[jrhorne]

Datalogging using Torque Pro App and BlackStealth’s OBD1 converter – Discussion, Q&A

This first post will be edited and updated as I gather more information, refine instructions, get questions answered, etc.

Lots of updates/additions on 5/20/18 - THIS POST BECAME TOO LONG (10000 char limit). Please see this link for the rest of the post and updates: Datalogging using Torque Pro App and BlackStealth’s OBD1...

The purpose of this thread is to discuss my experience setting up and using BlackStealth’s OBD1/MUT into Canbus OBD2 converter for use with Torque Pro App to display vehicle information from the ECU in real time and more importantly, datalog a variety of important sensors using the app. The datalogs can be exported (emailed, uploaded) as a .csv file so that you can view them in Excel or whatever other program.

Step 1: Buy the converter from BlackStealth, and get a OBD2 transmitter device so that you can transmit the data from the converter to your phone. You essentially have 3 options - bluetooth, USB, or wifi (mostly only for iOS users since they cannot use bluetooth). Apparently the cheap ebay $5 ELM Bluetooth devices work, but I opted for a slightly more expensive one that had high reviews on amazon (see below).
BlackStealth’s Converter (and any cables you need): Convert OBD1/MUT into Canbus OBD2 for Torque App
(If you are ever going to be adding sensors or making configuration changes to BlackStealth's Converter, you will need a serial cable. This is the one recommended ($14). I purchased it and was able to successfully connect to and reconfigure the converter for my boost pressure sensor: AYA 5Ft. USB 2.0 to Serial (9-pin) DB-9 RS-232 Adapter Cable FTDI Win/Mac/Linux 852660732384 | eBay)
OBD2 Bluetooth (BT): https://www.amazon.com/gp/product/B005NLQAHS/ref=oh_aui_detailpage_o02_s00?ie=UTF8&psc=1

Update: I was getting an average PID speed of only ~14 PIDs/sec with the OBD2 bluetooth transmitter. It seems that most bluetooth transmitters will be this slow unless you get a expensive one ($50+), and even then, it may not be that fast. This is important because this speed will determine the "refresh rate" for your realtime information display AND how often values are written into your log when logging. Lets say you are logging 7 PIDs (basically bare minimum for useful logging), this means values will only be refreshed about every 0.5 seconds in your log. Quite slow if you are trying to get near-realtime information when doing a WOT 2nd gear pull. 14 PIDs would only refresh about every second. Way too low.

To try and get more speed I bought a OBD2 USB adapter. Blackstealth reported speeds up to 200 PIDs/sec after "hacking" his OBD2 device as noted here (this is missing a few steps, see below in this post): New invention, convert OBD1/MUT into Canbus OBD2 for...
I opted for this OBD2 USB device for $12 from China (be sure to choose FTDI for the "color" option): https://www.ebay.com/itm/OBD-2-ELM327-Scanner-Autoscanner-V-1-5-Car-Diagnostic-Tool-Bluetooth-With-Chips/323180231234?ssPageName=STRK:MEBIDX:IT&var=512264744084&_trksid=p2057872.m2749.l2649
I also had to get a "on the go (OTG)" cable adapter to go from USB to micro USB to connect that converter to my phone (Samsung Note 5). Any recent andriod device that is USB OTG capable, and Andriod 4.0 or newer (I think) should be able to use the USB adapter with Torque. https://www.amazon.com/gp/product/B00YOX4JU6/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1

After buying those, I removed my blueteooth OBD2 transmitter and wired up the USB one. Before "hacking" the device (changing the baud rate) to achieve faster speeds, out of the box i got an average speed of closer to 40 PIDs/sec. Not quite what i was hoping for but still much faster than bluetooth. After a couple of days i decided to go ahead and modify the baud rate of the USB adapter to achieve faster speeds. Here is the exact procedure adding to what BlackSteal mentioned in the link above:

1. Open CoolTerm on your PC. Change connection settings to 38.4k baud (if this is the OBD2 device's original baud rate).
2. Plug in OBD2 to USB and power it up.
3. Hit connect with coolterm. Press Enter. Typing ATI and enter will confirm it is connected and repeat its version number.
4. Enter "AT PP 0C SV 23" (without the quotes) to set the OBD2 baud rate to 115.2k. It should respond with OK.
5. Enter "AT PP 0C ON" (without quotes) to tell the OBD2 to use the new baud rate. Should respond with OK
6. Power down the device and exit CoolTerm.
7. Reopen CoolTerm and set baud rate to 115.2k. Power on device, connect, hit enter, run ATI command, and confirm it responds with its version number using new baud rate to connect.
8. Change baud rate settings in Torque app to 57.6k baud, which is actually 115.2k baud.
9. Check your PIDs/sec in adapter status to see if you achieved faster speeds. You should now see closer to 200 PIDs/sec.

