We were standing trackside and talking about the new Acura NSX. My brother had just driven it to the overall win at the first-ever SCCA Targa event. It would have been nice if the NSX he’d been driving had also been a Targa; I believe they were called the NSX-T back in the day. No such luck. The new NSX does not (yet) come in Targa form. What can you do. We all have to face our own share of disappointment. Each worm to his taste, as the proverb says — some worms prefer to eat nettles.
My brother was off somewhere doing something so I was talking to a couple of occasional TTAC readers. They admitted that they skip my stuff and focus on the good solid tangible sales data from Tim Cain. I did not love them for this. But I discussed the NSX with them nonetheless, and at one point somebody said something along the lines of, “It’s a great car, but how are you going to fix something that complicated 20 years from now, without factory or dealer support?” And that was the sentence that triggered my Matrix moment, a mind spinning down a rabbit hole into a deja-vu past.
Fifteen years ago, I bought a massive collection of old Roundel magazines from a fellow BMWCCA member. Most of them were from the ’70s, which suited me just fine as those were the magazines I wanted to read. A few were from the late ’80s as well, and as I paged through them I remembered reading each of them for the first time when I worked for David Hobbs in 1989.
A common theme of Roundel magazine at the time was — wait for it, I’m going to be real with you here — how complicated and unfixable the modern BMWs were. According to various columnists and letter-writers, the modern BMW was a computerized toy made of brittle plastic that would be sentenced to junkyard duty in five years max. Ten, if you never drove the thing and lived next door to a sympathetic dealer. There were many reasons cited for this terror of the new cars. Electronic fuel injection. Emissions controls. Computerized dashboard computers that were computerized with blinky red computer lights. You get the idea.
The reader will recall that the 1989 BMW lineup consisted of:
- The E32 Siebener;
- The brand-new (and totally terrifying) E34 Funfer;
- The 635Csi and M6 that shared the old E28’s platform;
- The E30, in four-and-six-cylinder variants.
That’s right. The “disposable Bimmer,” the existence of which was bemoaned to the high heavens, was the fucking E30. Go to a club race in 2016, a full 24 years after the last E30 hardtop arrived on these shores, and you’ll see at least a dozen of them. They underpin every series from ChumpCar to AER to NASA Spec E30. To get a sense of how odd this is, imagine that you went to an SCCA race in 1988, the year that Roundel was whining about the E30, and the majority of the cars you saw were from the late ’50s and very early ’60s.
The same alarm was raised when the E36 came out; today, those cars are running ChumpCar with locomotive reliability. The E46 debuted to similar hysterics; you can catch them on the corners of every Cars and Coffee with 175,000 miles on the clock and duct tape holding them together.
You get the idea. The demise of the modern automobile as a durable, repairable proposition is much exaggerated and oft-predicted before its time. I’m reminded of the people who predicted that the world would end in famine by 1970 if we didn’t stop having children. The funny thing is that those people were right, given the data that was available at the time. What could not be predicted; the “Green Revolution” that would allow the Earth to overflow with people who like to eat but aren’t very keen on doing other stuff like participating in democracy or letting women speak in public or inventing airplanes or washing their hands after they go potty. Oh well. It’s a mixed bag, I suppose. Where was I?
Oh yes. I was saying that the predictions of famine were absolutely accurate, given the conditions of the time. So too were the predictions that the cars of the 1980s and beyond would be unfixable by their owners. What the Roundel crew could not predict was that the Internet would massively expand the access that the average old-car owner has to knowledge, specialized parts, diagnosis tools, and an aftermarket that is stronger than ever. Every E30 owner is on the Internet pretty much all the time from what I can tell, swapping troubleshooting information, learning about new and improved parts, and supporting Kickstarters for community-developed solutions to known issues.
Make no mistake. If your only tools to fix a 1986 325e amounted to the factory service manual and the phone number of your local BMW dealer, you’d be proper fucked, as Tommy says in Snatch. But those were the only tools available to people who were fixing their 1968 2002s in 1986, so there was no reason for anybody to expect that the situation would change.
Let’s call it the Red Revolution, in honor of all the people who have removed a transmission drain plug instead of the oil drain plug while following instructions on YouTube. The Red Revolution has made all sorts of very complicated cars fixable. Even I have taken advantage of it countless times to fix my Porsches and my other old cars, even though I’m very far from being a mechanic of any sort. Thanks to the Red Revolution, I don’t have to take my 2004 Boxster S to the dealer for a $150/hour molestation, I don’t have to pay main dealer prices for parts, and I don’t have to guess at how to perform basic maintenance.
But the Red Revolution has its limits. One famous example of it is the Lamborghini Diablo, which is chock-full of specialized parts that simply no longer exist. There are a lot of cars for which the wiring harness is becoming a unicorn item, as well. Sometimes, as with the V12-engined W140 Benzes, it’s because the harness is deliberately biodegradable.
