Cheap Catalytic Convertor getting P0420 same gas mileage

1st suggestion: Pony up the $5 for the pro version so you don't have to deal with the lite version limitations. You will probably want to get the PID pack for Toyota so you don't have to try and program in the PID addresses and calcs yourself. For the purpose of strictly OBDII data which you are collecting for this issue the PID pack is really not needed since everything we currently need to log is generic OBDII, but it will always be handy for other proprietary systems like HV system, ABS, Trans, inverter, and many other systems not covered by generic OBDII PIDs.

2) Get rid of all GPS data from your logging. It's cool I guess to be able to map where you drove with Google Maps, but you are collecting a lot of data that is useless to the problem we are collecting data for and each data point you are logging counts toward your maximum data capture rate of whatever BT adapter you use. It is easy to introduce aliasing to the key fast moving data like O2 sensors making the data worthless for analyzing. Limit the data collection to only the data discussed in previous messages in the thread. You had a number of drop outs in your data, this might be a result of trying to collect too much data at once and the BT adapter couldn't deal with saving all the data in the small buffer fast enough. BTW, did you notice the OBDII vehicle speed is MPH (the one we will use) and the GPS vehicle speed is meters/sec? Get rid of GPS data, we don't need it. All we need is generic OBDII data for this.

3) STFT was an optional data collection item. LTFT is not optional and it wasn't collected. LTFT is important so we can either see the ICE is running within parameters or not. If you have engine problems that needs to be determined and STFT is not good for that since it is just a fine tune adjustment of the LTFT either additive or subtractive to LTFT. While I could guess by the STFT numbers the LTFT is likely alright, we really need to see this. If the engine is running out of spec we would need to fix that before anything else... so it's important.

4) There is no O2 data. Kind of the whole point of this data collection. It needs to be there or we got nothing.

5)Might as well change the coolant from C to F. Since you are reading MPH instead of KPH, might as well switch that over. You could calculate this in Excel using another column, but just capture in F is easier unless you really like Celsius. Coolant data isn't absolutely required, but if doing a cold start test it's nice to see when the catalytic reaction starts. Cheap cats or degraded cats might not start the catalytic reaction when a good OEM would. This means monitoring might fail if the ECM expects the catalyst to be working when it's not ready yet.

6) Whatever else you do, make sure you are not capturing data by Frame or Synchronous. This won't capture useful data for fast sensor logging. This not a big deal for slow sensors, but is critical for things like O2 sensors.

Don't get discouraged, this is your first attempt and it will get better from here. Take the above as just a bit of training and let's get this right. Attached I have reduced your data collection to just the things that make sense to capture and as you can see there isn't even enough data to try to graph to make some sense of this, no LTFT and O2 sensor data. BTW, I know I mentioned this before, but I really prefer OBD Fusion for data collection. Might just be me, but I had a ton of problems trying to get useful data from Torque Pro with dropouts and difficulty getting it setup for meaningful data capture. You might want to try that. Whatever you do, get the paid version or either app and don't expect to get the best results from a freebie app. OBD Fusion is $10 I believe and there are versions for both Android and IOS.

 

On the log file 162100 it is very close, just a couple of changes. The sample rate is way too high. It looks like 50ms (20 per sec) Please change to 500ms (twice per sec). No need to sample that fast. I never thought I would say this... but too much data!!

Get rid of Bank 3 data. The Prius only has Bank 1 so that's all you need. You have that on both LTFT and STFT Bank 3 as well as Bank 1 data. We do need the Bank 1 data. Bank 3 data (all 0s) can be stripped out with Excel, but better not to have that in the data collection list.

No need for O2B1S1V and Lambda. You can pick whatever one you like. I like raw V data, so I generally just use O2B1S1V. Lambda is fine, but that is a calculated value and not raw data. But whichever one you prefer is OK. I just don't see a need for both. O2B1S1V data of 3.33 is equal to Lambda 1.0, so either gets the point across.

Other than that, the data collection looked OK but it is very small sample with no driving and the coolant temp was too low to be useful for analysis.

With the above couple of changes you're ready to do some drive testing that we can evaluate. I have attached your data formatted for graphing and with extraneous data removed as an example. You can order the columns any way you want, I generally choose this order to quickly see relationships between data, but there are no real rules for this. Just whatever you like.

 

Got a link to the cat you bought? Would be good to know the cheap ones that don't work, because most of 'em work without issue.