Category: Analysis

Anything that I have tested empirically, or explain with detailed calculations, or just generally bother to get all the facts straight.

Post #1922: Venn Diagram of Used Chevrolet Bolt Search.

A:   Not a salt-belt car.  Turns out, the majority of used Bolts for sale here in the DC area were sold new in the Northern U.S.  The cold isn’t the issue.  Multiple winters of driving on heavily-salted roads is the issue.  I don’t want a salt-belt car.

B:  No accidents, no obvious damage, no ludicrously excessive mileage.  I think the rationale there goes without saying.

C:  Dealer is not obviously a shithead.  And here, I’m not talking about the comments on Yelp (because those are always negative).  I’m just looking for a dealer where the majority of comments, on some mainstream site (e.g., cars.com), do not start off with some variation on “If I could give them negative stars … “.

As I sift through what’s listed within 25 miles of me, on Edmunds.com, this is how it shakes out.  This is what mathematicians call an over-determined system of equations.  Nothing satisfies all the constraints.  Or, the intersection of the areas is a null set.  Say it any way you like.

If I continue to pursue this, something’s going to have to give. At this point, I’m leaning toward buying a salt-belt refugee, from a seemingly decent dealer.  I mean, seriously, how much damage could three winters on salted roads cause?  Guess I may find out.

To be continued.

For the literal-minded of you, no, that’s not a proper Venn diagram of the situation. Some circles should overlap others, somewhere.  But it doesn’t look like a surprised face then, does it?  With the choice between literal mathematical correctness on a throw-away diagram, and some possible humor value, I went with attempted humor.  So sue me.  This is really more an expression of frustration over what ought to be a straightforward search for a commodity product.  But isn’t.

Post #1919: Salted Leafs and Bolts, an unexpected twist in my search for a used EV.

 

I’m in the process of narrowing down the used EVs I want to look at.

I just got a rude, but entirely logical, surprise.  It turns out that a lot of the late-model used EVs for sale in this area are salt-belt refugees.  That is, they were sold new in northern states, where they salt the roads heavily all winter long.  But were shipped south for re-sale as used vehicles.

The story.

Based on the ratings of car dealers on-line, I’ve focused on a couple of independent used car dealers in my area.  (FWIW, Kingstowne and Eastern’s Sterling).  I think maybe the phrase in italics is key, because these aren’t new-car dealers accepting trade-ins.  These are used-car dealers.

As I was doing my on-line due diligence, seeing what I can see about these cars by VIN, I happened to notice that one car I’m interested in — a 2021 low-mileage Bolt — was originally sold in Michigan.

Hmm.  Funny that this car ended up in Virginia.  But people move, and so on.  And yet …

I tracked down the original state of sale for the other two I’m focused on — 2020 and later, relatively low mileage.  Those were originally sold in Upstate New York, and Vermont.

One salt-belt car might be by chance.  But every car I’m looking at?  Highly unlikely that’s a coincidence.

I can guess what’s driving this.

EVs lose a lot of range in cold weather.  That’s a fact.  None of these cars has an efficient (heat-pump) heating system.   Also a fact.

I have to guess that:

  1. You have a lot more dissatisfied owners in cold-climate states.
  2. You get a much better resale price on these vehicles, in warm-climate states.
  3. So there’s a steady trade in shipping used EVs south for resale.

The issue isn’t that these were driven in the cold.  The battery management systems on these cars will all prevent the owners from damaging the batteries permanently by (e.g.) charging when the batteries are below 32F.

The issue is that all of these cars are salt-belt refugees.  That is, they were driven in the states where roads are heavily salted, for a significant fraction of the year.

After a few months of watching YouTube auto mechanics in salt-belt states (Watch Wes Work, from Illinois, and South Main Auto Channel, from upstate New York), one thing that comes through loud and clear is that salt is incredibly destructive.  Among the things I learned from those videos is the term “rust jacking”, which is when the accumulation of rust literally bends and breaks metal parts of the car.  Never seen that around here, and I’ve owned a lot of crappy old cars.

And so, once again, I need to stop and cool my jets, as I give this a re-think.  And look at what’s available as a used vehicle, from local new-car dealers.

Post #1918: Falling Leafs, fallen Bolts: The trend in used EV prices in my area.

