How Accurate is the TPMS?

Yeah you are correct. Instead of "samples" I should have said "update display." It has to sample continuously in case of a delta P from a puncture or something. Thanks Tony!

 

Depends on the gauge you're using too. I've found several PSI difference between different gauges I've used. I just know the dealership always fills them to 35. I haven't bothered to compare to my topeak bicycle gauge though.

 

Cheers!

 

I haven't noticed the TPMS signals adjusting or changing after a reset. I'm pretty sure the sensors output engineering units, like most TPMS units I know of, so all the calibration is internal to the sensor and changing it would require adjusting a gain value in the software of the sensor. Now you could easily implement a pressure offset if the truck measured atmospheric pressure somewhere and then adjusted the TPMS absolute pressures value to gauge pressures. This is very feasible. And that would help match pressures you measure manually as tire pressure gauges measure gauge pressures. I guess it could do this all the time or just during a reset. Or never. Just thinking out loud.

What does change when you do a reset is the low threshold pressure and this is why you set the pressures first. It reads the new pressures after reset and then creates the low threshold some fixed amount below those new baseline pressures. So your low pressure threshold triggers the warning based on the baseline pressure and not some fixed pressure.

 

Reminds me of the old saying:

The man with one watch always knows what time it is, the man with two is never really sure.

 

Does it have anything to do with tire size? Maybe mine are not sized for the tires I have. My tires are a little larger than stock, 265/70/17. I thought it just sensed pressure no matter of the tire size.

 

No.

 

I aired up today using a machine the auto cuts off at the set PSI. Set it to 34psi.
After 30
Minutes of driving my tpms read 32,36,29,33

How on earth could they all be so far off from each other…

 

I was wondering myself here on the Washington state coast I see a difference in 4 to 5 lbs of air when I travel from my house to work. Morning truck is at 31 psi and when I get to work 35psi

 

That’s because your tires heat up as you drive and your pressure goes up. Totally normal

 

Got sidetracked lol yeah I was just noting my results. I was also wondering how accurate our tpms is . I have multiple tire gauges and they all read different

 

I understand that my tires warm up lol

 

Lots of variables here. Do you know for sure that the auto cut off is actually cutting off correctly? You'd need to check that with a reliable gauge. If you verify cut off is correct, then there may be issues with the TPMS. Again, check the TPMS reported pressure against a direct read gauge. TMPS do go bad after awhile.

 

How many psi from normal to being too low before the dash warning led comes on? Seems this winter i checked my tire pressure and it was between 5-6 psi low from the 30 psi i set it at this summer.

 

There's a very good chance the auto off (cut off) is less accurate on cheap Chinesium compressors than a simple gauge or the factory TPMS. I've never had a vehicle that's been more than a couple psi different from known good gauge,,, but I'll still keep in mind, the TPMS system is designed for tire failure and to alert of significant differences for efficiency and early failure warning. If one needs to tune exact tire pressures for that track day balance, he/she should still be using a known good gauge.

 

Pick one gauge to set your COLD tire pressures. Be consistent using that gauge, then ignore the other gauges.

FWIW, the term "accuracy" is a misunderstood term. There are many things to define the "accuracy" of a gauge, such as Repeatability, Hysteresis, Linearity. The readout device (display) also plays a part, Resolution, Refresh Rate. The signal from the sensor to the antenna has errors also. The software translating the signal into units can have errors, 12-bit, 16-bit. I spent multiple years in a test lab handling gauges and data acquisition.

I suspect the sensor system has an overall error level of roughly +/- 1% of reading. So, a display of 30 psi the pressure could actually be anywhere between 29.7-30.3 psi.

The good expensive pencil pressure gauge is likely +/-2% ( +/-5% is possible) of full scale for accuracy. So a 50 psi pencil gauge (+/-2%) indicating 30 psi, the actual pressure could be anywhere between 29-31 psi. If the +/-5% FS, the actual pressure is between 27.5-32.5 psi.

My point.....don't get all wound up on the "accuracy" of these devices. You are overly concerned about small differences.

Example: I was running a test on a high pressure machine using 0.1% Full Scale accuracy on a 100,000 psi pressure transducer. The "gauge" had an error level of +/- 100 psi. The readout on the computer would fluctuate in the 100 digits, meaning 100,xxx the x would move. The design engineer was very concerned about the xxx digits. Did not matter to him that the system was operating within limits. So, I changed the program to use a larger sample of points in the signal averaging and slowed the display refresh to 1/2. The Design Engineer was now happier, because the "hundreds" digit settled for a few moments. The test pressures were in the 70-80,000 psi range. To me, the engineer was silly for worrying about 100 psi error on a 70,000 psi test. Oh well......I was just the Test Engineer.....what would I know......


Remember, as Winter temperatures fall so will the tire pressure. The same for Summer rising temperatures, the pressure will rise. This is all based in Pressure, Volume, Temperature relationship. In this case the Volume is fixed.

 

This winter / fall mine has read 31 psi with low tire light on . After a few minutes of driving it goes away

 

Well mine tire pressure has dropped as low as 25 psi this winter, but my warning light hasn't come on yet, so i have no clue whats up with this warning system. Maybe needs to drop another 5 psi...who knows.

 

On my, the threshold of Low Pressure (lower limit) is 23

 

You could always experiment with it and keep letting air out until it triggers the light then you would know for sure, however l think it's programmed to light up after about a 25% drop.