I can immedaitely tell that the realtime monitoring refreshes MUCH faster, unlike with the bluetooth where values would clearly be jumping around (i.e., not smooth) and lagging behind what was actually happening. To connect via USB you do have to make some changes in the Torque app to tell it that you are communicating via USB not bluetooth, set the BAUD rate for the device, and also select the specific device within Torque settings (you need to first plug it into your phone without Torque running so your phone detects it). Now when i plug in the USB cord to my phone after i start the car, Torque immediate launches and it connects after a few seconds. Basic instructions can be found here: USB - Torque OBD2 Wiki

Note: Another benefit of the USB OBD2 is that it will work with laptops. I have not yet tried any of the PC software to try and do realtime monitoring, logging, etc. I am just sticking to Torque for now.

Step 2: To get started, you only need to supply power to the converter, plug it into your car’s diagnostic port (in my case OBD1) using the cable purchased with the converter, and plug in your OBD2 BT to the converter’s output OBD2 plug. Of course later on you will probably want to wire in any other sensors you have like a Wideband O2, boost/pressure, oil pressure, etc, but that is optional and can be done later. For a permanent installation, a good place is under/behind the radio where the stock 1g radio amplifier is located. You will want to supply 12v that is only powered up with the ignition, such as radio supply power, or cig lighter power, both of which are in that area (you can just splice into the positive and ground to chassis somewhere nearby). For a temporary installation, or if you want to run this in multiple different cars, you can use this (I did for a few weeks before doing permanent install) https://www.amazon.com/gp/product/B072KDLH54/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1

Update: For my more permanent installation, I tapped into the rear cig power by the ashtray. I did not even need to use quick splice connectors - I actually just shoved the positive and ground wires into the connectors where they slide into the terminals on the bottom of the cig lighter. I then ran the power wire forward where stock wires already run, up to under the dash forward of ECU on top of shifter tables. I made a small box to house the converter and OBD2 transmitter so it would fit snug and not slide around, vibrate, etc. You can then run the OBD1 cables above your feed (ziptie out of the way) and over to the middle of the car. If you are going to be doing this install, plan ahead so you can wire in all your other sensors at the same time. I wired in my WBO2, and because of the way the WBO2 harness is, I actually now have the WBO2 +12v and ground going into the converter on the power supply terminals so that it is powered up at the same time. This prevented from having to extend the signal wire as it is in the same loom as the +12v and ground.
See this post for some pictures of the wiring (this was before i added in boost pressure sensor and switched to ODB2 USB): http://www.3si.org/forum/f35/datalo...iscussion-q-811626/index2.html#post1056240706

Step 2.5: Extra sensors
If you are going to be doing any real datalogging, especially for tuning purposes, you are going to need some extra sensors wired into the converter. A WBO2 like AEM will have a signal wire that you wire into the converter on the correct terminal as labeled, and then you will need to run configuration using the serial cable to select the correct sensor voltage to AFR range (see the converter's manual http://lcdbc.xp3.biz/CanbusDeviceManual.pdf). I also bought a MPX4250AP from BlackStealth so that I could get boost pressure into torque (I do not have LCDBC, nor do I have an electronic boost gauge that has a signal wire - you would not need this if you have those). Wiring that in is very easy as the converter supplies the +5V and ground to the sensor, and of course the signal wire going into the conveter as well, so everything can be loomed together from under the the hood where the sensor itself is mounted into the car to the converter. The unit is VERY compact, only about the size of a quarter, and uses 3-4mm vacuum host. You can just T into something else off the plenum to supply vacuum/boost to it. Mine is mounted where the stock EGR/boost/etc solenoids are mounted on the firewall. After wiring it into the converter you will need to configure for that specific sensor, again using the serial cable as described in the manual. This is why its a good idea to plan ahead when you are doing the install so that you can wire everything in at the same time and do the configuration all at once (its a pain to pull this out to reconfigure depending on your length of wires and how many sensors are wired into it).

 

[jrhorne]

Now on to my first set of logs, questions I have, and problems I think I discovered.