 

I don’t drive much.  I haven’t had a car for a couple of years now, and have gotten along  by borrowing my wife’s car, when convenient.

I’d like to get my hands on a nice, used EV.   That’s a good choice, given that I’m going to use this for a grocery-getter and little else.

Depending on the price, of course.  And I’m clearly in no hurry to buy one.

Back in July I looked at my local market for used EVs and narrowed my best option down to a 2018 or later Nissan Leaf.  That’s laid out in a series of posts around Post #1837, and the posts just prior to that.  The year cutoff was due to a change in the Leaf battery chemistry that year, to a much more stable (long-lived) battery.

I have been checking back occasionally ever since.

And I’ve been reading articles suggesting a steep decline in the price of used EVs.  I see talk about price declines on order of 30% per year.   This is almost always attributed to the fact that most used EVs are Teslas, and Tesla made some steep price cuts to their models this past year.

In other words, a falling tide sinks all boats.  Those Tesla price cuts are rippling through the entire used EV market.

But in addition, Chevy cut the price on the Bolt last year.  Both to spur sales, and maybe because the Bolt was plagued by a significant recall due to battery fire issues in a handful of vehicles.  Chevy claims that’s taken care of, but they ended up replacing the batteries in tens of thousands of cars.

In any case, when I went back to re-assess my local market for used EVs, it sure did seem like prices were down.  So I did my best apples-to-apples comparison between what I looked at back in August, and now.  As shown above.

By my estimate, asking prices for a used late-model Nissan Leaf fell 14% in the last five months of 2023.  Or … on-order-of a 30%/year rate of decline. 

More interestingly, I can now get a used Chevy Bolt for about the same price as a used Nissan Leaf.  This is a change from the prior analysis, where my back-of-the-envelope on a Bolt of this vintage, five months ago, put the average asking price at $21,000.

But now, consistent with the decline in the Leaf price, there’s been an even steeper decline in the Bolt price.

Objectively, the Bolt looks like a lot more utility for the money.

  • The Bolt has about 90 more miles of range than the leaf (about 250, versus about 160 for the base Leaf)
  • It uses a standard (J1772) plug, instead of the soon-to-be-obsolete CHADMO plug on the Leaf.
  • It has active battery temperature management, compared to the Leaf’s passively air-cooled battery.

The sole drawback from my perspective is that the Bolt looks like a tiny little car, where the Leaf does not.  To me.  They have roughly the same interior volume, and the Bolt actually has a higher curb weight than the base Leaf.  But the Bolt is shorter by about a foot-and-a-half.  Just enough that I notice how small it is, compared to (say) the 2021 Prius that my wife drives.

For either car, if you had little enough income in the year of purchase, Uncle Sugar will give you a $4K tax rebate for purchasing that used US-made EV.  (Yep, for purchasing a used US-made EV.  Part of the Biden Administration’s buy-American industrial policy intersecting with its global warming initiatives. So, thanks, Joe Biden. I guess.)

Rumor has it that the big drop in the Bolt price is due to Chevy rehabbing and re-selling a lot of those recalled vehicles.  I’m not sure how much that is true.  What I am sure is that the Bolt looks like a pretty good option, if you trust Chevy to have fixed that rare battery issue.  If you pick and choose, you can plausibly pick up a three- or four-year-old car, with about 10K miles on it, for a net $13K or so.

This, where the only expensive component — the battery — comes with a mandatory eight-year/100,000 mile manufacturer’s warranty. Which should, in theory, take a whole lot of the risk out of this used-car transaction.  Roughly speaking, you pretty much have to get at least five years of driving out of the car, or the manufacturer (not the seller!) has to replace your battery.

As used cars go, that seems like a pretty decent deal, regardless of the fuel source.  The fact that this is the low-carbon alternative is almost gravy, at this point.   To me, based on what I’ve been looking at, this now looks like it’s just a pretty good deal on a used car.  Period.

I have to confess that the first and last Chevrolet product that I ever bought was a Chevy Vega.  It was a traumatizing experience in many regards, as those of you familiar with the history of the Chevy Vega will understand.

I guess, going on 40 years later, maybe I can find it in my heart to forgive, and give Chevy another try.

Post #1911: LED Christmas light life expectancy.

