The custom sensors show up under the category you put them in. So if you make a sensor and call it "custom oil pressure" as a pressure sensor, then when you configuring the channels select pressure and you should see "custom oil pressure." If that not working, let me know ow and we can screen share and see what is going on. I have a bunch of sensors in this way and it works great.

 

After setting up and saving a custom sensor called "oil prs", I expected it to appear at the bottom of the "Channels" pull-down. It's not at the bottom of the list (the last entry remains "GPS Speed") nor is it anywhere else on the list.

I *was* able to edit and save the existing analog channel information on the "Channels" list but am apparently unable to add to that list

FWIW I did attempt to contact the AIM configuration experts last Friday but they were at the track. I'll be calling tomorrow.

[Edit] THE LIGHT HAS GONE ON!

 

THANK YOU MATT!

Yes, it's working the way you describe - I should now be able to proceed! Now you see the confusion new users face, not knowing where to find things. Once it's learned it becomes so engrained that it makes perfect sense to YOU guys, but it isn't described in any of the MXS or RS3 documentation where the new sensors appear.

Thanks again and the whining will cease, for now - though CAN interfaces still have a long way to go.

 

When I set up a custom sensor for temperature (resistor-type), a pull-up resistor is needed. RS3 allows entering the pull-up resistor value and defaults to 1000 ohms. When I tried changing it to the value I'm using - 2490 ohms - it complained that the value was invalid and grayed-out the calibration curve. By messing around with how the value was input I was able to get it to stop complaining, but the curve remained grayed-out.

The resulting readings "seem" to be correct - at least at room temperature - but the grayed-out curve worrys me. Is there something that has to be done to correct that? (If I set the resistor back to 1000 ohms, the curve stops being grayed out, but 1000 isn't the correct value.)

 

I would have to double check, but I believe it stops you from entering resistor values that the system can't deal with.

 

Understood, but in this case it doesn't make sense. After entering my sensor resistance values, RS3 showed a 1000 ohm pull-up. However, my sensor counts on a 2490 ohm pull-up (provided by the ECU); increasing the value only does good things - trust me, I'm a hw/sw engineer . It turns out that RS3's error checking is apparently running ALL the time, which causes problems. When the "1000" ohm figure is highlighted and the user starts to enter "2490", as soon as "2" is entered, it replaces the "1000" figure and RS3 instantly complains that "2 ohms" is invalid. Yes it is, but that's because RS3 isn't waiting for the Enter key - after the entire value has been input.

I worked around the problem by adding "2490" to the existing 1000 value, then backspaced over the 1000. That "seems" to work and the displayed temperature seems right (at least at room temperature). However, the plot showing the resistance curve remains grayed out. I'll have to do some testing to see if it's actually working throughout the sensor's range.

 

If you change the resistance reading to voltage using the resistor value, does the sensor work? Before RS3, I just used a spreadsheet that I made to calculate voltage values based on the resistor being used.

Are you doing this as an analog channel or trying to work through the CAN channels?

 

The only way to get scaling and units fixed was to give up on the CAN bus. I use it only for tach and coolant temp.

The resistance/voltage thing is often confused but the two are very related. I'm running the sensor into an ADC input on the dash, which measures voltage. However, because the pullup resistor value is also entered, I don't have to deal with volts since the two resistances form a voltage divider, and since RS3 knows the pullup value, and the sensor resistance, RS3 can figure it out. In fact Roger sent me an article on using this feature - http://www.trailbrake.net/featured-a...custom-sensors. It says, in part:

The highlighted part is exactly what I'm doing. It appears to be working yet still has the resistance curve plot grayed-out, but there wasn't any error msg once I figured out how to trick RS3 when entering the value.

 

I know all about the article - I wrote it. Does the sensor work when you input the different resistor value and have the grayed out graph? If you manually calculate the output voltage using your resistor, input those values into the custom sensor table under using the voltage, does that work?

 

I would have given you credit had your name appeared in the article!

Yes, the sensor works when I input the different (2490 ohm) resistor value, but the graph stayed grayed out. I'll have to go back and see if it still is, or if it's some sort of artifact from the pullup resistor entry error. I have not tried the voltage avenue yet.

 

The resistance page helps you figure out the best resistor value, but it's really just doing the voltage conversion for you.

 

Though the grayed-out plot is unresolved, I'll set that one aside.

Next question: How is WIFI setup? There doesn't appear to be any information anywhere other than the line item "It does WIFI" in their manual/advertisement flyer.

 

WiFi will be enabled in a future firmware update. The 802.11 chip is active and you can see an AP, but there is not yet a way to connect in the software to download or configure yet. The reason why it is taking a while is so we can ALL make sure that IF there are drops, there is error correction so as not to corrupt data or configs. It is coming...

 

Thanks, I've updated the FAQ above.

 

Is there a way to add filtering to incoming ADC variables? My fuel level reading works great... too great, swinging wildly as the fuel sloshes left to right. I'd like to add a filtering factor of at least 100, if not more.