1. What is “Intake Air Temperature” from the stock OBD1 computer? Temperature at the MAF?
2. My car only has 2 stock O2 sensors (the ones after the turbos). Is Bank 1 the front or the rear?
3. Knock logged through Torque seems to be way higher than what I see on my SAFC 2. I know they are calculated in different ways, but anyone have an explanation for this? And is the one logged through torque the “better” and more accurate value?
4. The fuel trims on each bank are essentially a function of the voltage the O2 sensors are seeing right? It looks like the fuel trims (%) go from -100% to +100%, is this a fuel adjustment based on the sensor seeing rich or lean condition?
5. Is “Turbo Boost and Vacuum Gauge” basically the stock center gauge in the car that is calculated using ECU information (i.e., it’s not running on a vacuum line)?
6. Can you get IDC (injector duty cycle) from Torque somehow? Maybe from Engine load – what is Engine load %?

Now, a quick log I need help understanding. I think I opened a can of worms here because not only do I have a lot to learn, but I think I am finding problems I didn’t know I had. First a foremost, it seems that something is wrong with my Bank 2 sensor. In every log I did so far under different driving conditions it NEVER goes above 0.05 V, even when my wideband is reading close to 10.0 AFR under a WOT pull. The Bank 2 long term fuel trim is also always 0, and the short term trim is 10.16 (??) and these values never change. My Bank 1 sensor seems to vary from 0-1V as it should. These O2s were actually replaced with OEM (denso I think? From 3sx) only a few years ago…

A little bit of a backstory on this. Before I installed a wideband at the beginning of this year, my car had a NBO2 gauge (red to green, no #’s) and it would ALWAYS show all the way red, even when, which I know now from my wideband, it was running nearly 10.0 AFRs. So, I assume this was tapped into the Bank 2 sensor which is reading low voltage all the time… That seems to add up.

Now another thing I noticed is that I am getting a lot (I think?) of knock even when the car is clearly running rich under boost. I read several huge threads over the past few weeks about tuning with SAFC 2, so I know that using correction factors causes the IDC to decrease and the ECU to use more timing than it otherwise would. I am running RC 550cc injectors and other mods listed in my signature. Based on all the threads I read I think my SAFC 2 numbers are way off… but ill save that for another post once I understand these logs more. It looks like the ECU is pulling timing the crazy once it starts to see knock. Right now the car is set to 9-10psi, and although my wideband isn’t wired in to log yet, trust me when I say its basically 10.0 AFR when under high throttle.

Log is attached.

 

[bluee3000gt]

Subbed

 

[jrhorne]

Alright so i extracted some of the information from the log i posted above and renamed the headings and reformatted so that it would hopefully be easier for people to see (especially on mobile). This is a 2nd gear pull with boost around 9-10psi and AFR on wideband (not in log yet) around 10.0. The IAT is from the ECU so i assume it is at the MAF inlet (see question above), and the Approx boost is from the ECU as well (not a boost gauge). Note that TPS of 14.5 is essentially 0% throttle. Load % is engine load from the ECU (see question above).

Trying to understand why I get so much knock (and then pulled timing) when i am clearly running rich... And also why my O2 on Bank2 (B2) seems to be unresponsive. Looking for input!

