TPS data

zechdz · Post by **zechdz** » Tue Jan 12, 2016 6:02 am

I need help trouble shooting.

Please look at lap 451 and 454, included in the link.

https://www.race-capture.com/events/tsu ... laps[]=453

(Need to rescale the graphs for them because I sent the wrong min/max during streaming)

I setup SusTravelF, SusTravelR, RPM, and TPS.
The sus travel and RPM look decent, although the sus looks a bit digital... not sure if its some sort of sampling artifact?

My main issue is the TPS. On the backstraight I'm pretty sure I'm doing full throttle from corner exit to final corner entry, but the TPS value dies off 1/4 way down the straight, and doesn't even hit full throttle for one of the laps.

It's hard to debug a live test like this, to see if the voltage is actually dying or if the data is just not getting recorded well. Do you have any tips for how to debug this?

I really want to trust the RCP, and be able to use it to actually know if I'm using the throttle correctly, lol.

My first course of action was to try logging the battery value and see if the battery is having trouble at those power loads, full throttle, high speed. Besides that I'm not sure what I can do because the throttle records fine when parked with the engine off, and when warming the engine up.

JMcDonough · Post by **JMcDonough** » Tue Jan 12, 2016 3:44 pm

What is the source of the TPS data? Raw voltage signal? OBD PID or CAN data? Electronic throttle?

stieg · Post by **stieg** » Tue Jan 12, 2016 4:43 pm

That is a bit strange, but may somewhat accurate. When you compare your speed against your TPS graph you can see that the car is gaining speed up until the 1.05 mile marker whereas the TPS seems to start dropping off around the 0.975 - 1.025 mile mark. Assuming we say that TPS drops off at the 1 mile mark (the average between .975 and 1.025), that is a difference of about 264ft (5280 * (1.05 - 1.00)). At the 1 mile mark you are doing about 126MPH which equals 184.8 feet/second (126 × 5280 ÷ 3600). So that is about a 1.4286 second gap in response time ( 264÷184.8 ) between TPS drops and when speed starts to decrease. Seems a little high indeed but hard to know what the cause is. Would be good to know how you are reading the TPS value. Also would you be willing to post your RCP config? May help.

zechdz · Post by **zechdz** » Wed Jan 13, 2016 2:56 am

I'll post the RCP config later tonight, but the TPS is from raw voltage (mechanical throttle) which I mapped while stationary, with the engine off.

This is on a 2010 CBR600RR.

I've attached the a subset of the harness blueprint to show the TPS connection.

stieg · Post by **stieg** » Wed Jan 13, 2016 4:37 pm

Ok. Will wait for the RCP config file. How did you map it exactly (the exact process you used)? How many sample points did you take? I will be able to see this from the config file. Also... I don't suppose that the voltage values shift with the engine running. Have you tried just to eliminate engine ground noise as your culprit?

momostallion · Post by **momostallion** » Wed Jan 13, 2016 5:11 pm

You could test the TPS voltage using a multimeter with the bike not running but in the 'ON' position to see if there is any voltage variance when held open for extended period of time. You could also see what level of movement of the throttle causes TPS changes.

toga94m · Post by **toga94m** » Wed Jan 13, 2016 5:47 pm

stieg wrote:Ok. Will wait for the RCP config file. How did you map it exactly (the exact process you used)? How many sample points did you take? I will be able to see this from the config file. Also... I don't suppose that the voltage values shift with the engine running. Have you tried just to eliminate engine ground noise as your culprit?

Just running the RCP playback while looking at the TPS trace and BATT trace (even though it's offscale, the numbers still show) it's obvious your battery voltage is dropping some during your straights. It's still over 14, so the engine will run fine, but if the TPS reference voltage isn't regulated, that could affect TPS reading. It goes below 14.2 also from 0.65 miles to 0.7 miles on the shorter straight, then from 0.85 to 1.05 when you reach the braking zone.

I can't see on the schematic what the Y/R wire is connected to, but it's also powering the MAP sensor. Doing a correlation between battery and RPM might give some clues, as well as battery & TPS.

