Misperceptions

[Blue Foam]

Now that my asbestos suit has cooled off (I wear one whenever I come here), I believe it's time to clear up some misperceptions about me and my not so humble opinions.

First, I do not and will not diss Mr. Barrow and his work as a race Merlin builder. Since Dwight Thorn's retirement, we can argue about the relative rank of various active Merlin builders. IM<HO Sparrow has to be placed in the top three and probably higher based on past performance and progress to date.

And THAT is what's missing from this discussion and what some don't seem to understand about my opinions. We know more today than we did yesterday. Therefore, we should expect to be better, smarter, and faster tomorrow than we are today. Without this, the world stops turning and we might as well all lay down and die in our sleep.

I expect more from Sparrow next year BECAUSE he has made progress in the past and I believe he's smart enough to continue to do so. "... a lot of people smarter than me..." Mike? I think you're selling yourself short. You pointed out something that I seriously overlooked in all the tech talk: risk is an independent variable. As an engineer, I have met many innovators and many creative idiots. What separates them are honest and thorough trade studies that take all factors into account. In terms of cost-benefit ratio, EFI and turbocharging seem to be a no brainer. However, if we add risk as an independent variable it gets less attractive. NOT on an emotional level; in a real-life sense, add risk to the cost and it is harder to justify against the benefit.

Now if the benefit was, oh, another 1000 HP without risk of failure, then we should look at spending that 75 to 150k to try it. It isn't. The benefit is theoretically a few hundred HP with reduced temps and a probable reliability increase if properly tuned. Clearly a less attractive trade than I proposed. So, I promise to apply risk as an independent variable in any discussion of technology from now on.

Doing so, what about knock sensors and piston cooling? An independent knock sensor costs $116 for the processor-indicator and $35 for the sensor. Inserted in the head (there's LOTS of convenient places on a Merlin!) the sensor listens for the unique frequency of unstable combustion that is the precursor of detonation. That means, for a few hundred bucks (both heads) a high end Merlin driver could get a warning that would do for detonation what chip detectors do for gearboxes.

My comment on Merlin reliability was not directed at Sparrow. We lost every Gold and a couple of Silver race Merlins this year to a variety of causes (and Sparrow is darned right to point out that systems failures were some of the causes). In case anyone thinks I'm turning somersaults, I will complete the thought I had after the Gold final: "This is the first year that only radials finished the Gold! Wow! That was one boooooring race!"

RRIII burned a piston. It happens (you want to know how many I've burned over the years in how many different engines???). Sparrow probably knows what caused it, or has a list of suspects; I do not for this specific incident. However, piston cooling is often a contributing factor and here we do have another low cost, low risk, high benefit solution. Ceramics and oil spray. RR themselves proved on the Crecy that connecting rod based oil spray was insufficient. All modern engines in other motorsports with high HP/in2 ratios use a dedicated oil spray from the cases to the bottom of the piston crowns. While time consuming to fab, there is no risk.

As far as ceramic coatings, they are used in virtually every other motorsport, and yes, even some air racers use them. They are cheap, not very labor intensive, and have no failure in service that I can find.

As far as my own experience and my qualifications to critique conventional air racing practices, I come from other motorsports. Ones where we all raced 6 to 15 times per year on the daily lunch budget of an Unlimited team. Yes, our racers were cheaper; consumables were higher. I don't think I'm God incarnate or just some mouthy rail-bird; I expect that I'm somewhere in the middle. I do know that I bring something different to the discussion since I haven't been looking at the same problem from the same angle.

I speak mostly with pilots. I listen to what they feel is coming back through the stick and try to explain it. I also take those explanations to a few top mechanics that I trust. Everyone adds something valuable. The theory must explain the data, not the other way around. When the theory matches up with reality from all angles, we have something to work with. I don't quote any of my sources here, since most of the information provided is in confidence. One Unlimited pilot gave me the most incredible and controversial information about his airplane and team during race week this year just to get my opinion of his opinion. In parting, he looked me in the eye and said: "Don't you go putting that on the internet." I didn't. If this means that people here can't trust my opinions come from some basis of knowledge and acceptance in the engineering levels of this sport, so be it.

People like Thorn, Nixon, Barrow, (and not too many more) have committed their lives to building the most powerful versions of the Merlin ever to fly. This required a lot of research and learning over the course of 50 years, some from the land speed community, some from hydros, some exclusively from air racing. To say that they have learned a great deal in past years is obvious. To say that they can't get any better in future years is sophistry.

I make no apologies for my opinions; neither should anyone else. Open discourse is often emotionally charged and unspoken truth sometimes comes out of the shouting. In this case, Sparrow reminded me of a critical factor in trade studies for air racing technology that I had neglected. Now, let's apply it.

[GeeBeeZ1931]

Re: Misperceptions

Eric;
Sometimes I've come here and felt my ears needed asbestos protection.

I agree about dissenting opinions but totally disagree about the yelling and hot heads.

Why can't we all express ourselves and deal with opinions as well as real knowledge and experience like civilized people? Without the yelling and hot heads?