Time	Load (%) RPM	mph	TPS	Timing	Knock	MAF(Hz) O2 B1(V) B2(V)	IAT(F)	Approx Boost
29:30.6	7.84	2781.25	28.52	30.59	30	0	283	0.75	0.04	78.8	-11.06
29:30.7	7.84	2781.25	28.52	30.59	30	0	283	0.75	0.04	78.8	-11.06
29:30.8	7.84	2937.5	28.52	30.59	30	0	289	0.75	0.04	78.8	-11.06
29:30.9	7.84	2937.5	28.52	30.59	30	0	289	0.43	0.04	78.8	-11.06
29:31.0	7.84	2937.5	28.52	30.59	30	0	289	0.43	0.04	78.8	-11.06
29:31.1	7.84	2937.5	28.52	30.59	30	0	289	0.43	0.04	78.8	-11.06
29:31.2	7.84	3000	28.52	30.59	30	0	289	0.43	0.04	78.8	-11.06
29:31.3	7.84	3000	28.52	30.59	30	0	289	0.72	0.04	78.8	-11.06
29:31.4	7.84	3000	30.62	30.59	30	0	289	0.72	0.04	78.8	-11.06
29:31.5	7.84	3000	30.62	30.59	30	0	289	0.72	0.04	78.8	-11.06
29:31.6	7.84	3031.25	30.62	30.59	30	0	289	0.72	0.04	78.8	-7.43
29:31.7	7.84	3031.25	30.62	30.59	31	0	289	0.72	0.04	78.8	-7.43
29:31.8	7.84	3031.25	30.62	30.59	31	0	289	0.66	0.04	78.8	-7.43
29:31.9	7.84	3031.25	30.62	30.59	31	0	289	0.66	0.04	78.8	-7.43
29:32.0	7.84	3031.25	30.62	30.59	31	0	540	0.66	0.04	78.8	-1.54
29:32.1	34.51	3125	30.62	30.59	31	0	540	0.66	0.04	78.8	-1.54
29:32.2	34.51	3125	30.62	30.59	24	0	540	0.66	0.04	80.6	-1.54
29:32.3	34.51	3125	30.62	70.2	24	0	540	0.82	0.04	80.6	-1.54
29:32.4	34.51	3125	32.75	70.2	24	0	540	0.82	0.04	80.6	-1.54
29:32.5	34.51	3125	32.75	70.2	24	0	798	0.82	0.04	80.6	-1.54
29:32.6	34.51	3437.5	32.75	70.2	24	0	798	0.82	0.04	80.6	4.07
29:32.7	43.92	3437.5	32.75	70.2	24	0	798	0.82	0.04	80.6	4.07
29:32.8	43.92	3437.5	32.75	98.43	18	0	798	0.98	0.04	80.6	4.07
29:32.9	43.92	3437.5	32.75	98.43	18	0	798	0.98	0.04	80.6	4.07
29:33.0	43.92	3437.5	32.75	98.43	18	3	1006	0.98	0.04	80.6	4.07
29:33.1	43.92	3781.25	32.75	98.43	18	3	1006	0.98	0.04	80.6	6.75
29:33.2	50.2	3781.25	32.75	98.43	18	3	1006	0.98	0.04	80.6	6.75
29:33.3	50.2	3781.25	38.32	98.43	7	3	1006	0.98	0.04	80.6	6.75
29:33.4	50.2	3781.25	38.32	98.43	7	3	1006	0.98	0.04	80.6	6.75
29:33.5	50.2	3781.25	38.32	98.43	7	19	1157	0.98	0.04	80.6	6.75
29:33.6	50.2	3781.25	38.32	98.43	7	19	1157	0.98	0.04	80.6	8.54
29:33.7	51.37	4281.25	38.32	98.43	7	19	1157	0.98	0.04	80.6	8.54
29:33.8	51.37	4281.25	38.32	98.43	8	19	1157	0.98	0.04	78.8	8.54
29:33.9	51.37	4281.25	38.32	98.43	8	19	1157	1	0.04	78.8	8.54
29:34.0	51.37	4281.25	38.32	98.43	8	19	1157	1	0.04	78.8	8.54
29:34.1	51.37	4281.25	38.32	98.43	8	18	1270	1	0.04	78.8	8.54
29:34.2	51.37	4750	38.32	98.43	8	18	1270	1	0.04	78.8	7.7
29:34.3	51.37	4750	46.17	98.43	10	18	1270	1	0.04	78.8	7.7
29:34.4	51.37	4750	46.17	98.43	10	18	1270	1.01	0.04	78.8	7.7
29:34.5	51.37	4750	46.17	98.43	10	18	1270	1.01	0.04	78.8	7.7
29:34.6	51.37	4750	46.17	98.43	10	17	1339	1.01	0.04	78.8	7.7
29:34.7	50.2	4843.75	46.17	98.43	10	17	1339	1.01	0.04	78.8	5.91
29:34.8	50.2	4843.75	46.17	98.43	14	17	1339	1.01	0.04	78.8	5.91
29:34.9	50.2	4843.75	46.17	98.43	14	17	1339	1.01	0.04	78.8	5.91
29:35.0	50.2	4843.75	46.17	98.43	14	17	1339	1.01	0.04	78.8	5.91
29:35.1	50.2	4843.75	46.17	98.43	14	17	1339	1.01	0.04	78.8	5.91
29:35.2	50.2	4843.75	46.17	98.43	14	17	1339	1.01	0.04	78.8	5.91
29:35.3	50.2	4843.75	46.17	98.43	14	17	1339	1.01	0.04	78.8	5.91
29:35.4	50.2	4843.75	53.8	98.43	14	17	1339	1.01	0.04	78.8	5.91
29:35.5	50.2	4843.75	53.8	98.43	14	19	1534	1.01	0.04	78.8	5.91
29:35.6	50.2	4843.75	53.8	98.43	14	19	1534	1.01	0.04	78.8	8.41
29:35.7	51.37	5781.25	53.8	98.43	14	19	1534	1.01	0.04	77	8.41
29:35.8	51.37	5781.25	53.8	98.43	14	19	1534	1	0.04	77	8.41
29:35.9	51.37	5781.25	53.8	98.43	14	19	1534	1	0.04	77	8.41
29:36.0	51.37	5781.25	53.8	98.43	14	19	1603	1	0.04	77	8.41
29:36.1	51.37	5781.25	53.8	98.43	14	19	1603	1	0.04	77	5.51
29:36.2	51.37	5781.25	53.8	98.43	15	19	1603	1	0.04	77	5.51
29:36.3	51.37	5781.25	53.8	14.51	15	19	1603	0.98	0.04	77	5.51
29:36.4	51.37	5781.25	61.22	14.51	15	19	1603	0.98	0.04	77	5.51
29:36.5	51.37	5781.25	61.22	14.51	15	5	94	0.98	0.04	77	5.51
29:36.6	51.37	5781.25	61.22	14.51	15	5	94	0.98	0.04	77	-13.74
29:36.7	0.78	5781.25	61.22	14.51	15	5	94	0.98	0.04	77	-13.74
29:36.8	0.78	5781.25	61.22	14.51	42	5	94	0.94	0.04	77	-13.74
29:36.9	0.78	5781.25	61.22	14.51	42	5	94	0.94	0.04	77	-13.74
29:37.0	0.78	5781.25	61.22	14.51	42	0	44	0.94	0.04	77	-13.74