 

This post goes way over the TL;DR line.  If you want to get to my summary on buying LED Christmas lights that will last a while, go to the Conclusions section in red, below.

Source:  Except where noted, images in this post are from the Gencraft.com AI with a prompt of “Christmas lights”.

Intro:  The ghost of Christmas lights past.

My parents had the same sets of Christmas tree lights for my entire childhood.  And then some, given that I was the youngest of four children.

I, by contrast, am getting ready to toss (recycle) yet another couple of strings of dead Christmas lights.   In this case, some elderly miniature incandescent light strings that started off the season dead.  Again.  And for which I am finally throwing in the towel. Continue reading Post #1911: LED Christmas light life expectancy.

Post #1909: Never eat at a place called Mom’s.

 

And don’t trust Grandma.  Or the Amish.

This evening my wife strongly hinted that I ought to replace the broken jar of black raspberry jam that was the focus on my just-prior post.  No fool I, I immediately got on-task.

So this is my second attempt to buy her some black raspberry jam, as a Christmas present.

This time I looked at every black raspberry jam offered on Amazon.

After my nth jar of jam, I came to a firm conclusion:

The folksier the name, the lousier the product.

Grandma’s Old-Fashioned Amish® homemade black raspberry jam.

You typically can’t find black raspberry jam at the grocery store.  The berries themselves are small and fragile.  That makes it an expensive crop to grow, per pound, compared to other berries.  As a result, black raspberry jam is typically priced at several multiples of (e.g.) strawberry jam.  And I guess that’s a non-starter for the grocery store shelf.  Which is why I’m ordering it off Amazon.

But not all jams are created equal.  They are some combination of fruit, sugar, water, pectin, and maybe an acidifier like lemon juice.  And sometimes other stuff.  Some are primarily fruit, with just enough of the other ingredients to sweeten it, hold it together, and keep it from spoiling.  Others are closer to berry-flavor sugar.

All I wanted, in my first cut, was to restrict this to jams for which black raspberries were the first-listed ingredient.  Ideally, I’d like jams where they were the majority ingredient, but unlike some European labeling, U.S. labeling law does not reveal percentages.  The U.S. simply requires that ingredients be listed in order of weight.  So I’ll settle for jams where sugar and water are listed after the berries.

(That’s assuming I could actually find the ingredients listed somewhere on-line.  It was surprisingly common to find jams listed on Amazon, but never showing the legally-required list of ingredients.)

So here’s a little quiz.  Based on the look of the jar, which of these do you think have the berries as the first-listed ingredient?

Source:  Amazon.

The trick is to toss out anything that says “Grandma”, “Amish”, or “Homemade/Homestyle”. Those are all the lower-quality jams where sugar outweighs berries.

We all know that none of these were actually made by Grandma, at home.  (Then sold in huge quantity on Amazon.)  The fact that the makers felt compelled to call them “Grandma’s Homemade” should have been a clue that they were compensating for something.

But as for the Amish, this is not to imply that the Amish make bad jam.  The issue is that “Amish” isn’t trademarked.  Anybody can make anything, anywhere, and label it “Amish.”  In fact, the last product (Kauffman’s) is made in Bird-in-Hand, PA, and so plausibly actually is Amish- (or maybe Mennonite-) made.  It just doesn’t try to sell the product based on that association.

Anyway, the ones with black raspberries as the first-listed ingredient are below.  Only for the two French ones (Chantaine, St. Dalfour — really, the same company) can you tell that black raspberries make up 51 percent of what’s in the jar.  For the rest, all you know is that there’s more black raspberry than there is of any one other ingredient.

Conclusion

Never eat at a place called mom’s.  And never buy jam made by Grandma.

Post #1894A: A minor technical followup on the NY Times/Siena poll results

I’m still looking for loopholes.  Hence, three remaining questions:

  • How was the sample selected, and in particular, did it require a successful match from voter record to cell phone record?
  • What was the overall response rate?
  • How well does this benchmark with the actual 2020 results?

L2 file?

After reading the end-notes on the detailed tabulations of the NY Times/Siena College poll, my main remaining question is:  What is the L2 file?

Survey respondents were chosen (in a sophisticated-but-neutral way) from persons on the L2 file.  That file is the “universe of observations” for the survey.