Many of the newest production cars, like the 2017 Acura NSX, rely on a veritable Beowulf cluster of supercomputers, sensors, and electric motors. Right now, it seems unlikely that you’d be able to pull a 2017 NSX out of a junkyard in 2042 and restore it to working condition. Yet there’s a new revolution coming to supersede the Red Revolution and to make exactly such a thing possible.
The harbinger of that revolution is the not-so-humble Megasquirt community-developed standalone fuel management system. It can replace the ECU in nearly every vehicle you can imagine. It makes engine swaps easier and offers a real, workable alternative to people who simply cannot source affordable ECUs for their vehicles. One example of this is the humble Plymouth Neon. Over the years, most of the Neons out there have fried their ECUs. The rare and desirable Mopar ECUs which permit a higher rev ceiling are pretty much all gone. It’s easier just to Megasquirt the car and enjoy control of the fueling system that the OEM computer couldn’t even dream of providing.
Let’s call it the Mirror Revolution, because it’s all about emulation. The Megasquirt works because it is thousands of times more intelligent than the dedicated-chip ECUs that it replaces, so it can be easily programmed to emulate any of them. If you’ve ever run a Nintendo or Atari emulator on your home computer, or played the MAME arcade games, you know the concept. Even the most pedestrian smartphone today can easily run a complete model of a Super Nintendo chipset in software.
The day is coming when computers will be so smart that they will be able to “black box” any existing component in a matter of hours. You plug your rare old ECU or any other computer into the emulator. It will offer every possible combination of input signal to the ECU and check to see what the responses are. Presto! In no time, you’ve got it modeled in software and can run it on any generic processor. All you need to do is fabricate the particular plugs or displays used by that system.
Fabrication, too, is becoming more widespread. Again, I tell you that the day will come that you can have anything from a Diablo wiring harness to a 944 Turbo control arm to the front wheel electric motors of the 2017 NSX fabricated just by having a brief chat with an “expert system.” You’ll be amazed what this makes possible. Brand-new “Offy” racing engines, replacement parts for a 1914 Cadillac, you name it. I’ve written fiction about this, but reality may soon outpace my imagination.
The only problem with all of this is that the parts and solutions offered by the Mirror Revolution will be as superior to their templates as the Megasquirt is to the Mopar ECU. That doesn’t sound like a problem, but wait until the day comes when you line up for a Spec E30 race in the year 2039 and you start to have a sneaking suspicion that the car next to you is entirely fabricated from carbon fiber and titanium, even though it looks exactly like a rusty old ’91 318is. What will we do, when a near-infinite number of available CAD cycles iteratively designs an E30 with the weight of a Lotus Elise, the chassis stiffness of a LaFerrari, and the shock tuning of a Formula One car? How can you compete against that? More importantly, what will the folks at Roundel have left to complain about?
If you went to an SCCA race in 1989, a large percentage of the cars competing were British roadsters from the ’50s and ’60s. Look for Road & Track coverage of the Runoffs if you doubt it. The bugeye Sprite owned H-production until more recently than that. Hardy Prentice was competing with a TR3 well into the ’90s in E production(IIRC), where he competed with newer Triumphs made before the TR7.
The smallbore production classes were still pretty much old Brit roadsters until the SCCA started classing some sedans in there, because no one save Mazda was making roadsters past 1980. I wouldn’t say they make up a large percentage of the entries, though. Count all the Formula cars, all the sports racers, Showroom Stock, and GT classes, you’ll find that they very much outnumber the MGs, A-H’s, and Triumphs.
Nice piece.
I enjoyed it as well. The car End-of-World scenarios are so much more visible today because of the internet.
Bet there were a lot of fearful American car buyers at the end of carbs who thought fuel injection would never work out.
About the only End-of-World situation I can see is the auto industries ability to continue market segmentation. Alfred P. Sloan’s brilliant means of selling the same car at wildly different prices continues today with consumers well aware that there are no real costs differences between cars, especially cars on similar platforms (leather is often cited as the *only* real difference. I expect that extremely complicated engines make a difference, while displacement certainly doesn’t – although it certainly adds to the price*).
Best guess is that consumers will remain preferring the existing system of paying dearly for an honest advertisement of wealth. Modifications will remain the domain of hobbyists and their creations will be as scorned as the “greasers” and “ricers”.
* I remain convinced that the CAFE was much a product of Detroit as it was of DC and used to kill muscle cars and put the “big fancy engine” as a reward for buying the big fancy car instead of being a option on the cheap car. CAFE let them artificially limit the supply of big engines and charge all they wanted for them. A win all around for Detroit (as well as whatever DC was doing as well).
Then car buyers wised up and started buying trucks, where they could have their big fancy engines for not a whole lot of money.
Well, to be fair, there were a lot of *horrible* early FI systems, from what I understand.
(The oddest thing, I think, seems to be all the mechanical diesel FI systems, which have roach-like durability.
