" may still have Boeing engineers as friends, or be Boeing engineers themselves, and may or may not have friends that weld/fabricate for a living. Obviously the computerized bypass valve might be difficult to procure, but I see no reason why those willing to put in the work in building their own shouldn't be able to, unless it's a legal thing, I guess.
Regardless of whether or not those questions are even answered, Ben's car is sick as hell, and Eric... all I have to say is keep up the damned good work. I can't wait to see the Team America R32 tearing it up! Also, is this the same Ben who's friend Elmo built that CRX you featured on here a while ago?
Tuesday, July 10, 2012 7:13 PM
@kenku, you can run just about any turbo, some will just live longer than others. Pretty much every WRC car uses Garrett Motorsports turbos along with many of the Rally Cross cars. Some of the Rally Cross cars make due with common Garrett aftermarket turbos.
Tuesday, July 10, 2012 7:19 PM
*This is just as I understand the operation from the pictures and description from NASIOC and the above post, so I am probably wrong*
So, what do we need to make this thing work?
-Air, the engine rpm is raised to to make the turbo flow enough useful air for the rocket to fiddle with.
-The leg from the charge pipe (pre-intercooler) feeds the inlet of the rocket, which inside I can only imagine an impeller blade much like a jet engine, but an impeller must be rotated by another force..
-What about that long silver tube with an electrical harness running to it on the "front" of the rocket. Could that possibly house some sort of electric motor? For sake of argument and to help me picture this, let's say yes.
-but we still need fuel from somewhere between the turbo compressor and the rocket inlet. Out of all of the pictures I saw I could never make out fuel lines of any sort, but we need fuel.. Let's say it's injected at the rocket, post (rocket) impeller.
-So we have everything we need, but how are we spinning the turbo?
1. Raise the engine RPM to shorten "launch time"
2. Begin injecting fuel into the rocket
3. Turbo RPM coming up
4. Start "rocket impeller" via electric motor (from prior argument)
5. Turbo is now capable of sustaining its self, close throttle body enough to allow charge piping to become filled with compressed air.
6. Turbo RPM builds, more air is flowing through rocket, being compressed and ignited, providing synthetic exhaust flow.
7. Turbo is now completely self-sustaining, rocket is providing exhaust energy through internal combustion principle while preventing back-flow through the charge piping
8. Turbo RPM limited by either ecu controlled wastegate or by simply the flow capability of the rocket.
9. Charge piping is filled with compressed air, engine unloaded.
10. Ready to launch?
Is this anywhere near close, or is there something else at play here?
Tuesday, July 10, 2012 7:45 PM
@Brap: Keeping in mind that I'm not an expert... no.
Think of a jet engine. It takes in air, compresses it, adds fuel, burns it and feeds it to a turbine connected to the compressor. The turbocharger has the turbine and compressor portion of this; that just leaves us with the middle parts; adding fuel and burning it. Well, the turbine is also fed by the engine exhaust, so you can just richen that up and voila, fuel for the 'rocket' to burn. No need for a separate impeller.
Using your idea of explaining the order of doing things...
1. Raise engine RPM to get turbo RPM up enough that it's moving air
2. Open valve to rocket, bypassing air to it; at the same time, richen engine mixture so the rocket has fuel to burn
3. Keep minimum engine RPM up high enough so that there's enough rich exhaust for the rocket to burn
4. Do whatever you want with the rest of the engine, continuously varying fuel over-richness and the valve to the rocket so as to both not overspeed the turbo, and have enough air actually going through the charge piping that the engine is seeing boost.
Much simpler. The only trick is, the combustor (rocket) design has much much different characteristics from any normal jet engine combustor.
Tuesday, July 10, 2012 10:45 PM
Definitely want to see the Rocket on the Team America GTR... an article about turbo compounding and MGU or a write up on hot and cold blowing diffusers when exhaust driven diffusers were legal would be great too
Tuesday, July 10, 2012 10:53 PM
Micah, that's pretty crazy, I was just referring to air pumps injecting fresh air pre cat with added fuel. Ive heard regenerations happen on newer large diesels, do these have combustors?
In regards to cost, you could probably replicate this general idea for next to nothing but making it last more than a few seconds in a Motorsport application would be the expensive part. I can't imagine the development costs of tuning, probably a number of turbos etc.
When I first heard a rocket aided impreza at the start of a stage, it blew my mind. There wasn't much information about it at the time in 03, but a friend had mentioned the "rocket" that gave 2.5 bar pressure from 2000 rpm. Nick Mann had done it previously with a bell emergency start turbine in his BDT powered hill climb car but Prodrive(Ben) made it possible with minimal parts/weight/space. Awesome!