I've always hated it and its affects on me and I know it does the same to many other men. It brings out our hormones and makes us want to fight and yell and PROVE that I'M RIGHT. When that happens we are no longer listening, thinking reasonably nor learning ANYTHING. Bad way to settle what shouldn't have been an argument in the first place.

[Wild Bill Kelso]

Re: Misperceptions

It never ceases to amaze me: we all love ourselves more than other people, but care more about their opinions than our own.

Marcus Aurelius

[440_Magnum]

Re: Misperceptions

I'm just gonna hit on one thing since (a) I'm in a hurry, and (b) its the only thing I have a halfway decent input on anyway

Regarding knock sensors, I'm skeptical of the benefit for two reasons. Shoot me down if I'm wrong. :-)

Reason 1: Reaction time. So you detect the onset of detonation... whatcha gonna do about it? In a full digitally-controlled engine, the engine managment system can take action on the very next cylinder that fires (or at a MINIMUM on the next cycle thru the firing order). But if you're just using knock sensors as a warning indicator on a classically configured race Merlin (carbureted, ADI, and magnetoes), can a pilot responding to an indicator, or even some simple analog system controlling throttle or ADI actually respond fast enough?

Reason 2: Detecting the knock as opposed to other mechanical noises in the same frequency band. My impression has always been that knock sensors had to be backed up by some digital signal processing that is tweaked to the engine in question in order to actually be effective. Otherwise, you get spurious knock indications from various rackets different engines make. A dirt-common example is the Chrysler 2.2 turbo from the 80s. Its still rather popular in pocket-rocket front-drive dragsters and nutty applications like 13-second minivans because you can get 300 horspower out of it without any mechanical changes to the long block, but its prone to spurious knock indication and the resultant spark retard and boost loss because wrist-pin noise in that particular engine mimics detonation noise. Is there any way of predicting how a knock sensor would respond to the internal sounds a Merlin makes?

[MRussell]

Re: Misperceptions

If you could go to ionic knock sensing (detecting knock through the current of spark plugs in each individual cylinder, perhaps you could cut down on the "false noise" mircrophone effect that occurs with old-style knock sensors.

[Apteryx]

Re: Misperceptions

Question for the Merlin guy's-

Having most of my experience with 0-200's, do the Merlins use offset wrist pins.
Had a conversation with a VW race motor builder, and he said it eliminates the 'slap' on the power stroke, and is common for auto racers. His theory was the rpm we turn, it would probably make the gain, negligible.

Other then as a package, I've rarely seen Merlin parts, that haven't removed themselves from the motor

Paul

[Wild Bill Kelso]

Re: Misperceptions

If you could go to ionic knock sensing (detecting knock through the current of spark plugs in each individual cylinder, perhaps you could cut down on the "false noise" mircrophone effect that occurs with old-style knock sensors.

Ever heard of a coil on plug ignition? The GP and F1 guy are using a COP system with the knock sensor set up you're talking about.

[stuntflyr]

Re: Misperceptions

My question to Eric was also about knock sensors and spurious signals. In the O-360 it is so noisy that I have yet to talk to anyone yet that has had success with any system in that engine. Hence, I have mags. Yet I know very little about the systems themselves.

Also, about airplanes versus cars and their systems. Dennis Sanders summed up the whole necessity of back-up or redundunt systems with a tale of cruising leisurely at altitude in the SX-300 over some not flat terrain when a little smell caught his notice. Soon, it was not so little of a smell and then, as this was a night flight, it was quite dark. He turned off the electric stuff and motored on with mags powering the spark plugs until a handy airport came into view and he landed OK.
I know that we often don't race at night, but the fact that airplanes aren't always in a position to glide back to the field makes the testing and integration of electrical components controlling the spark plugs a serious consideration.

I am the first guy to think that John Sandberg was onto something with the Hillborn FI and twin Garrett turbos on the Allison in Tipsy Miss. It was a great start and I think that there are some great things to be tried yet. The blocks and partial engines I saw of the Dart program were really cool, and I think their is a real possibility of things like that coming to fruition.
But like the 100 hours put on MAII before it blew it's tail off (the most dramatic of flight test failures in racing and the reason for my using it here), testing to reliability is going to be the major milestone of any new engine, airframe or component in the aviation arena. A whole lot of experience is locked up in every new racing Merlin already, owing to the fact that there are 150,000 engines built and countless hours of operation of that engines design. Z and Dwight, and every other guy that built a Merlin that stayed in one piece for more than 60 seconds at 120 in hg knows that he had a few good feelings about his mods because he knew there were a few guys before him that had done it and so there was some at least anecdotal data saying that it might work.

With a new technology, or engine that comes along there will be a long road for it to seem even remotely reliable. Especially on an airplane. Because it will be tested right in front of us at the race.
For all 'Ye Hardy's' that will tread that path we should always say "good luck and thanks for trying".

Chris...

[MRussell]

Re: Misperceptions

Ever heard of a coil on plug ignition? The GP and F1 guy are using a COP system with the knock sensor set up you're talking about.

Indeed, it only works with COP ignition.

