alfadriver said:
Knurled. said:
In reply to GCrites80s :
Why not the fuel pump circuit?
That turns on with the key, right? I'm pointing out that you need to avoid that, so that the risk of breaking the O2 heater from water drops hitting it is mitigated. Even a 5 second delay should make sure the area around the sensor is well above the dew point in the exhaust.
It shouldn't. GM will do a two second fuel pump prime, and after that it is only when cranking and when the PCM sees the engine turning. (There is also usually an oil pressure switch to turn the pump on when there is oil pressure, in case the relay fails)
I know that on the 80s/90s Mazdas, there are two separate fuel pump switches. The first is the from the ECU for the 2 second priming pulse, and the second is from the AFM after airflow starts. But in either case, you wouldn't necessarily have gotten the exhaust up above the dew point yet. I know a few people that cracked O2 sensors the way Alfa is describing when they tried to trigger off of that.
I would think something like this would work. The relay on it says it can handle 10 Amp.
In reply to WonkoTheSane :
Exactly what I was thinking about.
The Nernst cells that are the basis of O2 sensors are interesting devices.
Narrowband sensors generate a voltage that is dependent on O2 concentration on the inside (exhaust gas) vs outside (ambient), but it is not proportional, more like an on-off transfer function.
It was discovered that when an external voltage was applied the Nernst cell would actually act as a pump to transfer Oxygen.
This lead to the development of wideband cells, essentially a concentric cell within a cell having a chamber between them.
The inner cell voltage is monitored, and the the outer cell is then biased to either pump O2 into or out of the chamber until the transition voltage point occurs.
The amount of O2 pumped in to reach the transition voltage is used to calculate the O2 concentration of the monitored exhaust gasses and output to represent the O2 concentration, and so wideband O2 sensing is the result.
This is a simplified description, the real world implementation is more complex, temperatures affect the transition voltage, and must be compensated.
So most GM cars '90s and later use a heated O2 sensor, and it seems like many, many '90s GMs take the same heated O2 sensor part number. Do their circuits have a delay from the factory or was some other technique used by the factory to mitigate the condensation issue?
GCrites80s said:
So most GM cars '90s and later use a heated O2 sensor, and it seems like many, many '90s GMs take the same heated O2 sensor part number. Do their circuits have a delay from the factory or was some other technique used by the factory to mitigate the condensation issue?
The heater is controlled by the ECU. So it "knows". Early ones just had a simple timer, later ones used exhaust temperature models.
In reply to alfadriver :
IIRC they will control the current flow, not just "on" or "off".
There are two distinct types of GM 4 wire O2 sensors. The main difference being that one is designed to have its sensor ground to the chassis, and the other is designed to have its sensor ground to the PCM. (the former has a square connector, the latter has a trapezoidal connector, so you can't mix them accidentally)
Knurled. said:
In reply to alfadriver :
IIRC they will control the current flow, not just "on" or "off".
There are two distinct types of GM 4 wire O2 sensors. The main difference being that one is designed to have its sensor ground to the chassis, and the other is designed to have its sensor ground to the PCM. (the former has a square connector, the latter has a trapezoidal connector, so you can't mix them accidentally)
That depends on the sensor more than anything.
For most of the 90's, it's an on-off switch- as the heater was never so powerful that it would burn itself up just by being on.
In the late 90's, there were some fast, and ultra fast light off sensors that require power control so that they don't over heat. These sensors would be ready in 10 seconds, so that the system would be closed loop really quickly.
The rear CMS sensors have never been that powerful- they are just on-off switches. (well, up until recently- in the last 5 years, closed loop heater control was introduced to the rear sensor- from an emissions standpoint, I still can't understand why we spend that money).
WB sensors have always had closed loop heater control on them- and they are coupled with an super high power heater- so they can be ready in less than 10 seconds.
New developments in controls and sensors will have them on before you crank. And they will be robust to drops, too.
Edit- for those of you putting on NBO2 sensor into your cars with after market control, or want to have a faster heater- the late 90's era sensors that are fast light off work really well. They are powerful enough to be alive in less than 20 seconds, but not so powerful that they will burn out under high exhaust temps. Now that I've said that, I need to find some for you....
alfadriver said:The rear CMS sensors have never been that powerful- they are just on-off switches. (well, up until recently- in the last 5 years, closed loop heater control was introduced to the rear sensor- from an emissions standpoint, I still can't understand why we spend that money).
I don't know if Ford is doing it, but some manufacturers (I know for certain Chrysler and GM for the past 10-15 years) want the rear O2 to be warmed up and providing data as soon as possible because they want to see its output before the catalyst lights off, so they can calculate how much the front O2's output is skewed. The rear O2 sees less heat so it should in theory have output that is more stable long-term. (This leads to some interesting situations where the PCM trusts the rear O2 more, and will set a front O2 fault when in actuality it is the rear one that is bad)
Some automakers are also using wideband rear O2s, as well.
In reply to Knurled. :
I've looked into that, too. And you don't really need the ultra fast heaters to do it.
From an engine out and control from the start perspective, going from 13 seconds to 8 seconds on the front sensor is a big deal.
Dropping that same 5 seconds in the rear isn't. The front catalyst takes long enough to fully warm up, that a normal fast light off sensor is more than fast enough to see the catalyst do what it does as it warms up. And it saves a lot of money. As for using it to correct the front sensor- same thing, You can do it early, or you can set it up to do it via a fully warmed up system. You can use the information either way- even if it's a little different.
And a WB in the rear... I still see no real benefit to that. The kind of chemistry the rear sensor sees, well (there's not much there, relatively speaking)... there's more of a benefit to knowing a really accurate (and steep) stoich point than thinking that you can trust what the rear sensor thinks is center.
To me, the benefits are being sold by the manufacturers, like Bosch and Denso, all of which profit in many ways by selling you capability that you don't really need.