Also looks like the Gr/Bl wire is sensor common, which might have some voltage shift from chassis common (RCP power common) at speed/when charging. Again, can't tell where things are connected, the wires run off the side of the schematic.

Tom

Edit - found a similar schematic here http://www.600rr.net/gallery/files/3/6/ ... iginal.png and it looks like sensor power & sensor common return back to the ECU, not specifically grounded. So there's probably some voltage being pushed in (at high RPMs) between your RCP power common and the ECU's sensor common. Sensor common ties to the O2 sensors, MAP, IAT, TPS, ECT... the Y/R wire (sensor power) only goes to TPS and MAP.

zechdz · Post by **zechdz** » Thu Jan 14, 2016 2:29 am

I changed the config halfway through the day, so some of the values may be different. But I believe the TPS was the same all day.

config attached

zechdz · Post by **zechdz** » Thu Jan 14, 2016 2:33 am

I'm having trouble getting a readable PDF at the limit of ~200kb so I'm going to post a dropbox link to the full harness schematic. Sorry but eventually this link will stop working (whenever i clean up my dropbox public folder in a month/year/who knows when)

https://dl.dropboxusercontent.com/u/176 ... arness.pdf

I need to re-read the talk about common ground, but I'm not sure of the distinction being made.

The vcc and ground I connected to the RCP was that of the data logger connector.
Was I supposed to ground the RCP to chassis?

zechdz · Post by **zechdz** » Thu Jan 14, 2016 2:40 am

stieg wrote: How many sample points did you take? I will be able to see this from the config file. Also... I don't suppose that the voltage values shift with the engine running. Have you tried just to eliminate engine ground noise as your culprit?

I took just min and max, and then capped the values at 0 for anything 0v-min, and 100 for anything max-5v/100 and did a linear mapping (I will get a real mapping later, I just wanted to get things moving and see how the live telemetry worked).

How would I eliminate engine ground noise for the RCP?

I have a background in electrical engineering, but have been a software engineer for some years, so my knowledge is a bit rusty (and close to non existent for vehicles). I've heard of eliminating engine noise for speakers, not sure how that relates to my RCP setup. (In case this counts, I just made the precision of my throttle lower to eliminate the voltage fluctuations I saw coming from the raw voltage.)

toga94m · Post by **toga94m** » Thu Jan 14, 2016 5:49 pm

zechdz wrote:I have a background in electrical engineering, but have been a software engineer for some years, so my knowledge is a bit rusty (and close to non existent for vehicles). I've heard of eliminating engine noise for speakers, not sure how that relates to my RCP setup. (In case this counts, I just made the precision of my throttle lower to eliminate the voltage fluctuations I saw coming from the raw voltage.)

I'm not thinking it's a noise issue like in audio... I'm more suspicious of voltage shift on the common lines due to large power supply current in other parts of the bike's wiring. As RPMs go up, your coils and injectors are going to fire more often, which I would expect to increase the load on your charging system. So the currents go up, and any resistance in ground wiring could cause a voltage change between your signal ground (what the sensors are hooked to) and chassis/frame ground. I don't know where your RCP power hookup is, physically.

I do know that the bike's computer brings sensor ground out as a separate pin for some signals, rather than using the frame. That's signal SG1 (guessing Sensor Ground) on pin 18 of the A connector, all Gr/Bl color wires. Power to the TPS comes from Y/R wire (also to MAP sensor) from pin 9 of the A, marked as Vcc which would imply the computer's regulated power supply, likely +5V.

The datalogger connector lists a GND on terminal 2, wire color G/Bl, but it only goes to the OP FR SUS (front suspension?) and doesn't seem to hit ground anywhere else - probably a missing dot on the schematic. OP RR SUS does eventually tie to what looks like a ground lug, one of 3 shown next to the cam pulse sensor. And following up from another lug, there's a single G wire that goes to computer connector B pin 4, called LG.

So it's a little hard to chase where the ground float voltage might be coming from, if that's the real problem. If you can move your RCP ground connection closer to that lug with the G/Bl wires, that might help. It'll be tough to diagnose if it only appears at high revs, unless you can mount a sidecar with a technician riding along.

zechdz · Post by **zechdz** » Fri Jan 15, 2016 1:15 am

Would it be reasonable to connect the RCP ground to the battery in this case?