[440_Magnum]

Re: Misperceptions

Ever heard of a coil on plug ignition? The GP and F1 guy are using a COP system with the knock sensor set up you're talking about.

Not to burst any bubbles here, but coil-on-plug with ignition current waveform monitoring for misfire detection (and maybe knock detection too in conjuntion with microphonic knock detectors) has been standard on many production cars for YEARS. The first system I saw personally was on the 1998 release of the Chrysler 3.5L v6, and I'm pretty sure it wasn't the first to market.

[440_Magnum]

Re: Misperceptions

Also, about airplanes versus cars and their systems. Dennis Sanders summed up the whole necessity of back-up or redundunt systems with a tale of cruising leisurely at altitude in the SX-300 over some not flat terrain when a little smell caught his notice. Soon, it was not so little of a smell and then, as this was a night flight, it was quite dark. He turned off the electric stuff and motored on with mags powering the spark plugs until a handy airport came into view and he landed OK.
Chris...

Well, two comments. It all comes down to actual honest-to-goodness real-world numbers for the actual probability of failure. If you can get an electronic system reliable enough that its probability of failure is lower than that of a magneto, then in theory it should be safer than the magneto (of course on the time that it actually fails, someone's going to be wishing for a magneto anyway... )

Which brings up the other comment. Aircraft have dual ignition- so why not make one system EI and the other mag based?

[Blue Foam]

Re: Misperceptions

Aircraft have dual ignition- so why not make one system EI and the other mag based?

Timing. Spark energy is a big advantage of EI, especially at high boost. The other big deal is the automatic variation of timing with engine sensor inputs. The detonation margin of a multi-plug head goes down with the timing split, so any split would create a not-so-good operating point. If both were timed correctly at maximum boost, that might solve the detonation margin issue in the upper-left corner of the envelope (high MAP-low RPM).

If we wish to have a mag backup, then it may be better to switch at the leads. This way the EI (or EI's) time and control everything and relays allow for it all to go back to some self powered system (mags or otherwise) when electrical power fails. I realize this sounds a little wild, but high-power, fast pulse relays aren't expensive anymore and this would give the best of both worlds. The idea would require remote EI instead of coil over plug.

We still have to respect Sparrow's comment about risk. EI, like EFI, would require a lot of tuning across the entire engine map. In both theory and practice, it is easier to tune the fuel map than the ignition map if a mass-air-flow or O2 sensor are used. Ignition requires a lot of time on the engine, and these engines have VERY limited life. I'd just start with the knock sensor and see how much margin is there.

On that note, modern knock sensors are advanced enough to listen for a narrow frequency range. While it is possible that some mechanical transient would create a false positive, this would happen at all power levels and would be a cheap test.

Detonation is preceeded by unstable combustion that has a common frequency across all gasoline engines. It is the job of the modern detonation sensor to hear this frequency and give a warning as detonation is close, but before damage has occurred. The rapid response of an EI computer is essential for a 10,000 RPM road racer with fraction-of-a-second throttle transients. Not so for an Unlimited that pushes the throttle up once, and then takes .5 to 3 seconds doing it.

[440_Magnum]

Re: Misperceptions

Timing. Spark energy is a big advantage of EI, especially at high boost. The other big deal is the automatic variation of timing with engine sensor inputs.

I thought of that almost as soon as I'd hit "submit reply." If there were a way to slave the timing of the mag electronically- but heck, then you're into another whole development effort with maybe not a lot of payoff.

On that note, modern knock sensors are advanced enough to listen for a narrow frequency range. While it is possible that some mechanical transient would create a false positive, this would happen at all power levels and would be a cheap test.

Well, it doesn't necessarily happen at all power settings. In fact, it can be a combination of power level and RPM that sets up a particular noise. Not saying its impossible, just saying that its a known possible "gotcha" going in.

Not so for an Unlimited that pushes the throttle up once, and then takes .5 to 3 seconds doing it.

True, if you're just using the sensor to set the best operating point. I guess I was thinking of also using it as a safety net to shut the thing down if, for example, ADI flow drops or any other transient starts up that might lead to runaway detonation.

[MRussell]

Re: Misperceptions

Not to burst any bubbles here, but coil-on-plug with ignition current waveform monitoring for misfire detection (and maybe knock detection too in conjuntion with microphonic knock detectors) has been standard on many production cars for YEARS. The first system I saw personally was on the 1998 release of the Chrysler 3.5L v6, and I'm pretty sure it wasn't the first to market.

Coil-on-plug ignition has been around in cars since the late 80s, early 90s, but true ion flow knock control is relatively new technology, and I'm pretty sure it hasn't been on any Chrysler product yet.

[marada mx3]

Re: Misperceptions

From my experience in oval track racing, engine damage is caused by prolonged detonation. Any sensor that could detect the detonation would be helpful and allow the pilot to reduce power settings or whatever is needed to alleviate the detonation. We lived with detonation in our dirt track sprints and midgets. There was nothing we could do about it running the amount of compression and fuel we chose. It did necessitate frequent wrist pin bushing changes and inspections. Just my $.02

Tommy