[cgrey8]

I've got an EEC-V XDT2 that was pulled from a 97 Mercury Mountaineer 5.0L along with the original engine that I put into my truck. I kept it stored, but never used it since I retrofitted the engine to essentially be a Mustang 5.0L controllable with a GUFx era EEC.

I've been in pretty deep discussions with JSA about EEC hardware variations over the years which encouraged me to delid my A9P and XDT2 and get some pics.

The A9P had no surprises. The processor was clearly marked as an 8061 and the RAM 81C61.

However the XDT2 (following CQAB1 strat) had no obvious markings that detailed which chips were the processor, RAM/IO, or flash. We reasonably concluded which was flash (N71965FTCHCA), but we couldn't decide on which of the other chips was the processor or RAM/IO...or what kind of RAM chip it is. Here are the pertinent pics.

XDT2_Delidded.jpg

SC370673FN.jpg

N71965FTCHCA.jpg

EPIC_AM01EA.jpg

What's even more confusing is the crystal. It's a 12MHz crystal. So either this EEC-V runs slower than GUFx EEC-IVs or there's a clock-multiplier (possibly 2x) in place that runs the chip faster than the crystal. Any other EEC processors do this?

We can see traces going from the crystal to the larger EPIC chip which we thought might indicate that chip contains the 8065 however JSA recalls other EECs have crystals for the RAM/IO used to clock com ports. So that's not conclusive. It's quite possible Ford upgraded the 8065 micro-controller to utilize an internal clock (i.e. no need for an external crystal) hence why we don't see a frequency we would normally associate with an EEC-V of this time period.

Anybody else familiar with this EEC or similar EECs that can tell us more about the RAM/IO?

Is there anything in the BIN that can shed light on these details?

Xdt2.bin

I don't know of anything special about the vehicle. Like I said, the XDT2 controlled 97 Explorer/Mountaineer 5.0Ls, which were factory EDIS, used a cam position sensor, and were 4R70w on both 2wd and AWD. They also had traditional return-style fuel rails with FPR, so no fuel pressure sensor or variable speed pump. I don't even think the dash or console were fancy in any way.

What I don't recall is if the EEC/tune changed between 2wd and AWD and I don't remember which the donor vehicle was since I didn't need the trans, only the engine.

[tvrfan]

Stupid question, but you never know......
Is it possible that the 8065 (and other chips) are soldered to the OTHER SIDE ?
Some of the multi layer boards are thin enough, and chips got thinner and thinner.....

But you would have to take the whole thing apart to find out ? ..... maybe not a good idea if you want to use it at some point.

[sailorbob]

68 pin - 8065 (EDIT: ignore this, see two posts below)
32 pin - I've seen somewhere that it's 28F080 flash memory but don't know for sure (EDIT: ignore this, see two posts below)
160 pin - EPIC coprocessor

[tvrfan]

An 8065 made by (or marked as) Motorola ? Really ? Wow.....not seen that before.

[sailorbob]

Ignore what I said before, looking into it further I think the website I got the info from is wrong.

68 pin - just a guess, it could be the Analog Input Conditioner for EEC-V (AICE) chip assuming it isn't integrated into the EPIC too.
32 pin - not a standard EEPROM as people have tried to read it using 28F010 and 28F512 programmer settings to no avail, probably a MBus EEPROM.
160 pin - Electronic Powertrain Integrated Controller (hence EPIC) which is the 8065, DARC and DUCE all rolled into one.

Both Toshiba (e.g. their 8061 versions were called the 'T6126A-1' and the 40 pin 'T6127A-2') and Motorola have made chips under license for Ford.

[tvrfan]

Bob,
That seems logical ........... there is a '65' in the EPIC sequence numbers.
Just a couple of recheck questions...

Do you know if any of the EEC-V boards did have surface mount chips on both sides ?? It's a common build technique today of course, stuff often mounted both sides and chips are truly tiny. Just wondering.

I thought ALL the ROMS/PROMS/etc were custom Ford parts with M-bus logic, and so any practical change/retune is forced to use the J3 port with an adaptor ?

[cgrey8]

Also WRT to the 12MHz chip, is the EPIC using a 2x multiplier to run the actual 8065 at 24MHz? I can't imagine it is actually clocked at 12MHz. Although the Explorer/Mountaineer weren't exactly performance applications.

As for the chips, this is the 1st I've heard of EPIC & AICE. John mentioned DARC as a possibility, but I didn't know enough to ask an intelligent question about it. I think he said it is briefly mentioned in one of the handbooks, but not to great detail.

The EPIC is explained somewhat above, combo of 8065 uP, DUCE, and DARC. Since DUCE/CART/EBC and RAM were typically packaged together, is RAM also packaged in the EPIC?