I’m sure there was some reason or other that tech wasn’t applicable to gas engines, but I’m not guru enough to know what, other than “probably it has to do with the throttle”, because that’s the obvious difference in operation.)
We need for Scoutdude to come and opine.
The World doesn’t end. It evolves and adapts.
Tell that to dinosaurs.
They adapted as well, in modern times they are mostly noted for crapping all over your freshly washed car while passing overhead.
Maybe from someone who doesn’t turn a wrench or own anything complex, but just the writing and tuning of a ecu/tcu with the correct software like EFILive or HPTuners will make your head spin.
Try marrying a bcm to modern day car lately?
I ran FAST standing alone for the Grand National 3.8L on my Bertta back in the early 1990’s it was $1,000 for the computer system and $500 for a NTK wideband oxygen sensor. No abs, stability, traction control, VVT cams, or electronic throttle body to worry about back then. Just fuel, air, spark, and timing, hence FAST.
Modern day tuning is not that bad and the nice thing about Megasquirt/Microsquirt and other similar systems is that there are big communities out there and base maps to get you running.
You can pickup a Microsquirt for around $370 and adding an Innovate wideband will run you another $160. This will get you going for many cars.
The software is adaptable and many are already being used to control secondary modules and systems.
It is not simple to tune but is accessible for those that want to spend the time to learn it
Specific example – True to my username, I’m in the process of using that exact combo – Microsquirt / Innovate wideband – to replace the Bosch L-Jetronic on my Alfa Spider build.
Surprisingly there aren’t a lot of how-to / build threads for the Alfa engine at this point, which I’m trying to rectify. But when all is said and done the end result should be far more tunable than the stock Bosch EFI, have lower emissions and better MPG than a set of Webers, and be easier to repair and maintain than the old SPICA mechanical FI.
I believe Turkish and Mickey mention that phrase to Tommy during the hare coursing scene. I don’t think Tommy ever says it. #pedanticaboutminutiae
The folks reading Roundrel will always have map pockets and badges to bicker over.
Bmw owners of yore are the leasers of today. Just put gas in it and go to the auto car wash.
You make a good point.
However (and it is a big however), there is an unmistakable trend toward more and more small (or complex, in the case of electronic modules), bespoke parts. This is particularly true in the vehicle electrical systems.
I speak as a former electronics technician who has witnessed the increasing difficulty of debugging and repairing ever-miniaturizing circuits.
Running counter to this trend is the dependence upon software which, given the right circumstance, can decrease physical complexity. Of course, systems are becoming more complex at almost the same rate, so this may be a wash.
I suspect, that you are probably right, this complexity doesn’t exist in a vacuum and diagnostic tools and techniques improve at almost the same pace. The issue is that these tools can be just as narrow-focus as the systems they support, making them expensive and rare as we go forward.
One step closer to SkyNet…
I remember reading about a UK man 3d printing parts for an old E-Type, he was building a complete car from scratch.
And on Count’s Kustoms they were restoring an Imperial. The customer insisted on NOS taillight lenses not 3D printed ones. I just rolled my eyes.
Said customer sounds like The Old Man.
It was “The Old Man” – personally I might have asked to inspect the 3D printed lens before it was installed to ensure that it was good enough to be mistaken for the real thing but otherwise it seems like pissing money way to worry about whether it was the “real thing” or a “copy”
I wouldn’t try and make lenses from nitrous, myself…
(I slay me!)
Does that photo slyly introduce the author’s nickname? Until lately, anyway.
Death to the increasing number of “Body Control Modules” which create a fail point that whole systems to fail through one part.
I had an employee get stuck in Utah for an extra day due to a BCM failure in his 7 series, only took a dealer a whole day to diagnose it because the last thing they tested was the BCM. (How did a teacher afford a 7? He was already retired from another branch of government where he made far more money than I.)
The BCM on my “free” ’98 Voyager likes to freak out. For a while, the headlights would randomly come on when the car was parked, off, and hadn’t been driven in days.
Dealer said the battery wasn’t fully charged and reset it. Now I can’t get the headlights to to turn off when the car is on, but as long as they are off when the car is off, I’m fine.
I’m hoping it hits 20 before the next emissions test so I can get historic plates and not have to worry about it failing for lack of data.
I have a hard time feeling sorry for people who have broken 7-Series vehicles. They’ve been consistently pretty bad since 2002. This is known.
Trust me when I say that the alternative—a wiring harness as thick as your thigh—is so much worse. Having to do electrical repair on pre-BCM luxury car is evidence of why these exist.
Body control modules mean one subsystem fails when the BCM dies, which isn’t a bad thing: it means that problem areas are effectively firewalled from each other, and easy to diagnose.
The alternative would be going through one massive bundle of wires and a single, monolithic control box.
You’re absolutely right. We know two things about the future – it’ll be different and it’ll be grounded in today.
We have 3d printing, cheap cad cam, and 5 axis CNC mills that you can rent / send jobs out for a very reasonable price.
