the ignition module is installed INSIDE the intake plenum valley - very harsh conditions
Huh? It's there for cooling. It's way less harsh in there than it'd be just outside of it. If it's bonded to a metal mass, it'd also be subject to slower temperature changes. Sometimes those engineers aren't as braindead as it might seem, you know.
You use electromechanical controls. That piezo sensor generates voltage, you can use it to drive a sensitive amplifier valve in a pneumatic circuit. Yes, there would be some mass increase, but it all depends on how fancy you get with computer modelling and design of this stuff. We can pull of electromechanical assemblies today that would be nigh impossible to design and manufacture even 60 years ago. You can have perfectly adequate electromechanical amplifiers even for O2 sensor output.
Now they are out of targets where search space scales linearly with findings. They have an exponential decay in findings to fight with, and you can't with with an exponential by brute force.
Speak for yourself. I'd be the first in line to have such a biopsy done, were it to be routine in differential diagnosis for, say, depression. These days they have the anesthesia/analgesia down pat, one would have to be truly crazy to reject an objective measure just because there's a biopsy involved. I've had my septum straightened out under very local analgesia just to skip school. And I walked to and from the hospital uphill both ways in a blizzard:)
And of course if you look at a random electrolytic, you can immediately visually tell what grade it is, ha ha. Yes, sometimes you can. Most often you can't. What I find interesting is that many automotive qualified parts that I use, from legit sources, often have less verbose markings than more run-of-the-mill commercial parts. Many commercial electrolytics have the temperature rating printed on the case, none of the automotive parts I've used (not that many, admittedly) had such markings.
That's assuming that you can make sense of component markings, and this applies to large-ish ICs only anyway. Small packages (a couple mm across) can be marked with a couple alphanumerics only. Nothing to tell you what grade they are. Zilch. Nada. Same if you look at a ceramic resistor or a ceramic capacitor, there's zilch that tells you what the ratings are, whether it was automotive qualified. If it's foreign, like, say, Japanese Denso stuff, many IC markings are from another planet anyway and sometimes don't even show up in any web search results. Just look here and tell me what the chips with "JAPAN" on the markings are. And surely "if you know electronics" you can immediately tell whether they are automotive or even industrial rated or not. LOL.
Conformal coating mostly protects from condensation. If it's absent then perhaps the manufacturers have figured out that condensation isn't a problem. It's perfectly doable to have a "bare" board that can run acceptably between 5% and 95% relative humidity. Those aren't picoammeters. A bit of leakage won't hurt them.
I recently got some HVAC service done for half an hour of labor. My wife got nice boobs and she's friendly with everyone. Win-win for everyone. The guys at the service place got a babe to look at. I have a babe to look at. We have more money for vacationing:)
With EBD and yaw stabilization, you won't be fishtailing, or at least your fishtail amplitude will be kept to a minimum physically possible under the circumstances. It's as simple as that. Many large planes would be almost unflyable without yaw damping as well. Somehow you don't see pilots up in arms against yaw dampers, do you? So what the fuck is wrong with people who think that yaw damping in a car is bad, but in a plane is good? This is old tech -- it has been around in analog form since the 40s. It's high time it hit the road.
There'd need to be a clear case that such an upgrade would be useful. You'd need new firmware qualified to make use of the faster processors. It may well be that, given the fact that all the other pieces were designed to work with those "slow" 486 CPUs, the upgrade wouldn't solve any problems. Or maybe not. It's not just that "NASA couldn't permit it" with an implied "because I said so". It was more likely that they didn't want to spend any more money on that particular thing.
So what, you expect automotive-grade semiconductors and passive components to have bright "dude, I'm automotive grade" stickers on them? Protip: they usually don't look any different. The differences, besides just components, are in the board layout, derating of components, and EMC.
The modules you refer to were most likely built using caps that came from the infamous vendor who stole some trade secrets from their competitor, but not quite all of the needed ones. The missing part was what was the difference between good and bad caps. Country of manufacture got nothing to do with it.
Not only that, but on top of those 60C you'll have another 30-50C you must add for the actual die temperature of the hottest semiconductors. Those thermal resistances aren't going to vanish just because it's hot out there.
You haven't gotten the memo that the functions you all bundle together run on their own, individually hardened (or not!) CPUs, in their own dedicated assemblies. You don't want the SRS or ABS/traction modules sharing the guts with anything else. They have their own ride-through power, suitably sized, and don't have any unnecessary busses on the processor chip.
Even in a large city, nobody runs around patching things for a conditioned dry pair, I wouldn't think. You get a last mile T1 pair, this is then bundled digitally with a bunch of other stuff, hits the switch, get sent off to the other endpoint, eventually is unbundled and hits the last mile. It goes through a switch.
