Aussies at wits end.. PLEASE HELP -)

[Kristy_Kai]

Hello to all wonderful forum members,

Kristy and Kai here from Newcastle Australia... We are having non stop fuel usage problems with our 1999 Ford Explorer 4.0L OHV Manual and very much hope you can help us figure out our problem and make us happy again =). She's gone from doing 600 km's down to 450 km's per tank of 80Litres

It all started after we replaced the clutch... the minute it was done we ended up with a high idle problem that only went away after we discovered that the head gasket had blown as the result of a buggered thermostat.

So far we have done the following:

Replaced:

Battery
EGR Valve
EGR Valve to Intake "O" Ring
IAC
Full Head Rebuild, Valve Re-Lapping, Gaskets and Seal's (including new injector "O" rings)
Fuel Pump (including FPR)
Fuel Dampener (returnless system)
Spark Plugs
Plug Leads
Coil
TPS
Fuel Cap
Fuel Filter
Air Filter
Thermostat
Engine Temp sender and connector
EVAP Solenoid (under front bumper)
Serpentine Belt
Steering Pump
Idler Pulley
O2 Sensors both sides (vehicle only fitted with 2 on crossover pipes not pre or post cat)

Cleaned:
MAF sensor (sprayed sensor cleaner on and air dried for half a day)
Intake Manifold (WD-40 and rag)
Throttle Body (Throttle Body Cleaner)
EVAP Cannisters and EVAP solenoid (at rear of vehicle under spare wheel not the one under the front bumper which was replaced)

Checked:
Voltage to MAF (backcheck) = 1.00 V at idle (800rpm approx)
Voltage to MAF (terminal ignition on) = 12.46 V
Voltage to TPS (backcheck) = 0.5 V closed (no gap in butterfly throttle closed)
Voltage to TPS (acceleration) = 4.7 V Wide Open
Voltage to VSS (at connection ignition on) = 4.5 V (Manual says it should be 1.5 V)
Alternator = 14.24 V (engine on at idle)
Fuel Pressure = 65 psi at fuel rail
Vacuum Leaks = sprayed propane over all air hoses and seals (no change in engine speed or idle
Idle screw = sitting correctly unchanged
Throttle Cable = not kinked or catching
Hoses for EVAP = engine to Solenoid and to back of vehicle
No oil in water or water in oil

The only thing we haven't done is replace the DPFE only thought about that due to a post by "Wild Bill about water dripping out of the exhaust and fuel usage.

Also when we fill it up...(never going past first click of cut off at bowser) and checking fuel cap is on tight the fuel gage never goes up to the full mark (only ever 3/4) and then it bounces all over the place dropping up to 1/4 of a tank sometimes.

We are at our wit's end and would very much appreciate anyones help so any knowledge you can give us or even spots to look at would be awesome.. Thanks so much in advance.

Kristy and Kai
(from Stockton Beach Newcastle the worlds largest movable sand beach and epic 4WD heaven)

 

[budwich]

You might consider monitoring live run data to look at some of the fueling parameters along with other engine operating conditions. Hopefully, post head work, someone did a compression check to see if everything was "happy" mechanically. My bet is on for a replacement of the MAF.

 

[Kristy_Kai]

Hey Budwich,

Thanks for the reply, didn't do any compression checks unfortunately as we are lacking the tool to do that and ran out of money but we can say that the job we did was meticulous in the extreme when it came to putting the new gasket's on and torquing down everything to the correct tension not to mention pre-cleaning to ensure there was no dust or debris still left in the cylinders or galleries.

Haven't tried live data yet, was actually thinking it could be a grounding issue as there seem to be so many bugs in the machine with even the car's stereo and key in the ignition bell having a mind of their own not to mention the rear wiper coming on and off when it pleases...lol.

The MAF is a good point because that was reading high when it should have been under 1 volt which is strange... Do you know of any condition that would cause voltages to run high across the vehicle?