Surely the future will follow. You hear about “THE CHINESE” ripping off stuff by 3d scanning it and 3d printing it, but within one of my motorcycle’s communities, I got a bike with an usually intact side cover, and I was able to scan it and load the 3d files to a forum and other people over the world were able to 3d print the part so they could cover their batteries and stuff. A perfect reproduction is just a matter of cost.
Also, a lot of parts / connectors are getting standardized. Electrical harnesses could be the next thing to become easy to change. Most companies even use the same O2 sensors and interfaces these days!
I just wonder if there will be enough interest to make it happen. Electric cars are going to come online so fast and they’ll be so good once energy storage is good, who would bother with a 2wd gas car that needs work when you can get a carbon fiber, AWD torque vectoring monster on the cheap?
That said, we already have 250cc superkarts which aren’t just fast for their price, they’re fast. And people aren’t going crazy over the superkarts.
The benefits of electric vehicles are all 100% practical and objective. Cleaner! Simpler! More efficient! But they don’t exactly engender enthusiasm like a gas engine does.
I think for the majority of folks whose sentiments on cars and driving range from apathy to unbridled rancor, the EV revolution will be great. But for the oddballs who still drive stickshift and the like I’m just not sure EVs will ever cut the mustard. An extreme case is motorcycles- a lot of people exhalt e-bikes. If all future motorcycles were electric I would probably stop riding. W/o what the ICE contributes to the experience it’s not worth it to me.
I’ve been driving a plug in hybrid for the last two and a half years, and I am totally seduced. In addition to the quiet and lack of vibration in EV mode, having instant torque and such a linear response to the accelerator pedal is very compelling.
At this point, I don’t think I could go back to a conventional drive train, at least as long as most of my driving is in town.
+1
When people say they like rheostatic torque and instant throttle response, they don’t know they’re describing an electric powertrain.
If we had batteries as good as gas, we wouldn’t even be having this conversation.
Sure, some ICE engines sound nice. And some are fast. But electric motors are so devoid of tradeoffs – you can get good efficiency out of fast motors, you can drive along in the meat of the powerband, you can deploy electronic aids super easily, you can recapture energy during braking, ect. that many people on this site will become die-hard electric motor fans and they just don’t know it yet.
To this day, I don’t know anyone who left an electric car for a gas car for anything other than range issues. I don’t know a single person who missed the vibration, noise, or going to the gas station.
Like I said, electric power has plenty of upsides… all of them being very practical and objectively based. But nobody is passionate about Formula E. The Tesla S needs a 0-60 time in the 3 second range to be even remotely pique the interest of the average gearhead/enthusiast.
But at the end of the day, cars like the 911R still exist and sell out instantly. I’m sure Ferrari and McLaren could have ditched the gas engines in the LaFerrari and P1, but they didn’t. Bottom line, you can’t win over enthusiasts with practical concerns. It’s like trying to convince a girl to sleep with you because you’re a “nice guy”. That’s just not happening. When pure electric motors can get people to follow related motorsports or choose said vehicles over gas vehicles en masse then we can talk
They don’t garner enthusiasm yet. Frankly, a lot of cars don’t have engines worth enthusiasm. What joy is there in the Fusion’s 2.5 4 banger? Or that 200 horsepower V6 that graced so many jeeps?
For every 2000-2009 S2000 there are maybe 50 inline-4 accords.
A lot of random cars (Jack’s Neon excluded) are just kept alive to serve as relatively cheap race cars. Not because they’re excellent cars with a race pedigree (whatever that is on a street legal car) but because they’re cheap and good enough.
Agreed Sporty, I could care less if an EV could literally climb walls. Its just not the same. EVs are the ultimate expression of “being along for the ride” more so than even cars like the GT-R.
I’ve seen the term “drivers” and “driver’s car” come to increasingly mean people and vehicles that require a certain level of engagement when driving. Not that it implies any sort of superiority though.
The engine on my ’08 Civic (stick) may not be Mozart, but it IS Salieri to the Boxster’s Mozart.
Horse stuff is still a supportable hobby today so I don’t think autonomous or electric vehicles are going to completely kill gas-powered ownership. Historical cars are too culturally significant for people to just forget about them.
No doubt.
But if you pick up an AWD, 500 horsepower, 2000 lb monster for 10k, how many people would keep E46s running for autocross?
Direct software based emulation has major drawbacks. For games it’s obviously not a huge deal. But for a major control system in a car? That needs accurate emulation down to the transistor level. That is really really difficult.
If you are interested here’s an article from the author of one of those NES/SNES software emulators that strives for absolute accuracy and just how much processing power that actually takes and just how difficult it is.
http://arstechnica.com/gaming/2011/08/accuracy-takes-power-one-mans-3ghz-quest-to-build-a-perfect-snes-emulator/
One could easily extrapolate how difficult it would be to achieve something similar in 2046 with a car built in 2016.
