Thread: RRTMS - RoadRunner tire monitoring system extension plugin

Originally Posted by bes51659

I have just installed the system and I think the learning worked. But I have not installed RR yet as I want to make sure all hardware works first.

Can I somehow start the skin without RR? Or is a simple exe that loads the dll required? Does it exist already?

No, but you can use "TPMS Doctor" to test your hardware without RR.

Yes, off cause! Sorry.

I've tried it now and it worked. I must say that the TPMS has been the easiest to install of all functionalities I've installed.

I do have some problems getting updates from the sensors, aspecially the sensor closest to the reciever, but I'll read thru the threads and see if I can find something there.

Thanks

Originally Posted by bes51659

I do have some problems getting updates from the sensors, aspecially the sensor closest to the reciever, but I'll read thru the threads and see if I can find something there.

That is strange because normally are the sensors farthest that could have some problems, maybe you need an external antenna.

Hi Robby

Thanks for your effort!

I am considering an external antenna according to your previous drawing but I also think there might be a problem with the learning usecase....

I have relearned my system several times (without doctor and config up at the same time) and this time it is wheel 2 that causes me problems. In fact I get only one value at learning and then it is silent. (If I erase data it stays with question marks. See below)

[media]http://web.comhem.se/mulle4/TPMSDoctor.jpg[/media]

Q1) Could it be that collision makes the ID wrong and then the sensor is never seen again? Or do you have retries? Should I maybe deflate below 1,5Bar and back again to force sensor into "init mode" to get more frequent updates?

Q2) Why is this? I have observed (by accident) that the abort prompt sometimes goes away before I've even started to deflate the tyre, or VERY soon after I've started. I become very suspicious that a valid learning really has taken place.

Q3) If I have a problem with one sensor, do I have to relear all tires or can I just relearn that tire? Unless the ID is already used I don't see why not.

Q4) When I use TPMSDoctor on my 480x800 display the bottom gets cut-off entirely and I do not get access to the buttons at the bottom. I then have to violate the screen resolution with 600x800 (which will give the screen a nasty offset) when I need to use them. Can I please get a smaller layout version? (Am I really the only one that has had problems with this, or am I the only one rude enough to ask?)

Once I am at it, I would really like a timer on all sensors in TPMSDoctor to show have many seconds that has passed since last update. Or something else that has the effect that I can get an idee of how well the system is working.

I have read thru this thread at last and I have put some gold nugget post links here.

Edit:
I just seen in post one (:-)) that there is a 16 bit CRC in the message so the learning should not get confused by collisioned messages. (Why not show ID also in TPMSDoctor? Or save all traffic in a log file? Is it not sent from pic to PC?) Still, something seems to be wrong with "learning".

(btw. I only count 12 bytes)

Edit 2:
Since we know that we will get one transmission every 60s in normal mode we know if messages are lost. This could be used for finding best position for receiver/external antenna. A percentage calculation in TPMSDoctor maybe?

Originally Posted by bes51659

Hi Robby

Thanks for your effort!

I am considering an external antenna according to your previous drawing but I also think there might be a problem with the learning usecase....

I have relearned my system several times (without doctor and config up at the same time) and this time it is wheel 2 that causes me problems. In fact I get only one value at learning and then it is silent. (If I erase data it stays with question marks. See below)

[media]http://web.comhem.se/mulle4/TPMSDoctor.jpg[/media]

It sounds like you have a lazy sensor.
Have you tried in running conditions (driving the car) if the data are updated?

Originally Posted by bes51659

Q1) Could it be that collision makes the ID wrong and then the sensor is never seen again? Or do you have retries? Should I maybe deflate below 1,5Bar and back again to force sensor into "init mode" to get more frequent updates?

As you have already noticed, ther is a 16 bit CRC in the message, so each data collision or bad signal reception will be automatically ignored.
Also, deflating the tire below 1,5Bar and go back again may be helpfull if there is some difficulty during the learnig procedure.

Originally Posted by bes51659

Q2) Why is this? I have observed (by accident) that the abort prompt sometimes goes away before I've even started to deflate the tyre, or VERY soon after I've started. I become very suspicious that a valid learning really has taken place.

Generally, this problem is due to an incorrect cleaning of the USB buffer, the most likely causes are the simultaneous use of several programmes which have direct acces to the device, like TPMS Doctor or RRTMS Plugin and RRTMS Config.
When the "learnig" procedure is performed, any program, with the exception of Config Tool, should to be stopped in order to prevent any interference with the USB buffer.
Sometimes it may be useful to disconnected and reconnect the interface before to perform the learning procedure for each sensor.
What software (and release) gives you this issue?

Originally Posted by bes51659

Q3) If I have a problem with one sensor, do I have to relear all tires or can I just relearn that tire? Unless the ID is already used I don't see why not.

The system doesn't store any identical sensor ID, but you can replace any sensors with a new one (with different ID) without the need to re-perform the procedure for each sensors.

Originally Posted by bes51659

Q4) When I use TPMSDoctor on my 480x800 display the bottom gets cut-off entirely and I do not get access to the buttons at the bottom. I then have to violate the screen resolution with 600x800 (which will give the screen a nasty offset) when I need to use them. Can I please get a smaller layout version? (Am I really the only one that has had problems with this, or am I the only one rude enough to ask?)

I will see what I can do for it.

Originally Posted by bes51659

Once I am at it, I would really like a timer on all sensors in TPMSDoctor to show have many seconds that has passed since last update. Or something else that has the effect that I can get an idee of how well the system is working.

Originally Posted by bes51659

I have read thru this thread at last and I have put some gold nugget post links here.