The key is: the switch is old. I don't think anyone even still makes switches with analog switching fabric - maybe as spares, maybe small PBXes are like that. The last few switches I've seen, serving more than 50k customers each, didn't have anything analog. The relatively few lines that terminated directly at the switch all went to line cards that had analog on the front, digital on the back. Everything else came in via DS trunks, some copper, some fiber.
There's no such thing as a private point-to-point T1 unless you have laid your own cable/fiber. When you nominally get a "private" "point-to-point" T1 circuit from any major player, all that happens is that there is a virtual circuit set up for for you in a big network somewhere. Your traffic may be cross half the country back-and-forth for all we know.
Oh you're just silly. We just number people the right way and it'll fit on a post-it. One readable without any magnification except what's in your eye. Suppose there are X evil people, and Y not evil people. We number everyone evil between 0 and X-1, and everyone not evil between X and X+Y-1. Problem solved. You only need to write down the value of X. No need to store any bits. The first X bits would be 0, the next Y bits would be 1. Easy-peasy.
There's no uncompresed 22kHz audio anywhere in the phone network system. You simply budget 64kbits/s for one direction in a voice call, that's also called DS0. That's what the analog last mile gets converted to and hauled as worldwide. It's 8 ksamples/s, at 8 bits per sample, using nonlinear A-law or mu-law lookup table. Every fax or modem connection gets hauled that way as well, and it works by design:)
There's no extra need to convert anything to data. Today's phone communications are all digital in the backhaul whether you like it or not. Analog is only the last mile.
Where it doesn't work, some greasemonkey magic is all you need. Sometimes that magic is substantial (1000s of lines), but I've got sites that are IE-only by design and rely on IE APIs to work on both Safari and Firefox. I've even re-implemented some ActiveX controls using plain old javascript. Given the amount of effort (a couple weeks in the evenings) by someone who doesn't do such a thing very often, I think that the site developers should be publicly shamed. As in rotten tomatoes or eggs thrown at them in the middle of the city square, or something like that.
I'm making assumptions based on what you can actually buy or have custom made. You are referring to stuff that doesn't exist. I for one would like our tires to last at least 40,000 miles and survive a few curb hits.
Slashdot Mirror
the ignition module is installed INSIDE the intake plenum valley - very harsh conditions
Huh? It's there for cooling. It's way less harsh in there than it'd be just outside of it. If it's bonded to a metal mass, it'd also be subject to slower temperature changes. Sometimes those engineers aren't as braindead as it might seem, you know.
You use electromechanical controls. That piezo sensor generates voltage, you can use it to drive a sensitive amplifier valve in a pneumatic circuit. Yes, there would be some mass increase, but it all depends on how fancy you get with computer modelling and design of this stuff. We can pull of electromechanical assemblies today that would be nigh impossible to design and manufacture even 60 years ago. You can have perfectly adequate electromechanical amplifiers even for O2 sensor output.
Now they are out of targets where search space scales linearly with findings. They have an exponential decay in findings to fight with, and you can't with with an exponential by brute force.
Speak for yourself. I'd be the first in line to have such a biopsy done, were it to be routine in differential diagnosis for, say, depression. These days they have the anesthesia/analgesia down pat, one would have to be truly crazy to reject an objective measure just because there's a biopsy involved. I've had my septum straightened out under very local analgesia just to skip school. And I walked to and from the hospital uphill both ways in a blizzard :)
And of course if you look at a random electrolytic, you can immediately visually tell what grade it is, ha ha. Yes, sometimes you can. Most often you can't. What I find interesting is that many automotive qualified parts that I use, from legit sources, often have less verbose markings than more run-of-the-mill commercial parts. Many commercial electrolytics have the temperature rating printed on the case, none of the automotive parts I've used (not that many, admittedly) had such markings.
That's assuming that you can make sense of component markings, and this applies to large-ish ICs only anyway. Small packages (a couple mm across) can be marked with a couple alphanumerics only. Nothing to tell you what grade they are. Zilch. Nada. Same if you look at a ceramic resistor or a ceramic capacitor, there's zilch that tells you what the ratings are, whether it was automotive qualified. If it's foreign, like, say, Japanese Denso stuff, many IC markings are from another planet anyway and sometimes don't even show up in any web search results. Just look here and tell me what the chips with "JAPAN" on the markings are. And surely "if you know electronics" you can immediately tell whether they are automotive or even industrial rated or not. LOL.
Conformal coating mostly protects from condensation. If it's absent then perhaps the manufacturers have figured out that condensation isn't a problem. It's perfectly doable to have a "bare" board that can run acceptably between 5% and 95% relative humidity. Those aren't picoammeters. A bit of leakage won't hurt them.
Never mind the foul-smelling polluting exhaust of most american models even from the 70s. Yuck.