 

[BlackStealth]

1. Intake Air Temp comes from the MAF.

2. For 91-93 OBD1 cars, nobody knows if we have separate fuel trims for front/rear O2 sensors. The Mitsubishi MUT2 scantool does not report fuel trims for 91-93 cars, and the fuel trims that we know about was borrowed from the DSM datalogger. This means for 91-93 OBD1 cars, the fuel trims are identical. For 94+ cars we do know about separate fuel trims, so they will reflect that.

Bank 1 = Front O2 sensor
Bank 2 = Rear O2 sensor

3. Knocksum range is between 0-26 for 3S cars. SAFC2 reads only raw knock voltages, and no filtering -- this is why it is majorly different.

4. The ECU will adjust fuel trims to in order to reach that 14.7 AFR target, primarily based on O2 sensor voltages.
5. The Turbo Boost and Vacuum gauge that comes on by default on Torque Pro should be deleted, it is wrong and will only cause confusion.
Use the custom Boost PID to read Boost/Vacuum instead. This requires you to have a pressure sensor hooked up to the CanBus converter.
6. Injector Duty cycle -- I don't know, I didn't check. IDC calculates requires two pieces of data (RPM and injector pulse width), so if it was
possible to calculate IDC, there will be some error % in it. Mostly because there is latency between RPM data, and injector pulse width data.
Engine load basically describes how hard the engine is working from 0 to 100% range. I get most of my info from here:
https://en.wikipedia.org/wiki/OBD-II_PIDs

As you already noticed, the TPS is generally above 10% at idle (closed). If the TPS falls below (approx 10%), the ECU will trigger a code for bad TPS sensor.

Yes you are right, you have a huge knock problem, 19 is large number considering 26 is the max. Also, you can unplug your rear O2 sensor plug and see
if it still reads 0 voltage.

 

[jrhorne]

Thanks for the reply and info. I updated my first post with some of the information you provided about the sensors and whatnot so that other people reading will have all the info in one place.

As far as my car goes, I'm pretty convinced that my rear O2 sensor is bad, and has been for some time. I'm actually thinking that it has been bad off the shelf when I installed it a few years ago. I tried to look up a way to test of O2 sensors but I couldn't find anything (unless i physically swapped front to rear or something, but the front is a real pain). I will try unplugging it but what should the voltage read when it is unplugged? Wouldn't it still just read 0?