Nope. The problem with SNES is trying to match one specific type of hardware to a completely arbitrary piece of hardware (a modern PC). If you absolutely insist on using a 2046 iPhone, you will almost certainly have the same issues as the SNES emulator. If you ditch the iPhone and throw an FPGA in your megasquirt, you have a trivial (if relatively expensive) solution for the hardware (emulating the software might require acquiring some proprietary information).
Of course, a huge part of this is that auto electronics lags the electronics industry by quite a bit. If the auto manufacturers *really* needed cutting edge stuff (and were willing to design entire chips the way Nintendo did to get the most out of them) than it would likely require 2046 tech to do it (and since Moore’s law is in real trouble, it presumably won’t happen in 2046 with silicon devices. Maybe carbon nanotubes, but not silicon).
+1 wumpus
I was going to reply to say exactly this: 98% accurate emulation is not hard, but those last 2% are killer: you have to devote a huge amount of CPU time to emulate the weird bugs and interops in old silicon.
A 100% accurate NES emulator is tricky enough that Nintendo has had to use FPGAs to do it, and the NES is not a complex box. The same applies to the Commodore 64’s sound chip. More complex systems, like the Commodore Amiga, are impossible to emulate in software because they’re a total clusterf__k.
ECUs, though, are really not hard. They’re very simple silicon, with a limited number of inputs and outputs. You don’t need to worry about syncing the chip than handles sprites to the video scanline to the input sixty times a second.
Oh. My. Goodness. I shot the Wumpus on a teletype in the early 70’s. Who thought that silly thing was the future of computing.
” That needs accurate emulation down to the transistor level. ”
… It needs to accurately *enough* do the correct *output* for any given *input*.
It doesn’t matter *how* it does it, whether it’s a circuit-diagram emulation or algorithms based on an original spec, or whatnot.
(You don’t need a 3gz-level “perfect” emulator to actually run SNES software in a way that people using it will find accurate.
Likewise you don’t need to simulate the noise on individual transistors to make a given engine run, and even run correctly and well.)
Now, Jack’s wrong in that you can’t “just put in every input and look at the outputs” to black box it in any simple way – because so many things are *on timers* or rely on previous state and without at least an idea of the specs and states to model you won’t know what the problem space even *is*.
But Jack’s right in that once you have even a good idea of that you can get a good enough output that nobody will be able to tell from factory – and the regulatory requirements, EPA test requirements, and examples from other manufacturers, and what the shop manuals say it should do … they ought to get you “good enough” to black-box it.
Good enough really is good enough.
As long as there is a community of people who want to keep a car running, someone will make the parts available. On the other hand, if you’re trying to restore something that hardly anyone else is interested in, parts will be scarce and expensive.
This.
Jack’s examples are all cult-favorite cars (BMW, Porsche, Neon).
I have a 944, and I’m constantly amazed at the parts availability for a 30+ year old car. I’ve owned other old cars, makes that didn’t have cult followings, and Jack’s rosy assurances don’t apply.
Also, while the MegaSquirt is clearly a fabulous piece of kit, there’s a huge time investment required to properly tune it for your particular engine, as noted in several other comments. If you’re racing your car, great. However, MS is basically illegal on the roads of CA, unless your car was built before 1976.
Unfortunately for us in CA, the smog exemption isn’t rational, like “25 years old”, it’s “1975 or earlier”. Yes, I think the CA law is misguided.
Regardless, if your car has a cult following (BMW, Porsche), the future is bright. If you’re maintaining an old Peugeot, not so much.
I’d love to have a Peugeot 404 wagon (family got it when we arrived in Paris Aug ’65 for the year; it was the first car I drove legally, four years later). Trouble is, the 404 is virtually 404.
“Even the most pedestrian smartphone today can easily run a complete model of a Super Nintendo chipset in software.”
While technically true in some sense, most emulators are hacky pieces that prioritize performance over accuracy. Accuracy emulators require a ton of power. Try running higan (the 100% accurate SNES emulator) on anything less than a 3GHz i5 Ivy Bridge and you’re going to be in trouble. However, accuracy can be sacrificed for speed, and Higan in balanced mode (which still has a few bugs with specific ROMs) can do 99% of the work on a far slower machine. A smartphone emulator will play a lot of popular games (mostly) well, but it is an imperfect method.
Emulators have bugs and may not properly preserve the experience. However, FPGAs (field programmable gate arrays) are what will carry on the legacy of computerized automobiles. These (not cheap) devices are what will allow functional recreation of old computers. You could literally make a replica ECU-on-a-chip, given the engineering resources and testing and money.
The MegaSquirt is not really an emulator, it is more of a black-box re-engineering of a traditional ECU. It’s designed to act as its own brain for ignition systems. What I think people are concerned about on the NSX is the hybrid/electric motor systems, but truthfully I believe those problems will be solved eventually as well. At the end of the day these systems are created by humans and if a human made them, another human can recreate them given sufficient time and resources. It’ll cost you, and you might not be able to do it solo, but it can be done.