I have the connector to the data lines sitting right next to the battery (hope that isn't the cause for anything), so getting a direct line to negative terminal should be easy...

zechdz · Post by **zechdz** » Fri Jan 29, 2016 12:21 pm

So I just realized the purpose of the 5V reference:

5V Voltage Reference
This output provides a 5V reference to be used with the Analog Inputs. Up to 50mA is provided to supply power to active sensors, pull-up resistors and other purposes. This reference output is protected against short-circuits with an auto-resetting fuse. If the fuse is active, a low or zero voltage reading may be observed on this port. To verify an over-current situation, test the voltage on the port without any load connected.

I originally thought that the 8th analog input (interval battery voltage) being mentioned was this reference point, but after reading more carefully I realize it's meant to be used with sensors.

So I would like to know:

My sensors are coming through the bike harness and are not powered by RCP 5V reference

1) Could this be the cause of my noise?
2) (I'm guessing this is wrong but will ask anyway in case). Should I wire my bikes 5V reference point to this??

zechdz · Post by **zechdz** » Fri Jan 29, 2016 12:22 pm

I would also like to know.

When you guys connect what should be a stable input (aka throttle position 0), do you get voltage fluctuations in the RCP data?

When I connect my voltage meter the value looks pretty stable, but when reading through RCP even TPS 0 or TPS 100 has +/- 0.1 V fluctuations. Is this normal?

stieg · Post by **stieg** » Fri Jan 29, 2016 5:55 pm

Would it be reasonable to connect the RCP ground to the battery in this case

You can do this but it is not a requirement. Vehicle batteries always have the negative attached to the chassis, so as long as you have low impedance between your RCP ground wire and the negative terminal on the battery you are fine. You can measure this with a multimeter. If its higher than a few Ohms, then you fix the connection so that its as close to 0 Ohms as possible.

I have the connector to the data lines sitting right next to the battery (hope that isn't the cause for anything), so getting a direct line to negative terminal should be easy...

So long as nothing is shorting to the battery (check to ensure that all wires are shielded with no no bare wires exposed) this should be fine.

I originally thought that the 8th analog input (interval battery voltage) being mentioned was this reference point, but after reading more carefully I realize it's meant to be used with sensors.

Correct. Its meant to be used by sensors that are exclusive to RCP. In example, if you put in an extra temp sender, you could use that 5V reference plus a SensorX board (https://www.autosportlabs.com/product/s ... out-board/) plus one of the analog input points to create your own temp reader for the RCP. This gives the best results in our experience as its electronically isolated from vehicle circuitry and folks can use better senders (that have a wide range of resistance change) to get very accurate readings.

So I would like to know:

My sensors are coming through the bike harness and are not powered by RCP 5V reference

1) Could this be the cause of my noise?

Probably not. Wires close together can cause EMI (electro-magnetic interference) but in this case I doubt that is your issue.

2) (I'm guessing this is wrong but will ask anyway in case). Should I wire my bikes 5V reference point to this??

No. That output on RCP is designed to drive up to 50ma of current. Also... 5V on your bike? What reference are you referring to? Most standard references on motorcycles are 6V or 12V. I have a 12V system on my BMW R1150R in example.

When you guys connect what should be a stable input (aka throttle position 0), do you get voltage fluctuations in the RCP data?

When I connect my voltage meter the value looks pretty stable, but when reading through RCP even TPS 0 or TPS 100 has +/- 0.1 V fluctuations. Is this normal?

I see minor fluctuation occasionally, but that is usually do to me moving the sensor a bit or the vehicle battery being low. I would make sure that your power and ground wires to RCP are in good shape. Having a solid ground connection is KEY to being able to accurate measure voltage values (since voltage is a relative measurement, you must measure against the same reference, which is ground). As I mentioned before, ensure that the ground connection is solid. When you are testing the impedance between the negative terminal of the battery and the wire, giggle the wire a bit to ensure the impedance doesn't spike. If it does, it may indicate a bad connection.

Hope that helps.