Syvecs Forums

Not sure what your pictures are showing in relation to LSU?

The 4.9 is widely used by OEMs without issue, but it seems they do not lend themselves to aftermarket use. For reasons I've mused earlier.

The NTK (although a very old design) does seem ultra-reliable so, yes, the argument is "why bother" when it comes to the LSUs. I have not ever heard good things from the tuners I work with. But they are cheaper so customers like them so perhaps that's the reason.

From my point-of-view I have to work with many ECUs and very very few support the NTK directly. Hence my interest in getting a reliable solution for the LSU. If possible!

TimH wrote: ↑Tue Apr 21, 2020 10:25 pm Not sure what your pictures are showing in relation to LSU?

The listed specs by Bosch for maximum hex temp ranges are bullshit in reality.
In reality the LSU suffer from a very low temperature tolerance and they just are problematic when used in a high heat application. They suffer from rampant failures, the final solution to make them retain any sense of durability is to greatly reduce the temperature that the measured gas adds to the sensor body temperature.

So I came up with my own set up pictured that makes the LSU impervious to any level of exhaust temperature influence on its reliability.

We literally had thousands of users complaining about them, thus the HBX was invented, then I added to its capacity by making the bolt on Copper heat sink to augment its properties *pictured* but really the Bosch part is trash and the 'older' NTK is bullet proof thus no one I know wastes their time with the Bosch parts.

p.s. You can over heat the NTK though, and I do sometimes use that parts I developed to control the temperatures of the sensor which allows free sensor placement options

there is possibly a QA/QC component on bosch's part.

ntk make lsu 4.9 direct replacement sensors to the exact same specification as bosch, maybe they even come from the same factory (idk), yet I hear reports of people getting better life out of ntk boxed sensors compare to the bosch, particularly at elevated temperature environments. most likely a placebo effect.

in anycase, the lsu 4.9s can work just fine in oem locations with flash tuned oem ecu's where the boost is turned up to the moon on tiny hotsides. there is clearly a control aspect to this.

TimH wrote: ↑Tue Apr 21, 2020 11:36 am

rac wrote: ↑Tue Apr 21, 2020 6:46 am I've ordered some thermocouple bits and pieces to log both the egt and the pipe surface temperature at the wideband install location. will be interesting to see what the environment is like at cold start.

Will be very interested in the results of this - please share if you're willing

will do. if I can still afford to fill my tank by the time I get the thermocouples mailed to me, ironically enough since I work in oil and gas.....

Bosch LSU's remind me of Corona Virus, not much good you can say about either of them....... they are cheap though

RICE RACING wrote: ↑Wed Apr 22, 2020 3:22 am Bosch LSU's remind me of Corona Virus, not much good you can say about either of them....... they are cheap though

there is one good thing about covid19, my beer keg and the car are both in the garage and since the outbreak I've spent a lot more time with both.

but unfortunately as they say, money, time, heatlh, choose two....

what is the general consensus on denso sensors in aftermarket applications?

my s6+ could be SB modified for NTK, but only 1x, where as denso could be 2x.

So I'm pretty confident I've found the issue. I unplugged sensors that "stopped" working and plugged them into my aem gauge and they worked just fine.
Which led me to fault finding closer to the ecu and surprise surprise I have poor continuity within the potted pnp male ecu socket. I say surprise surprise because its already happened to me once on another sensor that I hoped was a once off. Now I can put tension and compression on those wires against the resin and cause massive changes in resistance from perfect to > 100ohms, that would explain the random nature of the behaviour and the sensor which played up more often also had the wires that were easier to manipulate the resistance by mechanical movement.

not good, anything could fail at any time in this scenario. will email syvecs and find out what my options are regards ecu connectors, maybe its best to wire direct to ecu.

^ still a shit sensor though
but seriously: great detective work !

FYI Where appropriate on my own stuff I solder pins then pot

Agree - flying lead looms we make from scratch (not many, these days) are always soldered/sleeved then potted. It's a b*gg€r if the solder joint fails when you pot it, but hey ho, and testing finds it! Using crimp terminals that vaguely fit the cropped pins on the header doesn't work, all the time, and I have seen fails for sure.

Not sure what the loom adapter is, in this case, though. Sod's law says it's one of mine