As far as my knock problem, I'm not really sure what to think... I am beginning to suspect that i am having a fuel problem with one or more specific injectors, and maybe that is causing me to get really bad knock in one or more cylinders even though my overall AFR on the wideband is rich. The more research i do, the more it seems like my SAFC 2 numbers are also way off what they should be. For example at warm idle it has a -30 correction and has a AFR of 17-18:1, but with 550cc a -35 correction should be the baseline for correcting back to 360cc? At partial throttle (15-20%) on the freeway it only has -15 correction and is in the 12.5-13:1 range and at WOT it also has a -15 correction but is at 10:1? Something seems wrong for sure.

FWIW the car was dyno tuned a couple years ago (and these numbers were set) by a reputable shop in southern California that does know 3000gts and is familar with SAFC2. I haven't really changed anything since then aside from repairs and maintainance, and i looked at the dyno sheet and they tuned it at 10.5:1 AFR at WOT.

 

[Merclb56]

Subbed

 

[jrhorne]

Also another question for BlackStealth or anyone else who has this device hooked up and is using torque:

What kind of refresh rate are you seeing for realtime ECU data? Based on my logging at 0.1 second interval it looks like values only change every 0.5-1 sec, and it seems about the same in realtime monitor. I tried enabling faster communication in the settings but that didn't make any difference. Is the bottleneck the communication between my phone and the OBD2 bluetooth device?

 

[BlackStealth]

When you disconnect the O2 sensor temporarily, it will give a 0 volt reading, use this method to identify which O2 is at fault

As for refresh rates:
My cheap $5 bluetooth adapter gives 28-30 samples/sec. And someone reported the "OBDlink LX" gives around 55 samples/sec, I forget where they posted this information. It is reported the OBDlink LX can give up to 100 samples/sec according to one website. When I used a $12 cheap ELM327 adapter with an (FTDI chipset) USB cable (non-bluetooth), it was over 200 samples/sec.

Oh, the more things you monitor the slower refresh rate you get. Best to monitor only the items you need to get the best speed as possible.

 

[jrhorne]

I know it is definitely the rear O2 that is at fault. Unplugging it made zero difference. I am also suspicious of the wiring though. See my posts here: http://www.3si.org/forum/f35/rear-p...ation-datalog-included-811794/#post1056237066

I am also getting a peak of 28 PIDs/sec with the adapter linked in my first post, with faster communication enabled. It was even slower without that. My average is still only around 14/sec though, which seems like barely enough to extract a meaningful datalog when you are doing WOT pulls. I assume this means if you are logging 7 PIDs that you will only get a refresh rate of about 0.5 sec in your log (on average).

Just to confirm by "monitor" do you mean the PIDs that you specifically select to log, or the number of things that you are viewing in real time? Can you actually disable the app from reading sensors that you are not interested in to speed things up? It seems like it is always scanning everything from the ECU, and you can just select a subset of these to display in real time, or to log when you start a log file.

I also confirmed that even when both my O2 sensors were working and cycling in closed-loop, the Bank 1 fuel trims (short term, long term, etc) reported in the logs never changed, so there is no reason to log those. Bank 2 do change, but probably better to just log Volts from each bank instead.

 

[BlackStealth]

I haven't logged anything with the Torque app, but I understanding how datalogging works with OBD1/OBD2/Canbus.

If for example you get 28 PIDS/sec, and you only monitor two PIDS (aka items), then they will only update 14 times/second. If you increase that to four PIDS, it drops down to 7 times/second. If you monitor 28 PIDS, they will only update once a second. If you add more items to your screen, they will slow down. As for datalogging, does Torque let you choose which items to datalog? If so, then the less you choose, the faster refresh rate you get.

Now if you're just doing tuning, you may just want to stick to RPM, Boost, WBO2, and knock. If you're looking for some other problems, then you can change those PIDS to reflect that.

 

[jrhorne]

If for example you get 28 PIDS/sec, and you only monitor two PIDS (aka items), then they will only update 14 times/second. If you increase that to four PIDS, it drops down to 7 times/second. If you monitor 28 PIDS, they will only update once a second.

This is exactly what i was getting at. I will need to do a back to back test with say 14 PIDs logged then only a few to see if I can clearly see that my log is updating more frequently.

If you add more items to your screen, they will slow down.

This is what I am not entirely sure about yet. There are many screens on the realtime monitor page, but you can only see one of them at a time. Right now I am really only using one of them that i customized to have the information i want to see while driving. Since I am only using this screen i guess i will go to all of the other screens and remove all of the "gauges" and things that are displayed on those to hopefully make the screen I am actually using to monitor real time refresh as fast as possible. This is what I am showing right now (will add WBO2 once wired in) - the car is not connected so no data is displayed.