Based on the U. Penn description, the L2 file contains public information on about 200M persons who recently voted.  And, about 95 million cell phone numbers.

The file itself was developed by L2.com.  Having dealt with mailing-list vendors before, I recognized much of the subsidiary information that they merged onto the publicly-available voter records.

But if that’s an accurate description —  95M cell phones, 200M voters — then roughly speaking, a bit less than half the L2 file had phone numbers attached to the voter data.

Did this survey draw from persons on the L2 file who had a phone number listed?  Or did it draw from all persons on that file.  The documentation simply says:

The survey is a response rate-adjusted stratified sample of registered voters on the L2 voter file.

I’m pretty sure they meant response-rate-adjusted, that is, they adjusted the likelihood of being sampled based on some prior estimate of likely non-response rate.

In any case, if the U. Penn description is correct, then this is a valid question to ask. Along with the obvious followup:  If it’s persons with listed cell phone, could that matching process — the process that added the cell phone number to the voter record — possibly have induced a bias?

Response rate?

The other thing not stated was the response rate.  They said that 94% of the people they called “were reached” on the phone.  Like this:

 Overall, 94 percent of respondents were reached on a cellular telephone.

But you’re left guessing as to what the actual response rate was.  At least, as far as I could tell, from the documentation cited above.  (The “reached” figure speaks more to the validity of the added phone data, than to the response rate. You can reach me, and I can say “no thanks”.)

Don’t people lie (on average) about how they voted in past elections?

That said, the big advantage this survey has is that it shows a modest win for Biden in these states, in 2020.  That is, it corresponds to the actual 2020 results.

Whatever their methodology goes, it accurately shows that Biden won the popular vote, by a small margin, in 2020.  It’s hard to say that the 2024 projection is hugely biased in some fashion, when you can see that no such bias exists for the actual 2020 results (as estimated from this poll).

Then I got to wondering:  Don’t people lie, after the fact, about having voted for the winner?

The problem is that if I Google anything near that topic, all I get is stuff about the 2020 election.  So any answer to whether or not this is material — if people tend to say they voted for the winner — will have to wait until I figure out some better way to find an answer to that.

 

 

Post #1894: Commentary on the NY Times/Siena College poll results.

 

I find myself grasping at straws, trying to explain away the NY Times/Siena College polling results showing Biden soundly losing to Trump in 2024.

This survey predicts Trump taking five out of six swing states in 2024: Arizona, Georgia, Michigan, Pennsylvania, Nevada.  But not Wisconsin.  Source:  This

Having spent some time down in the details, let me summarize:  A poll that correctly reproduces the 2020 Biden win (in six swing states) now calls for a big 2024 loss. 

Worse, it’s a good poll.  By which I mean, a well-executed poll.  I saw almost nothing in methods that I strongly disagreed with.  (And I used to be in the statistics biz.)

Insert rambling detail here

I have to admit that I woke up just totally pissed off about this poll. After my wife clued me in on it yesterday.  Her response to the results was “people suck and I hate them all”.  Seems like a valid viewpoint.

I just plain wanted it to be wrong.  That’s not science.  I looked for obvious errors, and didn’t see any.  So far.  FWIW.  It’s the gold standard — the best available estimate of how these swing states are likely to vote.

Here’s my take on the main message:

Biden’s too old. 

And other stuff, sure.

Weirdly, the main writeups seem to skirt this issue.  But to my eye, this is something that everybody agreed on.

Separately, smears work, disinformation wins.  Seemed like more than half of everybody think Joe Biden’s dirty, and has taken payments from China and Ukraine.   Which, as far as actual evidence goes?  In any case, one President makes his tax returns public, one does not, I’m gonna stick with the one who does.

Yes but.

And now, from the Democratic side of things, comes a string of “yes, buts.”

There’s some nuance to it, but I think I can boil them down as:

Yes, Biden’s too old.  But if my only alternative to Biden is Trump, then “too old” doesn’t exist.  If Biden’s breathing, I’m voting for him.

And I have some reasons for preferring Biden.  In no small part, it’s fair to expect Biden to assemble a far more competent team than Trump.  Fewer cronies, fewer toadies.  Fewer of his own children, for that matter.  And for sure, with Biden, we’ll likely have fewer Chairmen of the Joint Chiefs mockingly threatened with execution.