Am checking out with a Scan Tool II today so hopefully that will tell us something. Any other idea's are much appreciated and we will keep posting up here until we find it so anyone else with similar problems can benefit from the diagnosis... Thanks again =)

 

[budwich]

sorry didn't mean to suggest that your engine "overhaul" was an issue but just that "conditions" in terms of compression might show something (ie. significant differences between a cylinder or two). Anyways, live data would probably be your best bet at focusing your efforts. On the maf side, all things being equal, a single reading of a maf doesn't tell you much.... you need to follow it thru from idle to "high rev" to see if there is "strange things" happening along the way. Having said that, IF indeed it is "high", that would indicate to the pcm that you have more air coming in than "expected" and thus the pcm would increase the gas mixture to maintain the proper fuel to air ratio. Of course, if it was signficantly "out of whack", some codes would likely be set... so in your case, it is more likely just "off norm".... and hard to find... :-(

 

[Kristy_Kai]

LOL.... all good Budwhich. didn't think you were suggesting we did a bad job of the rebuild. Just running you through the pretty much **** things we did do when we were rebuilding it to see if anything stood out to you that we did wrong.

So today we took it to a mate of ours who has a ScanTool II and ran through the codes that were showing up and did a bit of a test of all the critical systems and the only thing that showed up was the old favourite codes of P0172 and P0175 which were both "Lean O2 Sensor Codes". (Codes now cleared as one thing we did find is that it was running in Open Loop Mode) which will cause the fuel to get used anyway as it is pretty much limp mode from what we understand.

We have however had the idea that maybe it could be an earthing issue as we did recently complete a 3" body lift... That said however the issue with the fuel usage existed prior to that operation anyway so we doubt it had much to do with it. The plan now is to put on a set of heavy duty earth lead's to the body and chassis as well as the ones from the back of the engine to the firewall and gearbox to the body. Also noticed that the PCM, VSS and MAF all have earth's which may be worth checking out as well.

The other reason why we are trying this approach is that we have actually noticed that sometimes when we go 4WDing or over speed bumps in the road the fuel usage suddenly comes good again so at this stage we are also going to look at a possible wire short as well since the Fuel Gauge, Rear Wiper, Front Speakers and Airbag lights, and Key in Ignition chime all seem to work when they feel like it...lol.

If ok with you though we would love to keep this thread and dialogue going as it's nice to have a friend we can run these things past... Often a second set of eyes will pick something up in what we are trying that we haven't thought of... Also we are going to try and find a new MAF through Rock Auto and replace that (just in case).

Look forward to hearing your thoughts on our thoughts. Hope your having a great weekend =),

Cheers,

Kristy & Kai

 

[Kristy_Kai]

Whoops correction... mean't to write "Rich Codes" Bank 1 & 2 not lean. Also found a little trick to test the MAF in a fullproof fashion other than backtesting the connections. Using a ScanTool if the Barometric Pressure reads 34.5 Hg then she's buggered as they say. So hopefully my friend from today can read that on the accumulated data he has.

 

[budwich]

well that's a lot more information than you started with mate. Actually, your "open loop" operation is your key. With the scanner, you should have been able to determine why the open loop was there. That causes the system to use default system parameters to keep things functional while it "waits" but in normal operation, the "wait" ain't too long so you need to look at that closer. Further, IF your codes are correct, I wouldn't run long without addressing these as then you will "eat" your cats assuming you still have them on.

As for your other "related issues", fixing any electrical paths is always good.

 

[Kristy_Kai]

Our thoughts exactly =) and funnily enough the Cats were one of the first things to go. We actually thought that they may be blocked previously due to bad O2 sensors so in the great Aussie tradition of she'll be right mate we cut them in half punched out the honeycomb and welded them back together which funnily enough didn't fix our issues didn't fix our issues with running rich.

We think the whole reason we have been chasing things round in circles is that the issue will sometimes resolve itself and all of a sudden we go back to wonderful gas mileage again then (and we think this is due to the open loop) things go back to bad again with a steady loss of gas mileage.

We are as you say however enlisting the help of many great friends who are like ourselves either mechanics by trade or just guys that have an interest in 4WDing in general. We are going to go online to RockAuto tomorrow and order a new MAF, DPFE, PCV and Hose and maybe a couple of O2 sensors although we think it is pretty unlikely that this is the issue.

Last but not least and one thing we forgot to mention previously is that when we pull up to lights the rev's are generally at around 1500 rpm then slowly drop down... So we are thinking that this may be linked to the VSS on the diff as it should be reading 1.5 v with the ignition off but is actually reading around 4.7v so yet another reason we are scratching our heads.

Don't suppose you have an idea why a car would have consistently out of wack voltages other than a bad PCM do you? And once again thank you for your knowledge and help it is all VERY VERY much appreciated =)

 

[budwich]

personally I wouldn't go ordering much more parts unless you know they are physically broken and / or can take them back for your money (my money is thin, my time is too but hopefully not short... ).

Removing your cats, gutting them or otherwise isn't going to solve your rich problem, they were likely a symptom that the problem (rich) has been around for a bit and was not addressed. You need to monitor your engine run time spent in "open loop"... that is a key to your low gas mileage... and the resulting cause of the likely "long open loop" is your actual problem. Related to this, again run time data is important.... what does the computer "think" the engine temperature is???

IF I had spare money, I would buy a cheap elm computer interface that would allow me to hook up my computer to the odbII port and use the corresponding freeware to monitor things. They are cheap from ebay for $20-$30... money well spent. Heck, when I first start this "method", I didn't even have a lap top, I hauled the my desk top out to the garage...

On your idle issue, I don't think the vss has any vehicle / engine influence... it deals with abs situation and maybe tranny shifting. In terms of voltages, I would suspect that perhaps you may have a grounding issue BUT a meter can usually readily find that. My take is that perhaps it is a "finger problem" with where you "think" you have ground.... you actually have to know (ie. meter) to be sure you indeed have a REFERENCE ground, otherwise any differences will cause reading issues.

As for the PCM, reseating connectors and cleaning connectors might help before your declare a "PCM dead" problem.

PS. My sister lives a few hours north of sydney so we are close...

 

[Kristy_Kai]

Not something we want to have to do either especially with our first born on the way in 6 weeks...lol so yeah the money thin thing is a reality for us to...hahaha.

The Cat removal was due to thinking the rich problem we originally had may have blocked them up which was actually a good call as they were definitely not showing much daylight.

Never heard of the ELM Interface before and since the scan tool doesn't show all data that is a wonderful idea and one we are only to happy to invest in... Computers, laptops etc are not something we are short of so that will be perfect for us and we will be on ebay first thing in the morning to purchase one =).

Idle issue... well glad to hear the VSS isn't something to worry about. It's on the diff in our truck which is a manual so no tranny but the juddering when taking off from first to second without sticking the boot into it was making us wonder.

Definitely going to get onto the ground situation once the bloody rain stops which is apparently here until wednesday. Have never been that adept with auto electrics as I stopped being a grease monkey when EFI was just coming in so without sounding like and idiot I would assume the best way to check for ground would be to set the multimeter to continuity test? And then hold a probe at either end of the strap?

PCM... will definitely give it a clean with some electrical component cleaner and go from there on that one...

We live in Thornton which is also about 2.5hours north of Sydney right near Stockton beach... Saw that your in Ottawa Ontario.... as in Canada? Or do you live with your sis near us...? Would be good to have another explorer lover close by... all these superior Range Rover Drivers think we are just Exploder Drivers... Even with our girl playing up we still showed them all up today on a track that none of their highly modified trucks could make it up.... even got it on video... (Upload to YouTube to follow...lol.. As proof)

 

[budwich]

I meant "not far" as in "mind set"... I am indeed in canada... don't get down under (lake manmorah district sp??) much as its far... but we do get together "mid point"... Hawaii.... yep a wonderful midpoint...

live data will probably help you alot... the elm interfaces are great and useful.

my guess is that you either have a temperature issue (sensor) or more likely a MAF issue based on the codes and my readings of various sites.

 

[Kristy_Kai]

Ahhh gotcha =). Well Oz is a pretty cool place honestly speaking although growing up in Banff allowed one of us to appreciate the beauty there to.

Would be lovely to take a trip to Hawaii... oh well one of these days...lol.

Ok so we ordered one of those ELM Scantool Interfaces today direct to Australia via Hong Kong. So hopefully within a week or so we will have the ability to read "Live Data". We have researched a bunch on YouTube and found that it seems to be a very cool tool indeed so kind of excited now at the prospect.

Now all we have to do is wait for this constant rain to disappear and the smell of cat crap to go away that we accidentally ran over in the front yard which is making crawling under the car to check earths a smelly proposition...hahahaha.

 

[budwich]

funny, we are even closer than you think (mind wise... , I am a native albertan... have fun your new toy when you get it, hopefully it will help you out. My truck just lit the cel light this morning so I am off to have some fun too.

 

[Kristy_Kai]

Grrrr they are just a constant source of love and pain at the same time huh...hahahahaha. Your welcome to tell your sis you have friends close by to if you want... can never have enough mates =)

Ok so updated info... live data tell's us that we have no voltage on bank 2 sensor 1 so after cross checking with other sensors (all of which could be stuffed as well given they were in there when the head gasket blew) we found that there wasn't much happening there.... pulled the whole wiring loom for the engine out and checked all points for continuity and fraying etc... nadda... all good..

Soooo whilst spending more needless money we replaced the O2 Sensor connectors, MAF Sensor, Serpentine Belt, Break Shoe's for Handy....lol. Once the new O2 Sensors arrive next week we will put them in and hope for the best... if still no good it's bite the bullet time and taking it down to the local dealer for a bit of testing....lol.

How have you been? How did the issue with your little Explorer go?

Cheers,

Kristy & Kai

 

[budwich]

mine "cel" was easy... tighten the gas cap (ie. large evap leak)...

back to you. when you are monitoring your live data.... what are seeing on your o2 sensors? You should see the voltages switch back and forth on the front two (ie. in front of the cat) and somewhat steady on the one behind (or two). What are the actual range of readings please? Ahhhh... I see a statement about you NOT having post cat o2... OK so what are the ranges for the pre.

another thing to test / try now that you might be close with the O2 analysis, is measure the resistance on both to compare results, might tell you something... might not

Lastly, you can try running by flipping the o2's around and see if the data changes / follows... then you can tell more about if its the sensor or wiring.

 

[Kristy_Kai]

Yeah have tried that one to...lol. Wasn't that. Even took the tank out and checked the pressure release valve etc... which is another part you can no longer purchase anywhere...lol. Ok so live data... well it's usually all in the negative range on both short and long term...

Bank 1 Sensor 1

Movi Live Data
2012-06-10 18:47:57

Time(seconds) Volts, %
0.000000 0.235, 42.19
6.865147 0.230, 42.19
13.021399 0.235, 42.19
19.114037 0.245, 42.19
24.986255 0.265, 42.19
31.071570 0.270, 42.19
37.121510 0.295, 42.19
43.222473 0.295, 42.19
49.351919 0.300, 42.19
55.491510 0.305, 42.19
61.497194 0.300, 42.19
67.622881 0.305, 42.19
73.738574 0.290, 42.19
79.858075 0.310, 42.19
85.870302 0.320, 42.19
91.859412 0.285, -23.44
97.849807 0.245, -23.44
104.002188 0.215, -23.44
109.984070 0.190, -23.44
115.990786 0.170, -23.44
121.867088 0.160, -23.44
127.840366 0.155, -23.44
133.869573 0.135, -23.44
139.865631 0.160, -23.44
145.841021 0.065, -23.44
151.843046 0.000, -23.44
157.986305 0.000, -23.44
164.120370 0.000, -23.44
170.086454 0.000, -23.44
175.959040 0.000, -23.44
182.107804 0.000, -23.44
188.227897 0.000, -23.44
194.212499 0.000, -23.44
200.200139 0.000, -23.44
206.338136 0.000, -23.44
212.344369 0.070, 4.69
218.498464 0.075, 4.69
224.363182 0.080, 4.69
230.491889 0.085, 4.69
236.464765 0.090, 4.69
242.468209 0.090, 4.69
248.591746 0.090, 4.69
254.692202 0.095, 4.69
260.617779 0.095, 4.69
266.752631 0.095, 4.69
272.739325 0.100, 4.69
278.746632 0.090, 1.56
284.720442 0.000, -23.44
290.722105 0.000, -23.44
296.842756 0.000, -23.44
302.998058 0.005, -23.44
309.091976 0.030, -23.44
315.215204 0.040, -23.44
321.210175 0.010, -23.44
327.343048 0.040, -23.44
333.331810 0.115, -21.88
339.492186 0.085, -14.06
345.509374 0.070, -18.75
351.616223 0.000, -17.19
357.717023 0.000, -23.44
363.854683 0.000, -23.44
369.971560 0.000, -23.44
375.995081 0.000, -23.44
381.956781 0.000, -23.44
387.867953 0.000, -23.44
393.841008 0.000, -23.44
399.835906 0.000, -23.44
405.968142 0.000, -21.09
411.974028 0.000, -17.19
417.971433 0.080, 7.81
423.973859 0.015, 42.19
430.113826 0.050, 42.19
436.236565 0.050, 42.19
442.250574 0.055, 42.19
448.250956 0.050, 42.19
454.210071 0.055, 42.19
460.214598 0.055, 42.19
466.220475 0.055, 42.19
472.369700 0.060, 42.19
478.332545 0.060, 42.19
484.335878 0.070, 4.69
490.480222 0.065, 4.69
496.349445 0.065, 4.69
502.490950 0.070, 30.47
508.526448 0.075, -23.44
514.490251 0.040, -23.44
520.486166 0.045, -23.44
526.619022 0.050, -23.44
532.591503 0.055, -23.44

Bank 2 Sensor 1

Movi Live Data
2012-06-10 18:47:57

Time(seconds) Volts, %
0.000000 0.625, -23.44
6.826907 0.550, -23.44
12.964030 0.550, -23.44
19.127664 0.555, -23.44
24.972276 0.625, -23.44
31.134350 0.575, -23.44
37.100079 0.605, -23.44
43.240782 0.415, -14.84
49.338851 0.445, 7.81
55.476826 0.465, -19.53
61.486867 0.430, -9.38
67.610071 0.395, 42.19
73.756839 0.480, -3.91
79.848827 0.500, -23.44
85.883358 0.510, -23.44
91.871967 0.525, -23.44
97.859991 0.535, -23.44
103.989334 0.525, -23.44
109.978583 0.520, -23.44
116.002618 0.520, -23.44
121.855059 0.535, -23.44
127.826844 0.535, -23.44
133.850622 0.540, -23.44
139.810036 0.535, -23.44
145.825086 0.475, -21.88
151.829944 0.510, -23.44
157.959006 0.525, -23.44
164.057927 0.535, -23.44
170.058555 0.545, -23.44
175.987149 0.545, -23.44
182.100367 0.530, -23.44
188.220068 0.540, -23.44
194.199811 0.530, -23.44
200.212275 0.560, -23.44
206.355129 0.525, -23.44
212.356278 0.505, 4.69
218.436693 0.490, 4.69
224.357030 0.500, 4.69
230.478981 0.505, 4.69
236.437993 0.510, 4.69
242.456668 0.510, 4.69
248.578727 0.505, 4.69
254.692140 0.490, 4.69
260.604038 0.480, 4.69
266.730585 0.500, 4.69
272.726748 0.490, 4.69
278.735250 0.490, -2.34
284.718391 0.505, -23.44
290.707915 0.495, -23.44
296.864623 0.500, -23.44
302.966586 0.495, -23.44
309.104786 0.510, -23.44
315.210196 0.515, -23.44
321.200244 0.505, -23.44
327.358086 0.505, -23.44
333.325240 0.440, -19.53
339.488534 0.475, -14.84
345.477264 0.400, -15.62
351.588202 0.450, -18.75
357.702089 0.480, -22.66
363.851147 0.520, -23.44
369.887038 0.510, -23.44
375.994205 0.515, -23.44
381.902368 0.500, -23.44
387.813172 0.490, -22.66
393.857472 0.450, -22.66
399.822353 0.480, -23.44
405.954087 0.405, -20.31
411.985727 0.460, -18.75
417.978857 0.410, 18.75
423.963091 0.365, 42.19
430.082278 0.380, 42.19
436.247744 0.370, 42.19
442.241842 0.390, 42.19
448.240845 0.395, 42.19
454.223106 0.395, 42.19
460.202627 0.390, 42.19
466.211843 0.415, 42.19
472.364853 0.350, 42.19
478.346586 0.410, 42.19
484.318867 0.415, 4.69
490.464852 0.415, 4.69
496.336183 0.425, 4.69
502.509252 0.470, 25.78
508.436623 0.505, -23.44
514.466869 0.485, -23.44
520.450906 0.555, -23.44
526.611729 0.575, -23.44
532.608066 0.585, -23.44


Regarding changing them over... We had that thought to and tried it thinking that it might be the Bank 2 side that was having an issue as that is the side that is returning 0.000V and is returning a code of P1132 and P1151 Lack Upstream Headed O2 Sensor Switch - Sensor indicates Rich/Lean Also got two other codes but both were basically saying the same thing... to rich or lean...

 

[Kristy_Kai]

Damn it...lol hit reply before finished typing... so basically end result was that no matter which one out of the 6 sensors we put on the left side there was no voltage... and checking the loom twice and back checking all the wires i.e. red/grey as common voltage to IAC, DPFE, TPS, MAF etc then also the Orange wire which is the common ground and the voltage wire which is Lt Blue and Red Strip... well you get the idea.. there was full continuity with all those... Once we get the new O2 Sensors on Thursday we will be able to back check the voltage on from the PCM and from the sensors themselves..... what do you reckon of that idea? LOL... am out of other ideas.