Tuesday, July 10, 2012 11:04 PM
Good that is has filtered down to road cars all be it that the RX6 still sells for about 9k pound and the rocket hardware for about that as well , but it is a good sign of the times.
Tuesday, July 10, 2012 11:06 PM
Also good to see you can get an ECU that can control it that does not cost a kidney .
Wednesday, July 11, 2012 1:07 AM
I would really like to see more of this!
Why no car launch :(
Wednesday, July 11, 2012 2:16 AM
Very interesting. I would love to see if this is incorporated into the Team America GT-R.
Wednesday, July 11, 2012 3:09 AM
As far as I get it, there's no fuel burned IN the exhaust, and the rocket itself is nothing more than a fancy name for a valve.
I think I figured out how it works.
He just revs the engine to get the turbo up to speed, so it can move some air into an external combustion chamber, called the combustor. When the throttle get's lifted, the Rocket-Valve opens and since the air is now trapped between the compressor and throttle blade it has only one way to go. Before it gets fed through the rocket into the exhaust manifold, this air gets partly burned off in the combustor. When enough energy gets released that way, it will spool the turbo, which then can self sustain this cycle. At this point you could actually shut off the motor and the turbo would still keep spinning (but kill itself in no time due to lack of oil supply). When the throttle is pressed again, the fuel supply to the combustor is shut off, and normal engine operation takes over.
The combustor consists out of an inner tube (the flametube) with some organized holes around it's circumference and an outer tube. The air that is fed into the combustor gets divided into 2 air streams. One part is directed into the flametube, where it is mixed with fuel and ignited by a spark. The second air stream moves between the outer tube and the flametube and get's diluted via the holes into the flametube where the air/fuel mixture is constantly burning. At the same time the air that is not burned cools the combustor & flametube down. So you end up with an extremely oxygen rich exhaust gas and an EGT of around 600°C celcius, which isn't even hot enough to kill a VTG turbo.
When you want to know more details about it, just have a look at how many people have build home-built gas turbines out of old turbos. Also there's a very good tech group on yahoo-Forums dedicated to DIY Gas turbines.
The old way of doing it in the Turbo Formula 1 and Group B era, was to feed fresh air directly into the exhaust on overrun. There it would get mixed with either fuel from the normal injectors or extra fuel injectors were placed in the exhaust. The heat from the exhaust was enough to ignite this explosive mixture and spool the turbo.
The disadvantages of that methad is that the combustion is by far not as stable as if a (good designed) combustor is used, and since almost all air has to be burned of in order to get a burnable air/fuel ratio, the resulting exhaust gas is super hot, killing manifolds and turbos in no time. Another negative is the waste of fuel to keep the turbo spinning, but that's the case with both methods.
With a sophistocated ECU-controlled system like the one in the video, I'm quite sure that it is also operating at small throttle openings. Balancing turbo speed, backpressure and boost pressure to get the most out of it.
I thought a lot about this stuff lately and I'm actually building a simplified version of this for my drift car right now, but instead of a "rocket" I will use a wastegate to feed the air into the exhaust.
If that works out, the only limiting factor would be the strength of the engine. In theory it would be possible to max out the turbo just above idle and hold that power level all the way to redline. Just like an electric motor.
Wednesday, July 11, 2012 3:22 AM
I just added a picture of Ben's engine bay. It's pretty damn clean.
Scott: some answers for you, 1) no, any potentiometer with detents can be used and must be configured in the Pectel ECU, 2) a standard SQ6 will run such an anti-lag system, 3) yes any other turbo can be used, but an RX6 is motorsport grade right out of the box, 4) it's all in the development and R&D. Of course a small shop could build one, but do they know how to build one?, 5) yes, this is Elmo's work once again. Notice how beautiful the bends on that inlet pipe are.
Logic: A SQ6M probably costs half a kidney. The standard SQ6 does not however, but is about Motec M600 priced (with a whole lot more features if you know how to use one).
Back to the Team America GT-R dry sump pan #2 machining now....
Wednesday, July 11, 2012 3:35 AM
Also, for you guys thinking of trying to do this yourself, make sure you use an ECU with a turbo speed control strategy. If your combustor works well, it'll fry the shit out of your turbo without turbo speed control.
Wednesday, July 11, 2012 8:23 AM
Thanks Eric for this great article.
I once had the privilege of having some of the engineers at Prodrive showing me and a few others here in Norway how some of the stuff around the WRC cars work.
They where supposed to show us the engine/drivetrain management system software, but there where some difficulties so we never got there. BTW, I think it was running EMS hardware by McLarren?
The Impreza was a few years old, don't remember exact, but it was the last model with the "super high tech" differentials.
Her is a clip with me in the passanger seat, and Henning Solberg in the driver seat. This was INSANE, and I can assure that the ALS works..lol. The thing that amazed me the most was the entry speed (going in to corners on snow), and the way he used the throttle/brake mid corner.
