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

[Jimvr4]

Do you communicate with Hamish? I'll bet his Evoscan has the wrong ECU IDs

[BlackStealth]

Do you communicate with Hamish? I'll bet his Evoscan has the wrong ECU IDs

No I don't communicate with him. ECU IDs are straight forward, so I don't think there is anything wrong with that. However, request codes and how to properly interpret some data is different on the 3S vs Evos. Most of it is the same, and some of the 3S specific items are missing. I'll add some of the new 3S specific items in soon.

[BlackStealth]

I stated earlier I saw an estimate 20% improvement in speed for OBD2 users (96-97), I did the math to see how much better it really is:
Before 8.91 PIDS/second.
After 12.16 PIDS/second.
36% increase.

Thanks goes to Melbowski for letting me borrow his 1996 ECU so I could verify my hacking results.

[Unlogic]

That is going to be great for those with 1996 ECU's. I remember the ordeals I had trying to log my 1996 ECU at high revs. It kept dropping out when the revs increased.

[BlackStealth]

That is going to be great for those with 1996 ECU's. I remember the ordeals I had trying to log my 1996 ECU at high revs. It kept dropping out when the revs increased.

Your 1996 Euro ECU communicated in MUT, the 1996 USA ECU is much worse because of OBD2. MUT vs OBD2, OBD2 is easily 10 times slower.

The dropping out at high revs was this reason:
1. Available CPU power on (1997 and older ECUs) would drop as rpms increased. So at higher rpms, the ECU didn't respond to the OBD requests in a timely manner any more.
2. When I increase the "patience factor" or my "time out" on the LCDBC it worked just fine from that point on.
3. The newer flash ECUs have a much faster microcontroller, so at higher rpms it was still able to respond to requests in a timely manner, and didn't have any problems.

It is very likely interrupt(s) were called more frequently (ie. Crank angle sensor) as the rpms increased, leaving less available cpu power to answer OBD requests. Not having access to a car with the drop out problem made it a difficult problem to solve. I wrote a special LCDBC software that would spy on the communication between the LCDBC & problematic car and fill up the RAM and then dump the results to a log. Once this data was submitted to me, I could see exactly what was the problem. I was so happy that the problem was solved with a simple software update to the LCDBC.

[2fnloud]

Tip: the wire harness or pigtail connector is hard to insert on these Chinese GM 3 BAR pressure sensor. I found it easier to disassemble the harness and insert one pin at a time.

Wire color information:
Red=5V
Green=Signal
Yellow=Ground

You wasn't joking, that is crazy. Sad part is I totally derped and didn't note what color was in what plug. I hope it is as pictured in the listing:

https://www.ebay.ca/itm/New-GM-STYLE...%257Ciid%253A1

A: Yellow-Ground
B: Blue-Signal
C Red-5V

Also what end do I need to make the plug to connect to the LCDBC brain?

[BlackStealth]

A: Yellow-Ground
B: Blue-Signal
C Red-5V

Also what end do I need to make the plug to connect to the LCDBC brain?

Looks right to me. Depending on how old your LCDBC is, it uses either JST XH or JST EH 3 pin crimp. You can order fastest from digikey.com, or order one from me. Send me private mail if you want one from me.

[2fnloud]

[BlackStealth]

That's a pic of the older LCDBC and that 3pin connector uses this part:
JST EH crimp pins, Part number SEH-001T-P0.6, or Digikey part number 455-1042-1-ND. I noticed Digkey no stock of the 3pin housing. I'll go check my stock.

[BlackStealth]

Ok I checked, I have about 20 of the old style connectors in stock. Send me private email if you want.