So you can “yes, but” it to your heart’s desire.  Yes Biden’s 80.  But Trump’s 77, fat, and his dad died following a period of dementia.)  Yes, Biden sometimes does old-guy stuff, but Trump rarely utters a coherent sentence.  If I gotta listen to one old fart ramble, please let it be Biden, and not Trump.

And there’s that whole fate-of-the-Democracy thing.  We got one joker in the Senate, saving up military appointments.  Hell, what worked for McConnell for the Supreme Court sure ought to work for the military.  Given how much Trump admires dictators, the idea of a military run top-to-bottom by Trump acolytes does not appeal to me.

But he’s stuck.

And I mean Joe Biden.

Is there any way that Biden could withdraw from the race?  I’m not seeing it.

Wouldn’t he then be obliged to support his vice-president, as the Democratic presidential candidate?

Do you think that America is ready to vote for a Black woman, to be President?  Separately, do you think Harris is a good candidate?

My answer is no and no.  I don’t see Harris as a viable winning Presidential candidate.  So Biden’s stuck there.  If he drops out in a normal and reasonable fashion, then the Dems lose in 2024.

So he can’t agree that he’s too old.  Even if he thinks he is.  And he can’t drop out, for that or any other reason.  He options are to fight one more election.  Or to lose.  No reasonable person can expect Biden not to fight for it.

Post #1893: Winter, firewood, soot, Canada

 

 

Normally, about this time of year, I’d start burning my way through two cords of wood, over the course of the winter.

This year, I’m not.

It’s complicated.

Soot uncertainty.

I went through the biggest global environmental problem in heating with wood back in Post G22-058.

In a nutshell, when I burn firewood for heat, the C02 that goes up my chimney came out of the air an average ten years ago.  For that reason, firewood is very close to a carbon-neutral fuel, when viewed over (say) a decade of time.  Over that time period, atmospheric C02 is neither increased nor decreased by the process of growing wood, then burning that wood.

As opposed to say, burning natural gas.  Typically, that was produced some time in the last half-a-billion years or so, and trapped underground.  The C02 from that source definitely adds to the current level of atmospheric C02.

But along with wood burning comes soot.  And even though that soot resides in the atmosphere for just a brief period (typically, two weeks), soot is incredibly effective at capturing the heat from the sun.  Dispersing a microscopic black powder through the atmosphere allows the atmosphere to absorb more light energy?  Who would have guessed that?

Source:  Myhre, G., D. Shindell, F.-M. Bréon, W. Collins, J. Fuglestvedt, J. Huang, D. Koch, J.-F. Lamarque, D. Lee, B. Mendoza,
T. Nakajima, A. Robock, G. Stephens, T. Takemura and H. Zhang, 2013: Anthropogenic and Natural Radiative Forcing. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United
Kingdom and New York, NY, USA.

Back in 1995, nobody quite knew what the net effect of soot was.  Even through 2014, estimates were uncertain enough that the confidence interval around the point estimate included zero.

That said, you have to go with the most recent evidence.  Based on the 2014 5th report (AR5) of the Intergovernmental Panel on Climate Change (IPCC), I estimated that the warming effect of the soot from my wood stove was just about large enough to offset any benefit from wood burning.  That’s the gist of in Post G22-058.

And that’s why I skipped the firewood purchase this year.

The IPCC sixth summary report was released in its entirety earlier this year.  So it’s worth taking a peek at that, as the estimate for black carbon involved a lot of uncertainty.

(First, though I have to note how different the public debate is now, for the IPCC 6th report, compared to nine years ago, for the 5th report.  For the IPCC 5th report, climate-change denialists went over it with a fine-toothed comb and found an actual substantive error, in a sentence, in a section of the technical portion of the report.  This had to do with the rate of melting of Himalayan glaciers. And, as is their habit, the climate-denial industry then proceeded to play the game of This Changes Everything,  So Believe Nothing You Have Heard.  For the sixth report, by contrast, the release was uneventful, and nobody tried to fabricate some made-up stink about it.  It’s almost as if everyone with sense now realizes that climate change is real, man-made, and causing problems.  And so there is little value in trying to generate new disinformation, because those who still deny that climate change is a real threat are more-than-satisfied with continuing to believe disinformation that was debunked decades ago.)

