Al Rentschler understands time.
With close to a lifetime spent in the clock manufacturing industry, over 30 years to be exact, you could call him a time expert.
The President of E.A. Combs Inc., a manufacturer and distributor of time pieces based in Lake Geneva, WI, has had a lifelong fascination with time, having assembled an extensive collection of clocks and other time pieces over the years.
“I’m not exactly sure how many I own, but it’s well over a hundred,” said Rentschler of his collection. “I started playing with time when I was just five years old.”
So Mr. Rentschler seemed like the perfect person to talk to in our quest to better understand the timing controversies swirling around Chicago’s new parking meter pay boxes which have sprouted up on city streets all over the city.
According to a statement on Monday from Avis LaVelle, spokesperson for Chicago Parking Meters, LLC, these new meters have their clocks updated every night via a cellular connection to a computer server.
“Pay boxes contact a central server every night wirelessly,” said LaVelle about the process the meters go through to retain an accurate time. “At that time, the pay box programs are updated and refreshed, and the pay box clock synchronizes with the server clock. The server clock is an atomic clock, and atomic clocks conform to the most accurate timekeeping standards.”
But according to Rentschler, their explanation does not add up.
“I doubt that it’s true,” says Rentschler about CPM’s atomic clock explanation. “If it were true they (the clocks) would all be right on the money. There is no reason to be off by even a few seconds.”
The problem is, the vast majority of the 60 Pay & Display meter units we’ve inspected over the course of the past few weeks, do not keep proper time and there seems to be no consistency to the amount of time each unit is off.
Rentschler explains that quartz controlled clocks are not perfectly accurate and concedes they can lose seconds and even minutes, over the course of time.
“Plus or minus two seconds per day,” says Rentschler. “That’s the maximum (time loss) allowable in the industry. State of the art clocks are in that range.”
So, if most of Chicago’s new high tech meters are chronically off by more than two seconds, how is it possible the new meter company is actually refreshing the clocks on their meter pay boxes every night?
“It’s apparent that they’re not,” Rentschler said firmly in regards to CPM’s claim. “That’s bogus . They (the units) would all be the same time in the morning if that were the case. A crystal should only lose one to two seconds per day.”
Jeremy Fischer agrees with Rentschler.
“Whatever their time source is, it is not working,” said Fischer, who does tech support for Franklin Instrument, his family’s 40 year old Pennsylvania-based clock business. “It sounds like something is not set up properly in the time configuration. You can tell by the large deviation in time. If this (nightly syncing) was actually happening, the times (of the clocks) would be off by just milliseconds.”
Fischer and Rentschler both believe that, because of the extreme deviance in time and the inconsistency of deviance between the different meters it’s impossible for CPM to be refreshing the clocks on their meters every night.
According to these two industry experts, without nightly external adjustment, each unit’s clock is left to the mercy of their inherently, but mildly inaccurate, quartz timing. Thus the wild time discrepancies between pay box machines.
But Fischer went even further in his analysis of Chicago’s meter timing problem.
“Based on the number of seconds the clocks are off, it makes me believe that these meters have not had their time adjusted since they were installed,” speculated Fischer. “If I got a ticket in Chicago I could beat it (easily).”
Fischer explained the longer a clock goes without a external time adjustment, the more extreme the time difference.
This would seem to explain why so many pay boxes in the Wicker Park neighborhood are so far off true time, to the tune of nearly two minutes in the case of four units we inspected.
That’s because some areas of Wicker Park were the first to receive these new units back in mid-April of this year.
Buffalo Grove resident Barnet Fagel is also baffled at the explanation given for the meter clock inaccuracies.
Fagel, who has 19 years experience in the GPS and tracking, an industry that is completely reliant on timing accuracy, is also Illinois’ Safety Advocate & Traffic Researcher for the National Motorists Association.
“How about a couple of nano seconds,” said Fagel when asked how much deviation a clock being synced nightly to an atomic clock server should have. “If the meter boxes get a time sync once a day from the atomic clock they shouldn’t be any humanly discernable discrepancy.”
“If they’re getting their sync from the atomic clock in Colorado, then it is strange their machines are off by so much,” Fagel elaborated. “If they are (syncing every night) it’s not an accurate sync. My point is, if they can make a $9 watch keep time and only lose a few seconds a year, why can’t a meter keep proper time?”
But Fagel thinks this meter timing issue requires further governmental scrutiny.
“Where is the regulatory standards (for these clocks)?” asks Fagel. “Shouldn’t they have to meet certain standards of accuracy? If you’re renting a metered parking space, the value of the space is based on the accuracy of the time.”
“Look at taxis and taximeters,” Fagel says. “The meter in the taxi must meet certain standards of accuracy. Why not these parking meters?”
Posted in 
Fantastic article. I love it when real experts can call out some lame-brained excuse.
Interesting article indeed. While Relativity prohibits absolute simultaneity, (a perfect system with receivers and one transmitter)inaccuracy would radiate out an exactly determinable and VERY SMALL time difference of mere microseconds. Light and radio signals travel at 186282 miles per second. A box 20 miles from the transmitter (say, on Sears Tower) would be SLOW by only 100 microseconds. And the city isn’t that big. Daley would be only gypped by mere tens on microseconds in the worst case.
Fun fact. When DOR employees with pre-set perfect clocks in this scenario, as they drive the clocks slow by some unmeasurable amount, again Daley getting gypped ever so slightly by Relativity!
For light to travel 2 minutes it covers over 22 million miles. It’s an obvious major malfunction on the part of the leasing company.
But this whole deal came down so fast that maybe Time slowed, or even stopped, for Daley, LAZ, and the Deptartment of Revenue, so that no amount of fixing will fix their clocks and they’ll always be behind. Or would they be ahead? And wouldn’t that work in our favor? I’ve never been able to comprehend Ralativity, so I’m out of my realm here.