Wednesday, July 11, 2012 8:51 AM
Nice! That looks like a blast. Looking at the body shape, unless they did a lot of tinkering, that's going to be after the active front and rear differentials got banned, but before the current rules set where there's not an active center differential anymore either. Funny thing is I've heard rumors that the Impreza lost its edge in WRC because the active front and rear diffs were banned; certainly the results seem to have gotten worse at about the same time but... well, rumors.
You have to love the tech that showed up in WRC... I have to kind of wonder what's going to happen to the knowledge of it though. After the stuff gets banned, how long before the knowledge of how to set it up and tune it goes away? Or am I overstating the issue?
Wednesday, July 11, 2012 10:30 AM
That is not true Kenku, the last couple subarus (S11-S12B) came on sachs dampers which offered no feedback tot he driver . The S14's geometry was all wron in the beginning then they did an upgrade with the subframes and made it a bit better (same car with gronholm leading portugal) but it was not good enough , too much anti dive designed into it amoungst other things .
Wednesday, July 11, 2012 10:49 AM
Will the 'rocket' make Team America quicker with it's anti-lag?.....Um yes, let's put one on there and have a back up handy! U-S-A....U-S-A
PS I would've been more excited if the video had a launch off the line!
Wednesday, July 11, 2012 10:56 AM
Well, it was just a rumor, and I never claimed it was the whole story. If the Sachs dampers were introduced with the S11, look at the results over the season for 2005 (S11) vs. 2006 (S12 with active front and rear diffs banned) for Solberg. I find the contrast interesting anyway. In reality of course, who knows; obviously there's a lot of factors.
At any rate, I still want a fully active car.
Wednesday, July 11, 2012 11:03 AM
Oh yes they were a lot of factors , believe you me.
Wednesday, July 11, 2012 11:34 AM
Thanks for the interesting post, technical info like this are not easily found anywhere else. I´m really looking forward to seeing more in-depth articles.
I would love to see the rocket system in the Team America GT-R also. Would probably be one of the first GT-Rs with real WRC type anti-lag?
Wednesday, July 11, 2012 1:15 PM
Bill, yes some of the trucks come equipped with them. The only ones I know of for certain are limited to refuse trucks by the OEM that I use to consult for. They use to keep heat up where the truck otherwise doesn't generate enough while running a trash route for instance. Regular trucks didn't need the burner system.
When I was talking expense, I meant the whole package of developing a motorsports ready package. The materials, fab and machining for the system as well as the development, tuning, and destruction (dare I say learning curve) of the system....so yeah, expensive lol. Back in FSAE we'd played with making turbos in to 'rockets' just being bored engineers. That and supercharging a briggs, putting it on a go kart frame with a Lazy Boy recliner lol
As for Eric, I've taken you advice and have gotten out of the powertrain consulting and am now running my company as a job and no longer a hobby. Think there's any chance you guys will have this at Buttonwillow? Have you asked Nads for a waver yet? I'm making the cross country trip and would love to see/compete with the car :)
Wednesday, July 11, 2012 1:23 PM
Eric , could you send me some info on the SQ6 and SQ6M?
Wednesday, July 11, 2012 4:13 PM
Slightly OT, but aren't there oiling issues with the turbo angled like that? I thought I read somewhere on the interwebs (so it must be true!) you don't want the axial angle above 20deg or something.
Wednesday, July 11, 2012 4:41 PM
No oiling issues for the RX 6 and the WRC setup .
Here are couple pics of the engine out with me working on it :D
Photo of the rocket plumbing
Wednesday, July 11, 2012 6:41 PM
Wow! Awsome stuff!
I guess I should have asked how do you take care of the oiling? Is it significantly different than a run of the mill turbo?
Friday, July 13, 2012 1:56 AM
@czubaka, IHI uses a pretty elaborate sealing system to allow the turbo to be at a very steep angle and not have oil leakage issues.
Re: diesels and a combustor in the exhaust. There are a few applications where it is necessary to keep the DPF and SCR temps up to allow for proper operation and emissions reduction; the catalysts have to be hot enough to work. So yeah, they basically dump diesel into the exhaust and burn it to generate the heat.
Friday, July 13, 2012 12:52 PM
@ Micah and spdracerut , What diesels use this combustor? Ive never seen that and Id like to. Youre talking about something beyond just the doser correct?
Saturday, July 21, 2012 8:57 AM
In the enginebay pic- What's that just before the throttle on the boost pipe? Looks like a MAF sensor to me?
If so, that would explain a few things.
Thursday, November 22, 2012 3:33 PM
How is this different than Porsche's implementation of Hyperbar in the early 80s?
MotoIQ Proudly Presents Our Partners: