So, the joke'* on me - here I was troubleshooting a non-closed loop problem for a non-closed loop calibration. Here'* the BIN file information for this vehicle where it specifies non-closed loop for export to Canada.
BCC used in 1987. (Scroll down for more info)BCC= AALU9288 Scan id= 1924 Part number= 16069287
Release date= 07/03/86 Engine size= 3.8 Trans Type= Auto trans
ECM/PCM: ECM #01227886
Used in cars:
Pontiac Oldsmobile Buick
Possibly used in:
BONNEVILLE BONNEVILLE LE DELTA 88 ROYALE
DELTA 88 ROYALE BROU LE SABRE LE SABRE CUSTOM
LE SABRE LIMITED
With C60 AIR CONDITIONER, FRONT, MANUAL CONTROLS
With FP3 TRANSAXLE, FINAL DRIVE, 2.73 RATIO
With LG3 3.8L GAS 6 CYL SFI V6 HO
With NM5 EMISSION SYSTEM, NON-CLOSED LOOP (CANADA)
I followed up with some on-line research and apparently GM was allowed to do this up until about 1988 after which the Canadian emissions standards were tightened. I guess the motivation for GM to do this would be to cut costs because they don't have to put a cat converter on these non-closed loop cars. I finally had a good look at the emission label under the hood and it specifically states that this is a non-catalyst vehicle so what I removed from the exhaust a few months ago was actually a resonator, not a cat. I still had it lying around so I had a closer look at it and yes, it doesn't really look like a cat - it looks more like a small muffler whereas the cats for this time period were much heavier and more of a pancake shape.
One of the puzzling things about this is that I have a 1985 Calais with a Buick 3.0L (very similar engine) which has a cat and definitely has a closed loop calibration and it'* an older vehicle so maybe it was manufactured for the American market originally or another possibility is that maybe GM had a quota on how many non-closed loop vehicles they could ship to Canada in any given year of production so some had it and some didn't.
Also puzzling is why it wants to go closed loop for 10 minutes and then give up and go back to open loop for the rest of the run cycle. The only thing it'* achieving is 10 minutes of trim adjustment for the BLM per cycle. Another possibility is that the short closed loop cycle was added to the calibration because it needs to go closed loop at least for a short period of time each cycle or the MAF sensor won't get a burn-off cycle after the run cycle - according to the OEM Buick manual one of the requirements to initiate a burn cycle is that it must go closed loop.
It'* also interesting that I can override the non-closed loop function by operating the vehicle in filed test or diagnostic mode where it stays in closed loop indefinitely and operates like a normal closed-loop vehicle. One other point is why put an O2 sensor in a non-closed loop vehicle unless you were expecting it to need it for a closed loop operation which again suggests it was calibrated to go closed loop for at least some of the run cycle time, possibly again just to get the MAF burn off later on.
I've checked out other EPROMs and it looks like I could swap out the ECM and EPROM for a U.*. closed loop calibration but not sure if I'll go there though - need to think about it for a bit.