Interestingly, they’ve revised their estimate of the warming impact of soot way downward, compared to the 2014 report.  (Though still within the 95% confidence interval of the 2014 report).

The best estimates of ... attributed to ... black carbon is substantially reduced. The magnitude of uncertainty in the ... due to black carbon emissions has also been reduced relative to AR5. (Section TS.3.1)

Source:  Page 42, IPCC, 2021: Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2391 pp. doi:10.1017/9781009157896.

And, in fact, it looks like the estimate of the warming impact of soot is back about where it was in 1995.  Which is about when I decided it would minimize my global warming footprint if I burned wood for (at least some of) my home heating.

Source:  Page 92, reference cited just above.

Without getting further into the details, at face value, the upshot of this most recent change is that heating your house with wood, in a modern air-tight low-emissions wood stove, probably offers significant net benefit in terms of global warming footprint.  Based on this most recent estimate of the impact of the resulting soot.  Assuming I did my calculations correctly last year.

Local air pollution

This year’s air pollution alerts from Canadian forest fires have made me a lot more sensitive to the issue of air pollution from wood fires. In the past, I’ve just turned a blind eye to that, mostly because as far as I can tell, I’m the only person within blocks that actually burns wood for heat.

Air pollution from an isolated wood stove does not have the same public health implications as air pollution from Canadian forest fires.  That’s because you have to be in the exhaust plume from my stove to be affected by it.  By contrast, you were breathing Canadian soot no matter where you were, and no matter when.  It’s the difference between a brief exposure, walking past my house (say), and breathing it 24 hours a day.

So, really, it’s more a question of what I’m doing to the air that I and my neighbors breathe.  And for that, the key question is how particulates generated by my wood stove, at my property line (i.e., entering the public domain) compare to the particulate levels we saw during the Summer 2023 air pollution alerts?

I suspect that the only way to tell, with this one, is to measure it.  Which I will, the next time I light a fire in my stove.

For now, let me work through the basics, given that this stove is EPA rated to produce no more than 2 grams of soot per hour.  A good round number for “too much soot in the air” is, say, 100 micrograms per cubic meter for total particulates.  That would trigger an “unhealthy” reading for PM 2.5.  To get down to that level, an hour’s worth of soot from my stove would have to be diluted into … 20,000 cubic meters of air per hour.

Or about 333 cubic meters of air per minute.  In order to dilute the smoke from my wood stove down below the “hazardous” level for particulate matter.  That’s a cube of air roughly 18′ on a side.  That seems like a high-but-plausible rate of dilution.

There is also a sense that if you can smell wood smoke, you are breathing in pollutants.  And that may well be true — the smell must come from somewhere.  That said, a quick look at some scholarly papers suggests that there isn’t a tight correlation between the smell of smoke and the density of particulates in the air.  (As evidenced, I guess, by the Canadian forest fires, where there was no smell of wood smoke in the air, but particulate levels were high).

So, before I even lay my first fire of the season, and get out my recently-purchased air quality meter, I’m guessing that this is an open question.  I can surely smell wood smoke, at ground level, at least part of the time that I’m buring wood.  The next step is to measure it and see if I’m pushing unhealthy levels of particulates out into my adjacent neighborhood.

Post #1891: If the on-line deal seems too good to be true, what do you do?

 

At what point is an on-line deal so good that you decide not to buy it?

And if so, why?

In the modern U.S.A., with markets dominated by cheap Chinese goods, is there still any such thing as a price that’s too low to be believable?

I have to write this one fast, as this amazing deal I’m looking at won’t last long.  I must order now, or I might miss out on the deal of a lifetime.  Continue reading Post #1891: If the on-line deal seems too good to be true, what do you do?

Post #1867: So, as I was changing a tire yesterday, the car fell off the jack …

 

Man, that’s embarrassing.

In order of importance:

  • All body parts and car parts remain attached.
  • It’s a piece of cake to recover from this error.
  • I was, in hindsight, being a complete dumbass.

The dumb-assity of it was clear the instant the car hit the ground, but it took a while to figure out the “complete” part.  This post explains why the car inevitably fell off the jack.  Because it wasn’t obvious to me, even after it happened.

