• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

Delayed Start of Wideband for Sensor Longevity

intothelabyrinth

Active member
Joined
Mar 20, 2012
Location
Bowling Green, KY
If you have ever looked at the reviews on the common bosch sensor used for AFR gauges, you will see a lot of people complaining about how quick their sensor broke. I thought there must be a reason all these people are messing up their sensors, so I looked into it and found a pretty well researched article talking about the most common cause of early sensor failure was thermal shock from beginning to preheat the sensor before the engine was started. I decided to come up with a pretty low tech way to turn on the sensor only when the car was running for several seconds. What I ended up using was a SPDT relay, paired with a commonly available 240 A/C relay.

I power up the SPDT relay with the ignition on, and the relay coil ground is the wire for the oil pressure sender which is only grounded when the car is off. When you turn the key on the relay clicks over to the normally open output which has nothing hooked to it. Once you start the car the relay coil ground is lifted, and the relay clicks back over to the normally closed output which then supplies voltage to the input and trigger of the Volvo A/C relay. After about 5 seconds of being powered up the Volvo relay then supplies voltage to it's output which starts the wideband warm up sequence.

I will report back with my sensor longevity results once some time has passed.

F0ik7cr.jpg

7Oqranbl.jpg
 
That's a great idea. in my experience my LM1 Bosch LSU4 sensor is well over ten years old and has never had any issues. Bosch makes a pretty good sensor so i would also be interested in where those people mounted their sensors and what the control circuit was in their unit.

I'm sure you'll be fine if you followed the mounting suggestions.
 
I'm not really sure how long the delay is exactly (I was just counting it off in my head, I didn't think my seconds were that long, but maybe they are!), but I'm sure you're right if you measured it. I also don't know how long is really enough. I'm pretty sure when the controller starts up it progressively heats the sensor over another set amount of time... I'm pretty sure I could substitute another delay relay in place of the A/C relay that is controllable. How much of a delay do you think is appropriate?
 
For MegaSqurt installs I wire the O2 power to the fuel pump relay.

This has the added bonus of getting the O2 controller noise away from the MS or other electrics.
O2 controllers control the heater temp with a high current switching module that is inherently noisy.
 
A power up delay on start up is not a bad idea as long as you don't go too long. Unpowered operation of the sensor while exposed to exhaust gas will also shorten its life.

The Bosch LSU application guide does not specify a power up delay. It does set out the heater control strategy which shows a delay period of low heater current which then ramps up to achieve the set sensor temperature. The application guide does not specify the delay period, indicating that it is application specific. Applications with more exhaust plumbing and turbo stuff between the exhaust valve and the sensor location would require a longer delay to bring all that stuff up above the condensation temperature. As an observation, most wideband controllers probably include some delay period in their heater control strategy.

The 14point7 free controllers have open source firmware. If you were so inclined, you can get the firmware, figure out what the reduced heater delay time is and then modify it to a value you are more comfortable with. Works if you are handy with coding in C or its siblings.

Your design would work well for a car that does not have electronic fuel injection. If you have EFI, connecting 61 of the A/C relay to the output of the fuel pump relay would initiate the time sequence when the fuel pump starts running eliminating the need for the separate SPDT relay. 15 would be connected to switched DC power.

I switch my wideband controller relay using the fuel pump control signal from my EFI. The fuel pump running signal is 'normally' a good indication that the engine is running and the heater control strategy in the wideband controller so far seems to provide adequate delay. My car is a relatively fast starter. If you have a 'starting challenged' engine where you need protracted cranking (fuel pump running; but, engine not running) to get the engine started, even with the delay in the controller the heater could be up to operating temperature before the engine starts. In such case, your additional time delay would be a very good idea. The delay you need might be more appropriately determined by how slow your engine is to start rather than the characteristics of the exhaust plumbing.
 
I'm not really sure how long the delay is exactly (I was just counting it off in my head, I didn't think my seconds were that long, but maybe they are!), but I'm sure you're right if you measured it.

Too many years ago, but likely I just counted like you did, and writing it down made it seem like truth after time passed. Anyway I know the purpose of the timer, so for AC, whether 5 or 10 seconds makes no difference, and measuring it would not be useful, given how old those electrolytic capacitors are determining the time elapsed.

I also don't know how long is really enough.

Neither do I. Sounds like others in this thread do, though.

I'm pretty sure when the controller starts up it progressively heats the sensor over another set amount of time... I'm pretty sure I could substitute another delay relay in place of the A/C relay that is controllable. How much of a delay do you think is appropriate?

I think my guess isn't as good as yours. I was just posting the guts in case you wanted to try to tweak the time delay by adding some capacity or resistance. I'm jealous you can see your AFR whenever you like (if the sensor is working) which is something I've done a lot of thinking about but nothing more than that. And I sincerely hope you don't blow those expensive wideband sensors so often that you can get meaningful answers to the longevity question.
 
I timed it with a stop watch and I'm getting 10 seconds approximately. It was so cool to be able to see LH2.4 starting it's learning in real time on the gauge, and really cool to be able to see when it went from open to closed loop. Pretty nifty thing. I can see how it'll help troubleshoot running issues, but the main reason I wanted it was to see if I was leaning out under boost, and know exactly how far I can push boost safely. That is a neat idea about tweaking that relay to delay it more or less... haven't even considered that!
 
Wanted to check in on this, as I'd like to do similar soon.

Is the delay still working out for you? Any issues or recommendations?
 
Back
Top