As for datalogging, does Torque let you choose which items to datalog? If so, then the less you choose, the faster refresh rate you get.

Yes, there is explicit configuration for which PIDs you want to log when you start datalogging. You go into settings --> data logging & upload --> select what to log, and then select the PIDs you want to be logged when you start a new log one at a time. I will do a few tests to see if I can tell in my logs that fewer PIDs definitely speeds up how often they are updated as mentioned above. That will confirm the correlation between # of PIDs selected to log and how fast they are updated in the log.

The other test will be to see if the # of PIDs shown on the realtime monitor screen(s) affects how fast they are updated there, and if this has any effect on how quickly they are updated in the log.

Now if you're just doing tuning, you may just want to stick to RPM, Boost, WBO2, and knock. If you're looking for some other problems, then you can change those PIDS to reflect that.

Agree with that which is what I realized after starting with logging about 20 PIDs just to get as much information as possible. This is how i found the problem with my rear O2 sensor. For now I am keeping on some extra stuff until I solve that, then will reduce to only essential PIDs for tuning.

As usual thanks for the quick replies! I am continuously updating the first post in this thread with info as we discuss and learn more about this.

 

[jrhorne]

When I used a $12 cheap ELM327 adapter with an (FTDI chipset) USB cable (non-bluetooth), it was over 200 samples/sec.

Coming back to this, how exactly did you get the USB adapter to connect to your android head unit in that way with torque app? Does your head unit have a USB port on the back?

I did a little searching and it looks like i would need something like this:
https://www.ebay.com/itm/ELM327-USB...I-FT232RL-Chip-PIC18F25K80-v1-4-/232058940291
https://www.amazon.com/dp/B0027YYMU6/ref=psdc_172544_t2_B00D8YZ2SA

With the 2nd item being the adapter so that I could use it with my android phone. Can you post the link to the "$12 cheap ELM327 adapter with an (FTDI chipset) USB cable" that you bought? Was it this with the FTDI option ("color") selected?
https://www.ebay.com/itm/OBD-2-ELM3...hash=item4b3f0afa42:m:mEy3sW91OQYv5bE_AHAFXsg

 

[BlackStealth]

Coming back to this, how exactly did you get the USB adapter to connect to your android head unit in that way with torque app? Does your head unit have a USB port on the back?

I did a little searching and it looks like i would need something like this:
https://www.ebay.com/itm/ELM327-USB...I-FT232RL-Chip-PIC18F25K80-v1-4-/232058940291
https://www.amazon.com/dp/B0027YYMU6/ref=psdc_172544_t2_B00D8YZ2SA

With the 2nd item being the adapter so that I could use it with my android phone. Can you post the link to the "$12 cheap ELM327 adapter with an (FTDI chipset) USB cable" that you bought? Was it this with the FTDI option ("color") selected?
https://www.ebay.com/itm/OBD-2-ELM3...hash=item4b3f0afa42:m:mEy3sW91OQYv5bE_AHAFXsg

My android head unit natively supported FTDI, so that is why I opted for FTDI chipset USB ELM327 adapter. It was as simple as plug and play. The last link you posted ELM327 v1.5 is what I have.

Yes my head unit has two USB ports, a full size normal USB on the back, and a mini-USB up front. I tried both ports, and they both worked fine (adapter required for mini-USB).

Also, torque software is buggy. The author has the baud rates all mixed up, and auto baud doesn't work, so you have to manually try them out until one connects. Secondly, the USB ELM327 version by default is configured for 38400 baud I believe. I bumped this up for more speed. I believe I documented this somewhere already.

 

[jrhorne]

I ended up just ordering the one from the last link i posted that you said you got. It's from China so its gonna take a while to get to me. I got a OTG adapter for USB to micro USB so I can hopefully use with my phone. I saw on torque forum that a OBDLink SX USB adapter was used with a Note 3 phone, so I hope the one i ordered works with my Note 5. Although i thought i remembered reading somewhere that only Android 4.0 was supporting (not that or newer, but specifically that one). I guess I'll find out.

I tried removing all the unnecessary gauges from my realtime monitoring pages and it didnt seem to make a difference with how often the ones on the page I am actually using/viewing are refreshed/updated. It looks like they update about every 0.5 sec or so and I have 10 on that page, so i guess that makes sense with an average PID/sec of 20.