Late to the party, but what about stuff like Adaptive Cruise, and stuff for semi-autonomous motoring that the automakers STILL can’t get right sometimes (like emergency braking that locks ’em tight when there’s not an obstacle in sight)?
That said, I think Jack’s onto something, and as a computer professional, I’m in agreement with him!
Good reading as always .
The whole 3D parts making is interesting , I can’t wait to see indie shops spring up making parts on demand .
-Nate
My personal feeling is that 96-06 is MY golden age of recent cars. Electronic ignition and fuel injection made the vehicles reliable and were making decent useful power, without being completely mired in electronic failure points. OBD II allowed for reasonable shade tree diagnostic. As an example, we solved the “cranks like mad but won’t fire” issue the Hyundai was having with a scan tool, which lead to a new CSPS from rockauto. They had cable operated throttles and switches instead of way too many sensors and modules.
Plus, they just looked better. Actual glass you can see out of! Low, flat belt lines! If they werent rusting away I’d buy another Alero and Sonoma in a heartbeat.
I really find these stand alone engine managements cool. I enjoy watching Roadkill, Hot Rod Garage and Engine Masters on the MT channel, and I think crate motors with modern electronics are sweet. I’d love to resto-mod something cool with a crate LS, fuel injection, and then have a modernized interior with A/C, an auxiliary input, etc.
Back before the Verano, I used to say as long as I had power windows, locks, cruise, sunroof, keyless entry, A/C and an aux input (or tape deck!) that I was happy. I still feel that way. I love tech but I think cars are getting out of hand.
Thanks for reading my rant!
You’ve got the dream of the ’90s.
I agree, I hate the high beltlines of current cars. About the only thing I like about cars of the last 10 years is the availability of AWD, that’s great in the winter. A 1996 XJ6 with AWD would be perfect – beautiful and functional.
tell me about those effin’ high beltlines. Sheesh!!!
Right on Dave, I feel the exact same way, and my mechanic brother I think has the same leanings.
My ’00 Maxima feels like a smaller car than my ’12 Civic and feels much easier to drive. The belt line is lower, the windshield isn’t some crazy slanted back thing, and the trunk isn’t riding high obstructing rearward view. my ’96 4Runner is likewise a peach to drive from a visibility point of view, and is a doddle to work on due to both the BOF layout as well as the very basic nature of the beast (no endless control modules or electronic throttles). Thankfully even something like my fiance’s ’12 Camry is pretty easy from an under-hood perspective. When I had to replace a battery last fall I could simply swap in an Autozone unit and get on with my life, no ‘teaching’ the car to recognize the battery. So if you pick carefully, there are still some newer cars with simple under-hood layouts, with port injection and transmission dipsticks.
gtem brother, if you ever come north we shall hoist a beer in favor of cable operated throttles!
I just realized you often head north, but then very far to the west… or I guess east?
I wonder how much of this kind of custom stuff will be legal, at least for on-road use. In my state, 20+ year old cars can get historic tags and don’t have to be emissions-tested, so it would be fine. But if more states start going the CA route and requiring everything to be certified for sale, then that custom flashed ECM or 3-d printed muffler is probably going to be illegal.
As far as I know, no megasquirt can be legally driven on US roadways. Of course, should you use one as described above (complete emulation) you will be faithfully following all emission protocols and if you “legally” flash the existing ECU with a custom tune you will likely won’t. Personally, I can’t believe anyone would go the megasquirt method and not tune the thing to their liking, but that’s not what the article describes.
But doing one brings money to companies allied with the EPA and one doesn’t.
Thanks Jack. It’s amazing how you manage to write pieces that feel custom written to my interests and history, until I read the comments. Who knew?
Just as generations of car folks have worried about cars becoming too complicated, throngs of VW aficionados have dismissively stated that all the problems of the previous generation have been fixed with the latest generation and there is nothing to worry about.
Yes, I know that was a cheap shot.
I’m concerned that if everything starts to develop for the Mirror Revolution, patent attorneys will make themselves a ton of money by going after end users and small-time builders in similar ways the RIAA did when MP3s started reaching into their pockets.
Depends on the object and how you copy it.
If it’s a design patent, the term is 15 years – and only the design, non-functional aspect can be patented, meaning a copy that takes care to change the non-functional looks can escape it.
If it’s a utility (normal) patent, the terms are longer.
But most parts of a car, especially mechanical bits you’d be copying, just aren’t patented – because there’s nothing novel in them.
(Likewise, unlike the RIAA, which had members who would be happy to sell you that music, there’s no *motivation* for e.g. GM or BMW to sue people for making parts *they no longer have any desire to make or sell*, for a tiny enthusiast market.
The RIAA sued people because it was bankrupting their members to have people Getting That Music For Free.
Nobody’s not buying a new car because they can print a replacement whatsis for their hobby car … so nobody’s going to care.)