What I do know is there are many who would love to fix Daley’s clock, or at least give it a good cleaning!
-John Adams
What I can’t figure out is 15 minute or loading zones.
(ie between 9 am- 6 pm ) do you then have to pay at the box or is it free parking. I see many people getting tickets at those locations.
The clock being slow on the meter gyps us assuming the DoR worker has an accurate clock. If the meter is accurate and the DoR worker’s clock is slow, Daley gets gypped. This assumes a laser stop device (like a dragstrip finish line) and you leave within the room of error. Consider this thought experiment:
The metered parking space has you in it a millisecond before the meter expires. You are about to launch within that millisecond. The DOR sets up the laser and turned it on waiting for your launch to be delayed by the meter’s millisecond to try to catch the violation. You blast off. If his clock is fast by a second, you’re caught before launch. Ticket. You were legally parked as you leave but get the bogus ticket.
If his clock is accurate matching the meter, you get no ticket as you leave with the less than one millisecond to spare.
If the meter is accurate and the Dor’s clock is the second slow, you launch with just less than a second of parking time as seen by the DoR. No ticket.
Fixing the clocks in realith will invariably mean they will always be a tiny bit behind (see above) so I stand corrected in that it’s us who get gypped by Relativity by the tiny amount. The thought experiment would involve a car with an impossible acceleration so the relativistic rip-off is virtually nonexistent.
Thanks for the great input guys!
I’m really interested in hearing what our PEA friends think of this little dustup and if they got any special instructions because of the problem.
Hey Jack!
Here’s the deal. Before, with the old single head meters, the loading zone was a free spot after or before the times posted.
Of course, during the hours listed, you could park for 15 minutes (hazards on) or so to load/unload, etc.
But now, when those hours are over, that loading zone spot is now considered a metered spot. It’s NOT a free spot anymore. Feed the meter or else!!!
You’re fully welcome, Parking Ticket Geek. If Daley or LAZ tries to blame Relativity, send them my two postings with the primer above.
Anonymous – speed of light has nothing to do with the accuracy. Any computer based time sync should be using NTP which takes that latency into account, so even though two clocks may be different distances from the source clock, it doesn’t matter and they should still have the same time within a nanosecond or two
Nanos is really hard to do, micros is more reasonable and millis is very easily done by ntp. For nanos you need to use something really advanced like precision time protocol or the like.
Is it possible these aren’t using a “clock” per se, but rather basing the time on the speed of the cpu inside the computer? That’s pretty common – usually it works okay, but clocks based on that can drift a decent bit, I think. Take a laptop or the like, take it off the network for a long time, and then hook it back up so it can sync. I’ve had them go off by a few minutes after a few open/close cycles. In some time intensive industries, it’s common to sync very frequently – I have an alert set up if mine are off by more than 10ms and am paged if they’re off by more than 50ms.
the laz spokesperson is a bold faced LIAR. she will now deny ever making those statements.
Don’t you feel time gaining on you?????
It’s like a predator; it’s stalking you…
Oh,you can try and outrun it….
but in the end, time is going to hunt you down…
They say time is the fire in which we burn…
My time is running out….
We leave so many things unfinished in our lives…..
I am aware of latency compensation. You need it in any multiclock timekeeping system. The optimum system for the boxes would have their server feed a transmitter and laa the boxes have receivers. The latency compensation that’s most obvious is receiving the original WWVB signal and processing it for the “WLAZ” time radio signal. You add that digital compensation (and encryption) to transmit.
Each box would then be set for miles from the “WLAZ” transmitter so all are in sync. While you would be hard pressed to get all of them to a nanosecond, you can get them to a handful of microseconds. Their cellular system is obviously now working, probably due to inadequate cellular modem capacity at the timeserver.
With an encrypted time signal you reduce hackers, like no doubt the cellular system tries to avoid. The advantage of my “WLAZ” system is the latency to the boxes is always the same. With cellular, the latency from box to server is variable to phone wire routing at the repeater. Plus, my proposed system would be less expensive. No cell phone bill.
Now, to address the PEA clocks, you issue atomic wristwatches (available at Walmart) eliminating inacuracy at street level. My point earlier is that speed-o-light delay is VERY small. They can’t use it as ANY excuse. Thanks though, I’m already aware of latency compensation.
NOTE: In “cellular system now working” “now” is supposed to be “not”. Sorry, my typing hand has poor buffering.
Hey Anonymous!
What the hell?!? Are you a electrical engineer or network specialist or something? Damn dude! You’re making me dizzy with your tech talk.
None of the above! I’ve been messing with computers since the early 1990s and always liked PBS science shows. As a kid I read a lot about astronomy, and Einstein. Oh, just for the fun of it, light travels 1 foot (30cm) in one nanosecond. I’m a lifelong electronics geek with some astrophysics thrown in for flavor.
One thing none of us will ever have to worry about. The Universe has a speed limit, being the speed of light. Don’t worry about getting a ticket with your starship, however. Why? Thanks to E=mc^2, the momentum makes the ship heavier, and the speed limit is self-enforcing.
Daley can’t set up THOSE speed cams!
Now, about LAZ Parking’s excuse???
hey anon…..good idea…..about walmart atomic watches…..but…..one minor detail……there has to be a bid process…..and shortshanks has tosee which….family member……gets the contract……
Damnit! I forgot about corruption during the dissertation. I was simply thinking about the tech end of what to do. The AutoCite system would hopefully use the same atomic clock standard as the Walmart watches. Great point! The NIST time standard is too damn easy to align clocks to. That’s why this LAZ stupidity is so egregious. Sure, they gave us 2 minutes, but anyone with half a brain or better knows it’s a PR gimmick. NEXT!