 

1:  Easy recovery.

Just jack it up by the other end.

On a modern unibody car (with no separate frame), with a standard scissor jack, you have to place the jack under the car, directly below a ridge of sheet metal.  Then make sure the jack end straddles that ridge, and proceed to jack up the car.

That means that if your car falls off the jack, with the wheel off, the brake rotor ends up on the ground, and there isn’t enough room to put the jack back under the car.

I literally couldn’t figure out what to do.  Pry it up with a 2×4?  Buy a floor jack and put it under the front suspension?

Then I looked up this YouTube video:

The answer is, jack the car up by the other end.  Turns out, that will raise the entire side of the car.  In my case, I was changing a front tire.  After chocking the wheels, I jacked the car up at the rear jack point.  This raised the front frame enough for me to put a brick under it.  I then let the car down, moved the jack to the front, and proceeded as if nothing had happened.

No harm done.

But why did it fall?

The parking brake wasn’t set.  I should have checked that, but … stuff happens.  Edit 1/27/2024:  In hindsight, the parking brake wasn’t working, something I only found out after yet another flat on that car.  I had the good sense to chock the wheels the second time around.

Even so, the car was in park, on level smooth pavement.

Why didn’t putting the car in park keep it from rolling?

I had the car in park, and one front tire was firmly on the ground.  Why didn’t this hold the car in place?

Turns out, on a front-wheel-drive car, park locks the output shaft of the transmission.  But it doesn’t lock the differential — the gizmo that splits the power between the two front wheels, and allows those wheels to spin independently as you go around a curve.

Because of that, if one front wheel can turn, then both front wheels can turn.  They turn in opposite directions, through the differential.

The upshot is, on a front-wheel-drive car, once you have one front wheel off the ground, there is nothing about “park” that holds the car in place.

Apparently, in northern climates, it’s well-known that if you park with one wheel on ice,  in a front-wheel-drive car, the car may slide even though the other wheel is on dry pavement.  That happens for the same reason.  As long as one tire can rotate, both tires can rotate, even if the car is in park.

My bad.  Lesson learned.

But how did I generate enough force to move the car?

I mean, I placed the jack very nearly vertical.  To within a few degrees, anyway.  How could that possibly have generated enough sideways force to get the car to roll off the jack.

This car was in my garage, on a concrete slab that is very nearly level.  It weighs about 3000 pounds.  My gut feeling is that it should take quite a bit of force to get that moving.  Right?

Wrong.  Based on a quick check of several sources, the coefficient of rolling resistance of car tires is about 0.01 to 0.015.  That means it should take somewhere between 1% to 1.5% of the weight of the car, applied horizontally, to get the car rolling on a smooth surface.

Or in this case, somewhere between 30 and 45 pounds of force, horizontally, could be enough to get the car rolling on level pavement.

How far off did the angle of the jack have to be, to generate that amount of force?   Assuming I recall my elementary physics right, that goes as the sine of the angle away from vertical. In this case:

The upshot is that I’d have had to eyeball the angle of the jack to less than two degrees of vertical.  Off by any more than that, and the geometry and physics of the situation would generate more than enough force to roll the car.

(On top of this, the garage floor has a slight slant, such that the car already has an estimated 15 pounds of horizontal force on it, just from sitting on that slant.  And, by picking up (say) 1000 pounds of weight off the car, and onto the jack, I actually only had 2000 pounds on the wheels, so I only needed 20 to 30 pounds of horizontal force to get the car to move.)

Conclusion

Before this, I had the vague notion that the car might roll off the jack if I didn’t take preventive action.

Now, having done the math, I can say that my front-wheel-drive car, in park, on level pavement, will roll off the scissor jack if I don’t take preventative action when changing a front tire.

In hindsight, it wasn’t a slightly risky, maybe-I’ll-get-away-with-it maneuver.  Falling off the jack was a near-certainty, under these circumstances.

FWIW, I’ve added a pair of folding metal wheel chocks to the bag holding the jack and tire iron.  Presumably, if I literally have those in my hand, the next time I get a nail in a tire, I will have the good sense to use them.

I think.

Illustrations are from the Gencraft and Freepik AI sites.