The 3D printing technology has already gone mainstream to the point of commercial airliner parts. Norsk Titanium is building a facility in Plattsburgh, NY that will employ over 400 and PRINT titanium aircraft parts. This technology has taken a giant leap in a very short period of time. Not long ago, you could print plastic knobs and small busts of George Washington.
I’m developing a commercial version of an invention of mine, working with a 3D print shop. While it’s quite a revolution in material processing, prototyping and manufacturing, it’s not magic. You can’t print over thin air, gantry structures have to have removable/dissolveable support structures printed first. The orientation of the print is also a critical factor, some things can only be printed in one direction. While the choice of filaments is expanding, there are still limitations on what polymers (or metals if it’s combined with laser sintering) can be used.
3D printers are cool, but they’re a few parsecs away from Star Trek’s replicators.
After the recent brouhaha between SEMA and the EPA, I wonder if replacing electronic components on a modern car with emulators would constitute tampering with the emissions control system. Put another way, you might be able to keep your 2017 Whatever running, but that doesn’t mean CARB will let you drive it on the road in California.
I’ve done a Megasquirt conversion to a 1966 Jaguar, and developing the fuel and spark tables in neither easy or quick. It takes a lot of dedication and investment in learning and time to be able to get a car to run. Many people are unable or unwilling to invent the amount of time and learning that is required. Like it or not, you’ll become an EFI expert after doing an MS install, and many people simply don’t have the technical aptitude to do it. There is a reason they buy a car instead of building their own!
The 80/20 rule applies, it’s not that difficult to get MS’s fuel and spark tables 80% of the way there, getting it the last 20% is much more difficult. Basically it’s easy to replicate a carbutettor, getting to 90’s level of EFI cold starting and idle qualities is much more difficult.
MS also only handles fuel and spark. There is a separate box for electronic transmissions, but the data is available for only a few transmissions.
Many later cars use canbus communication between all the modules, and MS can’t do that. You might be able to get a later car to run on MS, but with limited functionality of everything else. Even in a 90’s car, it’s amazing how networked everything is.
Jack, As you have experience with MegaSquirt, do you think it will solve the rough idle of my 1963 Thunderbird? I’ve rebuilt the carburetors and the block has been massaged from 6.4 ltr to 6.9 ltr with a lumpy cam.
With the modifications i’ve gotten it up over 8 mpg a few times to appease the Prius’ camped in the left lane as I pass.
Probably won’t solve the rough idle. It’s the lumpy cam that is the problem! MS needs to know the amount of air coming into the engine to meter fuel (as does a carb)and the problem with a lumpy cam is flow reversals – the air doesn’t move smoothy. You get flow reversals, so some air is measured twice, and that will mess up the fuel calculations.
I noticed a jump in torque going to EFI and the distributorless ignition, you might be able to go to a milder cam and then go to MS, which would allow you to keep the power levels of the more aggressive cam but have the smoothness that would make EFI work better. The fuel will be more accurately metered than a carb can do, and that gives mileage and power. On my 1966 Jaguar S Type I saw a road test in 1964 where they drove the car from London to Geneva and averaged a bit over 16 mpg. I did a similar length highway run in my car and averaged 30 mpg.
I think we’re transitioning away from a time (especially now that a lot of cars are partial hybrids) in which “working on” cars will be less about the powertrains and mechanicals, and more about a dead iDrive screen taking up space in the middle of your dashboard.
Blame the E65 which, while groundbreaking, seemed to be the first time BMW started putting prototype-grade technology into its cars.
Who knows what the remedy for that will be. Maybe people will come up with new interfaces to replace the OEM ones.
I’m not as sanguine about the future as Jack. The Golden Era of automotive microprocessing started with the ’77 Toronado’s MISAR and began its slow fade with the advent of OBD II. The important task back then was to simply make things work. Hardware ruled. Software was a footnote.
Nowadays, the Ghost In the Machine is software-centric and is probably as much of a lawyer as a technician. Software will become progressively more encumbered with encryption and booby traps (e.g. the current bootloader brouhaha involving the Samsung Galaxy S7) while hardware bits will be deliberately designed to thwart easy 3D-printer reproduction. The corporate bean counters can see into the future too.
I’d suggest the car buyer of 2016 is already in the same boat as the contemporary farmer. Automotive folks just haven’t fully realized it yet.
“Farmers Demand Right to Fix Their Own Dang Tractors” == http://modernfarmer.com/2016/07/right-to-repair/
I hope I’m wrong. But in case I’m not, I have a 1967 Cadillac Eldorado that I’m slowly restoring.
The popular thing around here in the 80s and 90s was to remove the computer / EFI systems and go back to the ol’ carb. More of a retro-conversion thing.
I was thinking of going that route when I was planning to plop a 383 in a stripped down ’91 Caprice. A project that ended up going nowhere.
The old ’86 Monte Carlo SS I used to have was modified the previous hillbilly owner – remove computer connection to carb (which makes it go full rich – derp!), and replace the exhaust – cat and the piping to the muffler – with headers.