 

[budwich]

ok.... let's slow back down.... and check back our steps as I am some what confused as to the state of testing.... especially the "flip around test".

So.... when you do your "normal run", you get data....I am not familiar with your logger... mine (free) present the information a bit different and doesn't allow me to actually capture a stream.... BUT I see just looking at the voltage values (not the percentage info cause I don't know what the means but perhaps its important / the indicator), I see the bank one sensor initially start to "switch" in some form going from some voltage to low and staying there... while bank two, stays in the mid range BUT appears to be switching in some form from .3 to .5 ish.... compared to mine which seem to run from .1 to .7/.8 (iirc).... your range is "thin".

Anyways, now if you take bank 2 sensor and move it over to bank 1 and viceversa, the big question and important point, is the data output the same... ie. does bank one data still look like it runs from .01 -.1 ... basically staying low while the "new" bank 2 sensor move from .3-.5 ish ?????

IF so getting a new sensor, is unlikely to solve your problem... :-( Not sure on the truck (mine included) how easy it is to wire the sensors backward... ie. is your wiring for the right side easily move / misdirected to the left side and viceversa.... perhaps a possibility but I would think not.

Next fuel trims... what are they saying?

Lastly, when you do your testing, I assume you are reaching "normal temperatures"... ie. the o2 sensors have warmed up and you are running closed loop status... right?

 

[Kristy_Kai]

Sorry it took us a few days to respond. Ok so the "Flip Around Test".... The basic idea there was that we were getting ok readings and voltage from the O2 Sensor on Bank 1 so to test if it was just the sensor on Bank 2 which wasn't operating or providing feedback Voltage we swapped them side to side. The result was however the same... no voltage no reading so that told us that it wasn't the sensor it was the wiring.

The percentage is basically what the O2 sensor is reading for the amount of Oxygen in the exhaust so what that was telling us is that there was tonnes of fuel and no O2 basically. And yeah your right the range was thin.

So now for the update...lol. So we ordered two new connectors for the O2 Loom side and two new sensors and put them in and now we have feedback voltage on both sensors however the engine is still returning codes P1132 and P151... however after detaching the MAF sensor which is also brand new we found that the readings improved significantly and all of a sudden she's running more lean than rich and no more fault codes come up after we cleared the others...

As far as wiring the sensors backwards goes it's pretty much impossible. We consulted both the manual and also on here as well as asking the local Ford Tech and as long as the colours for the four wires are in the right slots then it should work (which it now seems to be). And when we pulled them apart that thought occurred to us as well and upon checking they were wired correctly so we think it must have been that the pins on the old connectors were being pushed out when we pushed them together. We made sure this isn't the case with the new ones by marking them with a pen before we pushed them together and then making sure the lines above the connectors haven't been moved up =)

And yeah we checked the temp and we are reaching 88 Degree's C which is the correct operating temp apparently so there's not much wrong with the Thermostat.

We also tried pulling off the Brake Booster Line to see what if anything that did and basically she just stalled straight away... But we have once again ruled out an air leaks so it's back to the drawing board. Will try to get you some long term fuel trims tonight when we go for a drive and see if your eyes can see something ours aren't...lol.

Oh and BTW... check out the link we just found.... now this is what we call useful data... hope it helps you. We are going to buy the full copy and upload it to here for others to use to =)....gotta pay it forward right..=)

http://www.scribd.com/doc/13950965/21/Slow-Return-To-Idle

 

[fast_dave]

Kristy Kai,

I've read that one of the problems you're having is high gas consumption combined with a slow return to idle situation when pulling up to red lights.

To clarify, when you pull up to a red light, you're EX is at 1,500 rpm, and it eventually begins to lower the idle speed in defined "steps" to 600-800 rpm.

Have you ever replaced the (3) upper intake manifold rubber gaskets?

I've found that here in California, where we have high heat (like Australia) I have to replace these gaskets every 3 years.

I think what's occurring is your IAC Valve is trying to compensate for air leaks at these (3) rubber gaskets. The air leaks could also fool the Ex's computer and cause a running rich condition.

You could try re-torquing the upper intake manifold bolts, but - what I've found with these thin and fragile gaskets is that once they are compressed, and they subjected to high heat, they're flatten out and are toast...

If you have any questions - ask away.