pcmhacking.net

I run my wideband on 2 cars. Both have a plug ready to plug into.
There is a small variation between the 2 think the 5v is .1 different. I was thinking it might be different earth quality?
I think that's a great feature, but I've used no others to compare with.

Its a fantastic feature, makes it really simple to check/adjust the analog side of the wideband. Innovate products can be set to output a single voltage during the warmup phase and then a single voltage once warm, so you can sort of do the same. But it needs a PC and more time!

Ground location is the key to having small/no offsets. I have the heater ground to spot under the dash where many other things get grounded but the analog (or electronics) ground is to the PCM narrowband ground. Seems to be spot on like that.

Dylan wrote:I run my wideband on 2 cars. Both have a plug ready to plug into.
There is a small variation between the 2 think the 5v is .1 different. I was thinking it might be different earth quality?
I think that's a great feature, but I've used no others to compare with.

I found with my precision multi meter that the ECU sinks the WB 0-5v line slightly as soon as its connected to the ECU with all other connection connected and everything powered up already.

But when I pull the 0-5v connection out with multi on it, its jumps back to 5.00V at 20 AFR. with it connected its about 4.95V iirc ?

So after a bit of experimenting have come to the conclusion that the MTX-L works .... sort of.

I have two MTX-L units, both had readings all over the place with the sensors supplied by Innovate - one sensor has about 40k kms on it and the other maybe 2k kms. Using either sensor, on either MTX-L and the readings are not usable at all.

So got the 14point7 spartan installed, which came with a new sensor. Ran the new sensor on the MTX-L in my daily and all is ok, perfectly stable AFRs. So would think it was just faulty or old sensors right? Well the two sensors that play up on both MTX-Ls work perfectly fine on the spartan and LM1.

So my only conclusion is some difference in how Innovate are measuring AFR on the early LM1 and later MTX-L. They do control the sensor differently to everyone else, including how Bosch recommend. The 14point7 units use a different method again.

Here is a log and ADX comparing the spartan and MTX-L. Both now have new sensors, with both sensors installed in the same bank about 10cm from each other. Will get a driving log later and also hook up the LM1 for a shootout! I dont have three exhaust bungs in each bank though.

the thread I linked previously has a bit of discussion about how the Innovate and others measure lambda and drive the heater. Also the techedge website has lots of good info about how the sensor is driven, controlled and measured.

A mate of mine was about to get a MTX-L but may tell him to hold off for now...

A mate of mine tried ordering a 14point7 Spartan the other day and was told there out of stock at the moment but a newer design will be ready and in stock in 2 weeks. I have a spartan and very happy with it so far, I just leave it in the car permently as it's only my weekend toy

Yeah new spartan will be available soon using the LSU4.9 sensor

If you were starting again what would you recomend?

I like the idea of a gauge, and one that is not using an analog signal to get an AFR reading is even better (no problems with ground offsets making it read incorrectly, although the 14point7 startup sequence makes for a good check each power on).

I wouldnt go as far as not recommending the MTX-L but would like a longer sensor life, or that it was able to use older sensors that other controllers have no problem with. Starting again and knowing this would try a couple of spartans and gauge unit. Wouldnt mind trying a wideband controller using the bosch CJ125 IC (http://www.bosch-semiconductors.de/medi ... t_Info.pdf), there are heaps using this solution.


This is the spartan using my ancient and abused sensor. MTX-L still has the new sensor attached. The near stall was my fault, first start after the battery had been connected. As soon as it hit 2k RPM the IAC did its reset and pulled all the steps out! Anyway, an old sensor normally gets slow but this looks perfect still

Been playing around with the IAC settings to get the start just how I like it. Also due to the larger IAC ports compared to a Commodore all the warmup offsets needed a tweak along with a few other items to keep things happy in and out of gear and when coming to a stop while the engine is cold.

Pretty happy with this now, does not rev too high on the initial catch and has a nice decay to closed loop idle control. There is around a 40RPM increase as closed loop takes over so a couple more steps and should be just about spot on. And a video...