My LCD screen project, boost, WBO2, Knock, etc

Just had another thought,

I will remove the LCDBC and fit it too my wifes 95 TT. If the issue is the same then it MUST be software/lcdbc issue, if the issue goes then its an earthing/ecu signal issue on the car side.

I will report back in a few days when ive had a chance to swap over.

:coo3l:

ok so here goes... done some testing with the wifes car and a SECOND lcdbc (remember I ordered two) and the issues are identical in both cars.

My car is an 97 JDM NA which is converted to TT with piggy backs and fairly modded
My wifes car is a 95 JDM TT running a EU spec ecu and is pretty much stock.

BOTH cars connect perfectly on Evoscan with NO dropouts ever.
BOTH cars will NOT connect too Evoscan through the lcdbc unless you follow the 10 steps I listed on the previous page (which is not the correct way to do it, but the only way it will work)

BOTH cars will log on Evoscan through the lcdbc for a bit and then the log just goes mad and corrupts info
BOTH cars are using different lcdbc's and wired in to different power and earth points.
BOTH cars drop connection randomly and will NOT reconnect until ignition is cycled.


Im sorry to say but the issue is 100% with the unit/software/cables and not the cars..mine, my wifes and unlogics are all experiencing identical issues in an identical manner. The ONLY thing common to all three cars is the LCDBC.

If there is any more testing or things I can try too help pin point the issue let me know and I will do what I can.

Cheers
Elton :)

That's some great research Elton!

To do serial datalogging on SD card, there is this hardware $25.
SDLogger - Open Hardware Data Logger [DEV102D2P] - $24.95 : Seeed Studio Bazaar, Boost ideas, extend the reach
Then it needs TTL serial converted into RS232 serial and 5v power. So another $25 to fit all that onto a custom PCB and plastic case. If you want me to proceed let me know.

Any chance someone can scope the OBD2 data bus (pin 7)? The problem is I don't have a problematic car to examine. All the local cars that run the LCDBC don't have communication issues.

I can build a special software version for you guys to test out, that includes a longer pause before requesting data from the ECU. Maybe the Euro ECUs have trouble with responding to frequent ECU requests at higher rpms?

Quote:

---

Originally Posted by BlackStealth

To do serial datalogging on SD card, there is this hardware $25.
SDLogger - Open Hardware Data Logger [DEV102D2P] - $24.95 : Seeed Studio Bazaar, Boost ideas, extend the reach
Then it needs TTL serial converted into RS232 serial and 5v power. So another $25 to fit all that onto a custom PCB and plastic case. If you want me to proceed let me know.

---

I don't understand any of that im afraid but I don't want any extra's...I just too be able too log Evoscan through the LCDBC like it should.

Quote:

---

Originally Posted by BlackStealth

Any chance someone can scope the OBD2 data bus (pin 7)? The problem is I don't have a problematic car to examine. All the local cars that run the LCDBC don't have communication issues.

I can build a special software version for you guys to test out, that includes a longer pause before requesting data from the ECU. Maybe the Euro ECUs have trouble with responding to frequent ECU requests at higher rpms?

---

Again I have zero electronic knowledge so hopefully Unlogic can jump in, While I don't know electronics, I am an engineer so understand working out whats happening and where the issue may lay.

The issue is happening with three different spec ECU's..

Unlogic has a EU spec car and ECU
My car is a JDM car with JDM ECU
My wifes car is JDM with a UK spec ECU

While I appreciate its VERY hard finding an issue that's not apparent when you build and test these, there is definitely a connection/communication issue between the LCDBC and all three ECU's of different spec.

IM more than happy to try any test in order to resolve this issue so if you feel a different software is required then fire one over and I will give it a go :)


Is anyone else having ANY issues what so ever with the newer all in one box version ?? and is anyone else running a UK, EU or JDM spec ecu ??

Cheers
Elton

(1) Here's a version with extra Delays for fastbike76 and Unlogic to try out. This is to see if the Euro ECUs require extra delays before making a new ECU request. The "Delay=xxx" screen is located at the very end (near the Clear OBD2 section). The delays are real time so they go into effect immediately after you save.

http://lcdbc.xp3.biz/updates/DelayVersion.zip

(2) Notebook + USB OBD2/MUT cable works well, I believe this is because of the separate power ( notebook battery), and hence no ground loop interference. How about someone run the LCDBC off a 9v battery temporarily and see if it stays connected during the aggressive driving? The boost solenoid will have to be disconnected for this experiment (so we don't overwhelm the battery). If successful this will prove we're fighting a ground loop problem.

Also if you don't have a 9v battery, you could try using two 6V lantern batteries, put them in series for 12v.

Quote:

---

Originally Posted by BlackStealth

(1) Here's a version with extra Delays for fastbike76 and Unlogic to try out. This is to see if the Euro ECUs require extra delays before making a new ECU request. The "Delay=xxx" screen is located at the very end (near the Clear OBD2 section). The delays are real time so they go into effect immediately after you save.

http://lcdbc.xp3.biz/updates/DelayVersion.zip

(2) Notebook + USB OBD2/MUT cable works well, I believe this is because of the separate power ( notebook battery), and hence no ground loop interference. How about someone run the LCDBC off a 9v battery temporarily and see if it stays connected during the aggressive driving? The boost solenoid will have to be disconnected for this experiment (so we don't overwhelm the battery). If successful this will prove we're fighting a ground loop problem.

Also if you don't have a 9v battery, you could try using two 6V lantern batteries, put them in series for 12v.

---

Hi mate,

I will try the 9V battery and see what happens. My car is running boost control and WB logging through the unit, on the wifes car though I literally just connected the power wires and nothing more.

Once ive tried that I will then try the flash with the extra delays to see what changes that makes.


Thanks for your continued efforts :)

EDIT...

my battery is in the boot and easily accessible...would it be better if I ran the power wires direct too the main battery terminals as it 'shouldnt' see any interference there ??

Screenshots of OBD2/MUT communication on my bench. I wanted to see what difference there was between the OBD2 chip that's found inside the boxed version LCDBC, and the USB OBD2/MUT cable. My test probe is set at 10:1 attenuation ratio, so 500mv division really represents 5 volts. With the bench battery producing 11.03 volts, an ideal waveform is 11.03v for high signal, and 0.00v for low signal. Radio waves or noise shows up as thick green lines (2 volts).

The USB OBD2/MUT cable has a better high signal by 2V margin than the OBD2 chip. When my PCB arrive I will repeat the test and see how the transistor/resistor OBD2/MUT adapter behaves.


http://i782.photobucket.com/albums/y...ipdetailed.png

http://i782.photobucket.com/albums/y...e2detailed.png

What does the above mean mate ? Is the USB/obd cable or the OBD chip better at 'resisting' noise ?

:)

The USB OBD2/MUT cable can resist noise better than the OBD2 chip version by a margin of 2 volts. When you activate your turn signals, you can visual see a voltage drop as the dash turn signal goes bright, dim and bright again. So if 12volt pin on the OBD2 connector experiences a severe voltage drop momentarily it will corrupt data.

It could be that: those that have no trouble with OBD2/MUT communication have more than 12 volts available (ie. 12v-15v); those that do have trouble have less than 12 volts and are more susceptible to electrical noise.