A couple of questions about O2 sensors

[RadioTek]

Please forgive me if this has already been covered, but i couldn't find it by searching.

I'm in the "Engineering" phase of a Megasquirt install; that is, I've decided to do it, and am assembling the pieces and parts.

My plan is to build a MS1 v2.2 to install onto a '75 Mercedes 450SL. I plan on doing a 'fuel only' system, as i have already upgraded the breaker points type of ignition with an electronic/optical retrofit. (Crane/S&S system) The existing Bosch D-Jetronic system is having some problems that prevent me from passing Emissions, and the replacement parts are 'prohibitively expensive'. Manifold pressure Sensor is $325.00, The Pulse Generator field Repair kit alone is $495,00. I think I can do a 'squirt for less than that. In fact, the D-jet system seems like one of the forefathers of the MegaSquirt, they seem to have a lot in common.

The building of a kit is not intimidating to me, as I'm an Electronics Technician in real life, with more than 35 years of experience.

One thing that I have a question about is the O2 sensor. On this model and year, the catalytic converter was actually a part of the exhaust manifold: The cats are bolted directly to the heads, and the exhaust pipes start out from the ends of them.(one on each head of the V-8)

What i would like to know is if a (heated, of course) O2 sensor would be able to work "downstream" of the cats? I have an existing crossover pipe that seems to be the place to weld a bung for the sensor. Does anyone have any suggestions or comments?

Do I need to have a wide-band sensor, or would a narrow-band sensor work?

I'm trying to do this as economically as possible, but I know that sometimes you have to pay a little more now to prevent paying a lot more later.

Thanks,
Scott

-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
Why pay some confused person parts and labor to screw things up when you can screw it up yourself for just parts? --My Dad
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

[DonTZ125]

That's a strange one - I can't say I've heard of the cats being right at the heads!

I'm not really as up on O2 sensors as I probably could be, but I'm fairly sure the cats will give an incorrect reading. Your cross-over pipe (if upstream of the cats) sounds like a better idea.

[trakkies]

IIRC, cats. require the AFR to be kept close to stoichiometric to give a good life, and this is usually done by an O2 sensor before the cat. Modern cars often have an additional one after the cat. too. A wideband is great where you can vary the mixture for maximum power and economy - where the AFR may vary from say approx 12:1 to 16:1 - but a cat. won't last long with that.

If i were you, I'd go for an MS2. It does cost more - but not that much when building it. And it gives you the option to get rid of that dizzy at a later point. Or you could go for a MS1 V3 which is identical to the MS2 excepting the earlier processor, which can be updated easily later.

[RadioTek]

My problem is that it's virtually impossible to get an O2 sensor in front of the cats; The cats ARE the exhaust manifolds The manufacturer only perpetrated this for 2 years; the heat under the hood is REALLY rough on wiring. At 35+ years old, I'm not sure how much catalysing they're doing anymore,though.

It's possible, but expensive, to put manifolds from an earlier year on, and replace the pre-resonator with a 2-in 2-out cat, but that would require a LOT of luck to find the proper manifolds and have the exhaust system rebuilt.

Do I hope that the cats are no longer functional and go ahead? Or will a wideband at my crossover pipe (after the cats, before the first resonator) give enough info to be useful from 'catalyzed' exhaust.

Thanks,
Scott

[RadioTek]

Sorry I haven't been on recently. I had a furball at work coupled with a hard drive crash that ended up losing my bookmarks. Then there was the annual knife fight with the I.R.S.

I had an idea in the middle of the night a few weks ago: Is it possible to weld/braze a bung onto the exhaust manifold hext to the head and get a good input? Aside from the physical complications involving the metals, would it be possible to get the O2 sensor TOO hot?

I know that there would only be info from one bank of cylinders, but it may be the only way I can get a reading.

Ideas, anyone????

[trakkies]

http://wbo2.com/lsu/position.htm

[RadioTek]

Thanks, that leads to a lot of information. It sort of looks like I could almost pull this off. It seems to say that I could put either type of sensor (wide or narow band) in the manifold, if I can make the bung long enough to get the sensor element back out of the exhaust gas stream a bit. It looks like I can start moving forward on this project, now.

[kjones6039]

Hmmmmm.... How far out of the stream are we talking here?

Ken

[RadioTek]

From what I gather, I just need to get far enough back to keep it from getting too hot, about 10mm. The gasses in there are pretty turbulent, so it sounds like there should be no problem for it to get a valid reading.

[kjones6039]

IMO, I would not be concerned with ~10mm. I was just concerned with getting it way too far out of the stream.

Ken