Aftermarket TPMS Ideas

[BriteBlue]

The indirect TPMS does not use valve stem sensors. It compares all the tire rotation speeds using the speed sensors for each wheel. If one wheel is rotating faster that probably means it has a smaller radius than the others because it's low on air.

Sponsored

 

[Gruber]

I know that, my point is all the indirect hardware is already there with wheel speed sensors and ABS. Its literally just a flip of a switch.

What costs money is warranty coverage and unnecessary complaints from customers that something seems to be not working.

https://community.cartalk.com/t/indirect-tpms-driving-me-nuts/110645/15

 

[CastorX]

What costs money is warranty coverage and unnecessary complaints from customers that something seems to be not working.

https://community.cartalk.com/t/indirect-tpms-driving-me-nuts/110645/15

I seriously don’t understand why so many people have problems with itpms. My 8th gen civic had one and always worked properly. Now in my 10th gen it’s working as expected. The only occasion it caused truble was when I resetted the system when the tires were warmed up. Even tho i knew I shouldn’t do. It must be reset before you go driving, and it must be driven straight in different speed intervals for a few minutes on a good quality road without too much load in the car.

 

[CastorX]

Daily temperature changes do not matter too much actually in itpms systems. If the external temp changes then the internal temp changes in all 4 tires. And in EU (most countries) we change the tires 2 times a year. But bad roads are not a good thing for those sytems, at least during calibration.

 

[bcrichster]

I was on the highway, one minute everything's great, next I get just the slightest wobble, pull over and it's too late, bands are showing on the sidewall

THAT right there is why I didn't get anything with bigger wheel than 18's.. I been there, Lol.

 

[Gruber]

Daily temperature changes do not matter too much actually in itpms systems. If the external temp changes then the internal temp changes in all 4 tires. And in EU (most countries) we change the tires 2 times a year. But bad roads are not a good thing for those sytems, at least during calibration.

Wrong!
The second generation iTPMS like what's in 10 gen civics will detect even all four tires going low by the same amount. It detects in real time wheel speed fluctuations as the wheel turns and memorizes the rotation speed fluctuation frequencies during the calibration drive using quite sophisticated signal analysis algorithms. That's why the calibration takes a relatively long driving time (like 20 minutes or so at speeds in about the 40-60 mph range, iirc). It's quicker with direct TPMS.

It may also trigger because some weird road surface or high temperature, snow, damage, would change how the tires flex.

So yeah, it works very well for me, because I know how it works. The light will usually pop up on the first cold day in fall/winter. Just like it does in my CRV, which has direct TPMS (still works on original batteries after 11 years ). In the civic once or twice it seemed like the calibration didn't succeed the first time. No problem, try again.

But people who don't take time to figure it out or find out may get pissed off occasionally. In civics the tire icon on the dashboard usually will trigger shortly after buying a new car, probably because the dealer forgot to calibrate after adjusting pressure, or, even if they did initiate calibration, exactly as they should, the customer then failed to perform a sufficient calibrating drive in their new car. Then the customer freaks out about a light popping out in a "brand new car" even though the tire pressures seem to be fine.

 

[CastorX]

Wrong!
The second generation iTPMS like what's in 10 gen civics will detect even all four tires going low by the same amount. It detects in real time wheel speed fluctuations as the wheel turns and memorizes the rotation speed fluctuation frequencies during the calibration drive using quite sophisticated signal analysis algorithms. That's why the calibration takes a relatively long driving time (like 20 minutes or so at speeds in about the 40-60 mph range, iirc). It's quicker with direct TPMS.

Nice summary.

It may also trigger because some weird road surface or high temperature, snow, damage, would change how the tires flex.

So yeah, it works very well for me, because I know how it works. The light will usually pop up on the first cold day in fall/winter. Just like it does in my CRV, which has direct TPMS (still works on original batteries after 11 years ). In the civic once or twice it seemed like the calibration didn't succeed the first time. No problem, try again.

But people who don't take time to figure it out or find out may get pissed off occasionally. In civics the tire icon on the dashboard usually will trigger shortly after buying a new car, probably because the dealer forgot to calibrate after adjusting pressure, or, even if they did initiate calibration, exactly as they should, the customer then failed to perform a sufficient calibrating drive in their new car. Then the customer freaks out about a light popping out in a "brand new car" even though the tire pressures seem to be fine.

Believe or not I know how it works. I just didn't want to start talking about the details... again. I HiL tested that thing for 2 years by continental and a friend of mine HiL tested the civic's TPMS system a few years back. Conti sells the same SW for multiple OEMs but finetuned fo a specific car. I think he also mentioned that for the civic the frequency analysis is active only on 2 wheels (rear axle maybe), but he wasn't completely sure about this part. The BASE calibration takes a few minutes. This is called DDS by continental (Deflation Detection System) The wheel speed comparator algorithm learns much faster than the "quite sophisticated signal analysis algorithms" (called DDS+ by continental). That "sophisticated" algorithm uses FFT (fast fourier transformation) to find the energy peaks in the in the wheel speed sensor's signal. And as far as I know they even calculate with the external temperature and estimated the vehicle load. It was quite a hassle for the colleagues to make it run as fast as possibe. During development the the mudule often caused ECU loop timeout errors because of the FFT process. Noone was happy when such thing happened...
The rough surface recognition (or whatever it was called) deactivated the DDS and DDS+ modules in rough road.
On the HiL where we could create ideal circumstances where the System calibration of the Plus system together with the base system took about 3-4 minutes or less! And as far as I remember the completion of the base calibration was usually enough NOT to throw an incomplete calibration warning at least for the first XY miles, I don't know how much XY exactly is.

 

[Gruber]

Nice summary.