As far as reducing the number of items datalogged to check if that speeds up how often they are updated in the log data, I still need to test that, but I'm pretty confident it will. I'm still trying to fix an issue with my rear O2 sensor and find the time to wire in my wideband to your converter, then ill change my logging settings. Right now I have my logging interval (how often they are written into file) at 0.25 sec with 14 PIDs logged and it looks like they are updated roughly every 0.5 sec, so consistent with my average/max PID/sec speed i was seeing through bluetooth.

I also still have a big knock problem even though i turned my boost down to wastegate (6-7psi), am running rich (10:1 AFR) and its cold outside (ambient ~60 and even MAF temp is only 70F). Will dive deeper into this after i fix my rear O2 and wire in the wideband. For now just staying out of boost.

 

[jrhorne]

So i finally got around to doing a more permanent install of the converter today. I also took the time to wire in my Wideband O2. It took several hours, but I was taking my time and taking breaks to plan out exactly how i was going to run the wires and doing lots of tests (measure 100 times cut once approach). Yesterday I also installed a new rear oxygen sensor, which seemed to fix the problem with my rear O2 reading (near) zero volts, which was also preventing the car from going into closed loop - see this thread for details on that: http://www.3si.org/forum/f35/rear-p...losed-loop-operation-datalog-included-811794/

I chose to supply power to the converter from the rear cig lighter near the ashtray. The way my wideband is wired now means that these power wires are supplying power to both the converter and the wideband (it made the most sense to do it this way with how the wideband harness is). Shouldnt be a problem though as the wideband is rated at 1.3 amps max and the converter takes very little power.

I took some pictures and will update the first post on this thread with more details on my install procedure once i have more time. In summary i removed the factory 1g amp behind the radio, tapped into that cig power to supply power to the converter and the wideband, ran the harness from the wideband in a gauge pod down under the dash over to behind the radio to get power and supply the 0-5V (white wire on AEM) to the converter for input/logging in Torque, made a small box to hold the converter and OBD2 bluetooth, and then put everything back together.

These 3 pictures summarize the install. Again, more detailed writeup coming soon on the first post in this thread.

Power wires tapped in:

White wire is the power wire supplied by BlackStealth

Wires connected before positioning box behind radio.

 

[jrhorne]

Hey BlackStealth quick question:

Since I dont have a digital boost gauge that has a signal wire, I'm looking for a inexpensive way to get a signal for boost to input into your converter (and thus to Torque). I am wondering if something like this would work:
https://www.ebay.com/itm/Pressure-t...1000-psi-for-oil-fuel-air-water-/312004049546

Specifically i would get the 30psi since i never plan to run more boost than that. I could then just fit this to a vacuum/boost port somewhere on the plenum.

I see on your converter you have some "+5V sensor output" and "Ground for sensor" pins - are these what i would use to supply the +5V power to that sensor (it specifically says it needs 5V signal not 12V). And then i would use the "Sensor 2 signal 0-5V" to take the signal in from that sensor so I could get it in Torque?

I assume then I would need a custom PID to convert that sensor signal input from voltage to psi.

 

[BlackStealth]

That particular sensor does not work out of the box, as the canbus converter has a list of supported boost pressure sensors. I have a 3 bar GM pressure sensor you can have for $20 + shipping that is supported.

The three wires goes to:
+5V, Ground, and Boost signal.

Next run configuration menu inside the canbus, and select GM 3 bar pressure sensor for boost sensor, and you're good to go.

 

[jrhorne]

Ah i see. So is it something like this? So essentially just the 3 wires to the converter and a vacuum line for boost? Do they also read vacuum or only boost (e.g., positive pressure)?
https://www.ebay.com/itm/New-GM-Sty...tive-Motec-Megasquirt-With-Plug-/302055807310

Of course I am skeptical of anything ebay that seems "too cheap" and most of the low priced ones on amazon also have bad reviews. Is that different from the one you have? I'd rather not risk getting one that doesn't work properly to save a few $, so if yours is tested with the converter and is better quality then I am interested in buying that.

 

[BlackStealth]

Wow that's even cheaper than the one I have. If you prefer quality, then the MPX4250AP is pretty good, (non-chinese made) but only goes up to 22PSI. Both the 4250AP and the GM3 bar reads vacuum also.

The MPX4250AP is available from me for $18 USD.