Nice work.

Originally Posted by bes51659

Edit:
I just seen in post one (:-)) that there is a 16 bit CRC in the message so the learning should not get confused by collisioned messages. (Why not show ID also in TPMSDoctor? Or save all traffic in a log file? Is it not sent from pic to PC?) Still, something seems to be wrong with "learning".

Each sensor ID is stored internally to PIC E2PROM only for internal usage and is not possible to read it.
It would require a change in the PIC firmware.

Originally Posted by bes51659

(btw. I only count 12 bytes)

You are right, sorry, my bad.

Originally Posted by bes51659

Edit 2:
Since we know that we will get one transmission every 60s in normal mode we know if messages are lost. This could be used for finding best position for receiver/external antenna. A percentage calculation in TPMSDoctor maybe?

Even this is not possible, because each time the PC does a data request to the PIC, this last always returns the latest information received from the sensors, so is not possible to determine how much time has elapsed from the last RF reception.

Thanks for your thorough answer! I don't know why I did not see it until now!

I did have some problems with reception that seems to be ok now. I do not know why. Maybe the sensor was dormant for some reason. It was not the learning procedure as it started to work all of a sudden without me doing anything.

I have not rerun the learning so I do not know why learning completed so fast. I definitely did not run any other TPMS sw than the config app. Do all apps share the same USB buffer? What if it received an update from a different tyre? Would it do like this when it discovers that ID is already used for a different tyre?

To bad the ID is not part of data sent from PIC to PC. It could have been helpful when "learning" for some reason pics the wrong sensor.

480x800 would be great, as well as an intelligent guess when sensor data was last updated. Like time when pressure, temp, or battery data last changed. Great help optimize receiver position. But for me, as I now have reached the point of using a skin, its no longer nessessary.

----

Only problem I have left is a problem I started to get when I moved the receiver from the USB in the back to the hub under the dash. I get "unknow device" and "not connected" a lot, but not always. If I disconnect and reconnect the plug it works fine. Rebooting PC or restarting TPMSDoctor is not enough.

On the same hub is only my OBDII and a cable for extra's like ipod etc. They work fine. The hub is active but the DC/DC (15W) also drives the DVD (10W max), so there should be sufficient power. I have cut 5V and GND in the USB cable from the back to the front of the car to avoid ground loops.

I have had difficult USB problems so I have looked into that a lot, and cannot see what is wrong, and I am 95% sure its not a power problem. I have 10 or more device not giving me any problems anymore. How much power does the receiver need?

Originally Posted by bes51659

Do all apps share the same USB buffer?

No, but as the USB bus is basically a serial bus, if two differents apps sent two differents data requests to the device, this last will sent back two different replies that will be recived from both apps which would be "confused" by unexpected data (especially during the learning operation).

Originally Posted by bes51659

What if it received an update from a different tyre?

Each data paket sent from the sensors allways contain the 32 bit sensor ID, so even the PIC allways knows (after it has been successful paired) wich sensor is.
If the data paket contain an unknown ID (not paired sensor), the data are simply ignored.

Originally Posted by bes51659

Would it do like this when it discovers that ID is already used for a different tyre?

The firmware inside the PIC is designed to not save any identical sensor ID, so it's impossible to pair twice the same sensor.

Originally Posted by bes51659

To bad the ID is not part of data sent from PIC to PC. It could have been helpful when "learning" for some reason pics the wrong sensor.

Due to the reason explained above, it is virtually impossible that the PIC takes the wrong sensor.

Originally Posted by bes51659

Only problem I have left is a problem I started to get when I moved the receiver from the USB in the back to the hub under the dash. I get "unknow device" and "not connected" a lot, but not always. If I disconnect and reconnect the plug it works fine. Rebooting PC or restarting TPMSDoctor is not enough.

On the same hub is only my OBDII and a cable for extra's like ipod etc. They work fine. The hub is active but the DC/DC (15W) also drives the DVD (10W max), so there should be sufficient power. I have cut 5V and GND in the USB cable from the back to the front of the car to avoid ground loops.

I have had difficult USB problems so I have looked into that a lot, and cannot see what is wrong, and I am 95% sure its not a power problem. I have 10 or more device not giving me any problems anymore. How much power does the receiver need?

If I remember well, the receiver requires about 80 mA, so it should not be a power supply problem.
I had a similar problem, but I solved it by replacing the HUB with a new different one (D-Link USB 2.0 7-Port).

Originally Posted by mnwcsult

I get "unknow device" on one of my computers the car computer. I have tried this device on three computers and on two them it works. I have tried thorugh direct connection and thourgh hubs.

Does the device work differently through USB 1.1 and USB 2.0 devices? My guess is ??? I would not think that it would work through multiple hubs. In my setup there are two chained hubs to the front of the car. There are direct connection at the back of the computer.

On the direct connection at the back of the computer I think those are USB 1.1.

Any ideals?

Sounds like you have the same type of problem as me. Or similar. Does it work if you connect the TPMS when the system is up and running?

Hope there is a solution to this. I have tried a number of hubs and this one I use now have not given mne any problems apart from this device. I've heard of "devcon", a way do enable and disable USB devices apone restart or reentry from hibernation. I will try that in a week or two.

Originally Posted by Robby BMW

Each data paket sent from the sensors allways contain the 32 bit sensor ID, so even the PIC allways knows (after it has been successful paired) wich sensor is.
If the data paket contain an unknown ID (not paired sensor), the data are simply ignored.

Thanks for your replies Robby. My questions in this matter was regarding learning process and problems that might occur _during_ mating, so we missunderstood eachother there.