It would be certainly possible, just that it'd be way more expensive than modern electromechanical controls.
I recently got some HVAC service done for half an hour of labor. My wife got nice boobs and she's friendly with everyone. Win-win for everyone. The guys at the service place got a babe to look at. I have a babe to look at. We have more money for vacationing :)
With EBD and yaw stabilization, you won't be fishtailing, or at least your fishtail amplitude will be kept to a minimum physically possible under the circumstances. It's as simple as that. Many large planes would be almost unflyable without yaw damping as well. Somehow you don't see pilots up in arms against yaw dampers, do you? So what the fuck is wrong with people who think that yaw damping in a car is bad, but in a plane is good? This is old tech -- it has been around in analog form since the 40s. It's high time it hit the road.
I've been in a side collision and the side airbag was all that stood between me and a good concussion.
There'd need to be a clear case that such an upgrade would be useful. You'd need new firmware qualified to make use of the faster processors. It may well be that, given the fact that all the other pieces were designed to work with those "slow" 486 CPUs, the upgrade wouldn't solve any problems. Or maybe not. It's not just that "NASA couldn't permit it" with an implied "because I said so". It was more likely that they didn't want to spend any more money on that particular thing.
So what, you expect automotive-grade semiconductors and passive components to have bright "dude, I'm automotive grade" stickers on them? Protip: they usually don't look any different. The differences, besides just components, are in the board layout, derating of components, and EMC.
The modules you refer to were most likely built using caps that came from the infamous vendor who stole some trade secrets from their competitor, but not quite all of the needed ones. The missing part was what was the difference between good and bad caps. Country of manufacture got nothing to do with it.
Not only that, but on top of those 60C you'll have another 30-50C you must add for the actual die temperature of the hottest semiconductors. Those thermal resistances aren't going to vanish just because it's hot out there.
You haven't gotten the memo that the functions you all bundle together run on their own, individually hardened (or not!) CPUs, in their own dedicated assemblies. You don't want the SRS or ABS/traction modules sharing the guts with anything else. They have their own ride-through power, suitably sized, and don't have any unnecessary busses on the processor chip.
Even in a large city, nobody runs around patching things for a conditioned dry pair, I wouldn't think. You get a last mile T1 pair, this is then bundled digitally with a bunch of other stuff, hits the switch, get sent off to the other endpoint, eventually is unbundled and hits the last mile. It goes through a switch.
The key is: the switch is old. I don't think anyone even still makes switches with analog switching fabric - maybe as spares, maybe small PBXes are like that. The last few switches I've seen, serving more than 50k customers each, didn't have anything analog. The relatively few lines that terminated directly at the switch all went to line cards that had analog on the front, digital on the back. Everything else came in via DS trunks, some copper, some fiber.
There's no such thing as a private point-to-point T1 unless you have laid your own cable/fiber. When you nominally get a "private" "point-to-point" T1 circuit from any major player, all that happens is that there is a virtual circuit set up for for you in a big network somewhere. Your traffic may be cross half the country back-and-forth for all we know.
Analog POTS is just the last mile. It gets converted to a digital form before it hits the switches.
Oh you're just silly. We just number people the right way and it'll fit on a post-it. One readable without any magnification except what's in your eye. Suppose there are X evil people, and Y not evil people. We number everyone evil between 0 and X-1, and everyone not evil between X and X+Y-1. Problem solved. You only need to write down the value of X. No need to store any bits. The first X bits would be 0, the next Y bits would be 1. Easy-peasy.
There's no uncompresed 22kHz audio anywhere in the phone network system. You simply budget 64kbits/s for one direction in a voice call, that's also called DS0. That's what the analog last mile gets converted to and hauled as worldwide. It's 8 ksamples/s, at 8 bits per sample, using nonlinear A-law or mu-law lookup table. Every fax or modem connection gets hauled that way as well, and it works by design :)
Uncompressed voice data going on phone networks is in a fixed format worldwide anyway. 64kbps per voice channel. It's been like that since forever.
There's no extra need to convert anything to data. Today's phone communications are all digital in the backhaul whether you like it or not. Analog is only the last mile.
Where it doesn't work, some greasemonkey magic is all you need. Sometimes that magic is substantial (1000s of lines), but I've got sites that are IE-only by design and rely on IE APIs to work on both Safari and Firefox. I've even re-implemented some ActiveX controls using plain old javascript. Given the amount of effort (a couple weeks in the evenings) by someone who doesn't do such a thing very often, I think that the site developers should be publicly shamed. As in rotten tomatoes or eggs thrown at them in the middle of the city square, or something like that.
I'm making assumptions based on what you can actually buy or have custom made. You are referring to stuff that doesn't exist. I for one would like our tires to last at least 40,000 miles and survive a few curb hits.