Believe or not I know how it works. I just didn't want to start talking about the details... again. I HiL tested that thing for 2 years by continental and a friend of mine HiL tested the civic's TPMS system a few years back. Conti sells the same SW for multiple OEMs but finetuned fo a specific car. I think he also mentioned that for the civic the frequency analysis is active only on 2 wheels (rear axle maybe), but he wasn't completely sure about this part. The BASE calibration takes a few minutes. This is called DDS by continental (Deflation Detection System) The wheel speed comparator algorithm learns much faster than the "quite sophisticated signal analysis algorithms" (called DDS+ by continental). That "sophisticated" algorithm uses FFT (fast fourier transformation) to find the energy peaks in the in the wheel speed sensor's signal. And as far as I know they even calculate with the external temperature and estimated the vehicle load. It was quite a hassle for the colleagues to make it run as fast as possibe. During development the the mudule often caused ECU loop timeout errors because of the FFT process. Noone was happy when such thing happened...
The rough surface recognition (or whatever it was called) deactivated the DDS and DDS+ modules in rough road.
On the HiL where we could create ideal circumstances where the System calibration of the Plus system together with the base system took about 3-4 minutes or less! And as far as I remember the completion of the base calibration was usually enough NOT to throw an incomplete calibration warning at least for the first XY miles, I don't know how much XY exactly is.

Well, the owner's manual says the calibration takes approximately 30 minutes of cumulative driving at 30-65 mph (48-105 km/h). Probably a surprise to some owners.

And yes, by "sophisticated algorithms" I meant mainly the frequency analysis by Fast Fourier Transform computational methods. What exactly is in the Honda algorithm, only Honda knows. The possibilities for programming tire rotation frequency analysis are endless. Definitely not something an auto repair mechanic needs to know.

So when it works, the Honda iTPMS works fine. When something is actually broken though, it can be hard to diagnose. But an owner or a mechanic doesn't need to worry about the FFT, it's enough to go through the troubleshooting procedures with a scanner or without, and keep replacing expensive parts until it works.

The limitation of the iTPMS (as well as direct TPMS without explicit tire pressure display) is that it triggers too late. Some drivers think that because they have TPMS, they don't need to check tire pressures. That's not the idea, except for direct TPMS with each tire pressure display.
TPMS warning will prevent tires from going seriously flat as a result of slow leaking, but this is not enough. Going by TPMS warnings only, means running with underinflated tires most of the time. I check and adjust tire pressures before they trigger the TPMS.

 

[CastorX]

What exactly is in the Honda algorithm, only Honda knows.

Not even remotely. I'm pretty sure they don't know how it EXACTLY works, they dont even have the source code. Like I wrote before it's an "off the shelf" solution from Continental (Or the competitor NIRA Dynamics AB for some other OEMs). Honda (or any other OEM) has not much to do with the detection algorithm. The brake system comes from ATE ( a Continental AG company), the Brake ECU is from Continental and the iTPMS software was developed and fine-tuned for the civic by Continental Engineering Services.
Ususally the customer just tells what CAN signal in which frame should be set to what value when a deflation warning must be shown and what kind of reset strategy should be implemented (What signal, for how many seconds, and so on). And of course they send a test vehicle to fine-tune the system. I've seen a lot of those specifications and sometines there are some hilarious things in there. Mazda for example completely forgot to translate it for us and sent the Japanese version. Then we had to wait 2 additional weeks when they finally translated it(with some serious Google Translate aftertase).

You are right, mechanic doesn't have to know that.

And the orginal (at least main) goal of the system is to prevent increased fuel consumption due to low tire pressure. Strange, but true, iTPMS was never ment to be a safety related system.

Btw. I asked my old colleague and he confirmed that the external temperature is also monitored, so the changing external temperature theoretically should not trigger a low tire pressure warning.

 

[Gruber]

Daily temperature changes do not matter too much actually in itpms systems. If the external temp changes then the internal temp changes in all 4 tires.

OK, I see you are well versed in the second generation iTPMS and you knew that the civic has te 2nd gen. So why would you ever say the above?
The main purpose of all the frequency analysis is to be able to detect when all tires lose presure evenly. Otherwise all this is unnecessary. Only the simple 1st gen. iPTMS has a problem with all tires deflating equally.....

 

[CastorX]

OK, I see you are well versed in the second generation iTPMS and you knew that the civic has te 2nd gen. So why would you ever say the above?
The main purpose of all the frequency analysis is to be able to detect when all tires lose presure evenly. Otherwise all this is unnecessary. Only the simple 1st gen. iPTMS has a problem with all tires deflating equally.....

Because I think first I clicked on reply started writing and then I read that BriteBlue wrote "It compares all the tire rotation speeds using the speed sensors for each wheel" and I think I replied somehow mixed to the wrong post at the end. Now reading back again it sounds strange yes. What I ment is the rotation speed comparator is not affected by the ambient temp.

But how do you know that much about it? Did you work with these kind of systems or just read about it?

 

[julianzh]

I was going to try out those Bluetooth tpms that screw over the valve but looks ugly and gave up that idea.

 

[BriteBlue]

I've always checked my own tire pressure, but having a TPMS is just added insurance. What I've noticed the last couple years is when cold weather first hits there is a long line of cars at Discount Tire. Initially I thought they sure were busy. But then it dawned on me that probably all those TPMS indicators are going off & people are getting their tires checked & topped off with air. Makes me wonder how low the tires have to be on some vehicles before the TMPS warning indicator comes on, and if that's the only time some people check (or have checked) their tires. It seems without some form of TPMS there'd probably be a lot under inflated tires out there on the roads.

My other car has direct TPMS and it will update the readings before I get to the end of the block in the morning. So it's quick. Regarding accuracy, one tire sensor reads one psi higher than the others even after rotating. IOW the reading is following the tire.

Sponsored