Gas mileage was terrible and the cruddy 305 was a pain to start in cold weather. Mileage and performance shot up when I replaced the 305 with a 355 and an Edelbrock (Edelcrap! as the joke goes) carb.
My friends are literally building their own parts with 3d printers and a lathe. They use CAD and digital scanning. Of course they’re running megasquirt to run it all.
Just wait until the ferro-carbon metals are widely in use for engine components as well – as strong as steel, as heat resistant as iron, as light as aluminum.
The hard parts are always the obstacles. There are people out there that will rebuild your Trofeo’s VIC to perfection. But try finding a headlamp door. Or a back glass for a Reatta. Stuff like that.
Bought an ’89 Camry V6 2.5l a few months ago. It had been parked for three years due to a neglected timing belt/water pump — and the owner being smart enough to park a complex interference-engine vehicle.
The repair was beyond his capability and would’ve taken me a week to do. So I paid a garage to do it. Guy said it “was a good engine and the job wasn’t terrible”. Few days later I did notice a few more greasy fingerprints in the service manual (sitting in the car) though…
Several hundred dollars later, an oil change, bleeding the brakes, and a good wash I have a reliable vehicle.
Now… Have a friend who as a child always lusted after an E28 635-whatever. He eventually wound up with three. None were ever reliable. Cracked dashes, faulty guages, clattering valves, sticky shift linkages, etc.
Everything in my 27-year old Toyota simply works — down to the stupid electric seatbelts.
You know what? I love ODB-2.
I love that I can plug a $5 chinese Bluetooth code reader into my car and pair it with a $5 Android app to get a rough approximation of what the hell is wrong with my car.
I don’t need exotic tools, I don’t need to mess around measuring voltages at various relays, I don’t need to throw parts at it. I can get 90% of the way to a solution very quickly, and as Jack notes, I can get the rest of the way through publicly-available research materials.
This might not be a huge thing for an experienced mechanic, but for a regular owner, it’s a huge benefit.
The Megasquirt is the ECU of choice for NA/NB Miatas (1990-2005). You can get kits that require you to assemble and wire them, or you can get a plug-and-play version complete with a base map that’ll run the car on the first try. It’ll also handle forced induction and allows owners to bolt on a snail and exhaust system to double their power output without so much as taking off the valve cover. It’s an amazing tool, although definitely seems to take time and work to master.
“There are a lot of cars for which the wiring harness is becoming a unicorn item, as well. Sometimes, as with the V12-engined W140 Benzes, it’s because the harness is deliberately biodegradable”
Isn’t there a service that replaces the wiring harness in cars like this (and your average V12 Jag) with body computers so as to avoid harness hell?
I hope there is!
Luckily, there are re-wiring services out there that will do work like that but they are usually pretty expensive since it is time consuming work to build them. The LS swap type harnesses are cheaper since most of the shops will build jigs that will make the work easily repeatable.
I’ve had experience with w124 wiring and it is a pain. The throttle body wiring fell apart and prying it open to rebuild was almost impossible.
I would like to note that there was also one person that was standing there that said that they read something I wrote about the WRX
Love the megasquirt. Only way I could get my old supercharged Miata with a hot side blower to run worth a darned. I was a novice with standalone ECUs and managed to get it working. No piggyback solution ever got close. not a cheap solution though once you factor in dyne tuning and all that.
The folks at Roundel might still be complaining about run-flat tires. Andy they will moan about cars being too heavy and complicated anyway.
Tech talk is great and I like the stories about road trips, but the complaints about disposable cars and nostalgia for older cars is very tiring. Interesting to hear that it was the same way almost 30 years ago.
Some people have a hard time thinking beyond the moment. Nature always finds a way.
Hi Jeff!
AFAIK Megasquirt does not handle (at the time): electronic throttle pedals/throttle bodies properly, no CAN functionality (your dash and other things like the TCM rely on engine data _very_ heavily)
Not that your point does not stand: we simply don’t have that technology yet, somebody is bound to develop it at some point in time.
Jack Jack Jack, you never worked on a ’67 VW Type II or IV back in the day, did you? We didn’t have a clue what to do with those computer boxes, and we didn’t even really know all that they did. Just when all other troubleshooting came to naught, we swapped boxes with a wrecked car (it helps that I worked at a salvage yard too).
Then along came L- and K-Jetronic, along with Nizzan’s (or was it Datsun?) imitation Bosch injection systems, and those were pretty straightforward. Sigh of relief. And they were easy to troubleshoot because mostly it was the fuel filter or the pressure regulator (located on the engine block, yeah).
I have a 30 year old C4 that thank goodness my own experience stops at. I went back into aviation in ’86. But at least that car is, mostly, still fixable by me. No electronics in the suspension, just the funky VATS and the ECM, basically.
NOW you tell me along comes this Squirt company and makes it all easy again? Dammit, Jack, I was just starting to feel young again!