02 Sensor getting killed in lightning headers
 

so just got my EFI system up a running, using the NGK afx wideband 02 sensor kit for tuning only, megasquirt computer, anyway have killed 2 02 sensors in about 1 hour of run time, seems fine till ran pretty hard then go dead...one was a brand new one and the other was a known good one we pulled off our race car, pulled it out after running..no sign of water..but maybe just hit them on deaccel ?? and then look fine ?? these are 5 wire heated ones and are not cheap...the bung is in about 2" past where tubes go into collector, way forward of the water dump...it is a 500HP with a procharger..assuming it has to be a water issue ? anybody make one that will stand up ?? thanks Rob

Mild Thunders idea!
 

Mine get wet when I`m idling so I wait till I`m running for about 5 minutes or so .. must be dry by then, then I power em up. If you`re way rich that will also kill them fast

which ones are you using ? we may be some rich but not a point where runs bad at all so ??

Wideband sensors do not like heat. They might be too close to the exhaust port. On my car engine I have been using the same AFX sensor for over 5 years with no issues with an a/f in the 10s. It has to be either water or heat killing them.

It is about 2.5" back from where collector starts, so do not think it could be heat...also the last one came off our race car and had been on there for over a year...so still thinking water...bung is in the factory location where lightning installs them...my real question then does somebody make one that water does not kill in mins, I know the ones in std cars live in moisture from the cat convertors...

When you say it kills it do you mean outrageous readings or an error code on the reader? I am getting outrageous reading and believe it must be water so I tonight did a little work on the tailpipes to stop any reversion that will hopefully fix it. It seems like water splash even though I see no signs of that.

works fine for about a hour then reads full rich, you can tell it's not..them in a min goes all lean...then pull sensor out later, looks fine, try to cal and will not..dead...killed 2 in just hours...

I've run into that before on standard Lightning headers. Wasn't even running a big cam, but reversion kept killing sensors.

I have the same problem, I will be watching this thread.

seems somebody would have to make one that would live...