I didn't show it, but the XDT2 also has a DIS 8 chip. However I'm assuming it'll be communicated to via one of the outputs, not via MBus. It was mentioned that it might be SAW or IDM, neither of which meant anything to me. Are those simple pulsed protocols or protocols that require a full-up COM peripheral like the CART/UART/EBC? Also I'm assuming the DIS 8 chip, gives feedback via those protocols, so the EEC can be made aware of whether a cylinder is misfiring or if a coil impedance is off. I know that's possible since my 97 Ranger V6 gave me trouble codes about a specific cylinder shortly after changing the plugs. Whatever was wrong with the defective plug caused a horrible misfire during a trip back home from Florida (I live in north Georgia). Fortunately it didn't start acting up until I was relatively close to home. The EEC picked up on this, threw the CEL and via OBD-II can tool, was able to tell me which cylinder was misfiring. That was particularly helpful as all I had to do was replace that cylinder's plug and voila, the problem was solved. Point being, if the 97 Ranger V6 EDIS EEC could detect and report this from a coil pack, then I'd think the 97 Explorer/Mountaineer XDT2 (also an EDIS EEC) could/would do the same thing.

All this does have my curiosity high enough that I'm tempted to de-lid the 97 Ranger's EEC and get pics of it just so we can see how similar/different two EEC-Vs in 97 were. If anybody is also interested, I'll try to get that done this weekend.

[sailorbob]

The only details on the AICE I have seen is in patent number 5,506,777.

EDIS is pretty well documented, e.g. http://www.dainst.com/info/edis/edis.html and in more detail patent number 4,922,874.

[cgrey8]

A Google-search only found 1 entry for that Ford patent and it appears to be a rather poor quality scan of original paperwork:
Official Gazette of the United States
Download the PDF and go to p652.
Here's an image of the text:

FordPatent5506777_AICE.jpg

I could zoom in and get the image bigger, but it didn't really get any clearer. What's worse is it appears to be written in English, but it doesn't make much sense to me.

[cgrey8]

Here's the pictures of the XAD2 EEC (97 Ranger 4.0L 2wd Automatic).
I expected it to be very similar to the XDT2...but no.

Here are the relevant pics:

XAD2_Delidded.jpg

N8065FB.jpg

N81C66DB.jpg

N70005FWCHCA.jpg

And there are 2x of these chips. One (pictured) next to the 18MHz crystal and the other you can see just the pins of in the picture:

N81C69BC.jpg

Since I can only attach 5 pictures per post, I excluded the closeup of the SC370674FN since you can see it clearly in the delidded view of the entire EEC.

There is the EDIS 6 chip up close to the wiring harness connector.

So the questions here is what are the N70005 and 81C69s? And given there are 2, seemingly identical 81C69s, is there any chance they are using Bank Select lines to differentiate them?

I don't have a BIN file for the XAD2 nor do I know what strategy it follows.

[sailorbob]

The '81C' part of the 81C69 name suggests it's RAM, no idea if there's extra functionality added.

The 32 pin chip looks like it's a variant of the 32 pin chip in the XDT2 (maybe 216 kB instead of 112 kB?).

The XAD2 uses the ANAJ7 strategy. I haven't seen the calibration from that either.

[AuroraGirl]

The 1997 PCM should have no issue identifying the cylinder misfiring as long as its OBD2 compliant because OBD2 regs require the PCMs to have more strict monitoring of catalyst-damaging misfires(misfires post a certain threshold, esp the flashing CEL kind)

And the monitoring for the cylinder doing it is to help get the issue fixed. codes for the cylinder can sometimes pop up, but the misfire tally is going to be helpful a lot!
My 96 4.9 i6 had 4 5 and 6 misfiring only and I found the wires were separated internally and that means had to jump internally to continue to the plugs. Crazy

[jsa]

I have spent some time ohm meter probing the 8065 boxes I have available.
I've added the part numbers above for XDT2 and XAD2 but do not have them to probe.

8065 IC Table.PNG

MBus goes from J3 to uP, PROM and RAM.
MBus does not go to the SC370695 type I/O chip.

Some part number information can be found in;
Handbook K2-K4.
Hardware Reference 1-1,1-2, 12-5 and D-1 to D-3.

EPIC is using a 7500 series part number which is designated for RAM in the hardware reference.
A couple of boxes have IC number marking as per the handbook. EPIC has the uP number IC1H.

I'm of the view EPIC is an 8065/RAM and other functions.

There is information on AICE and DARC in the available strategy documents.

XDT2.bin has code for AICE.

The documents indicate DARC is on MBus, so it would be good to probe and get the BIN from XAD2.

The documents suggest the C in 81C is CMOS, F would be Flash.

[jsa]

The complete AICE patent;

US5506777 AICE.pdf

[jsa]

@TVRFan,
Yes, some have IC's on the reverse side.

20221223_202822_resized.jpg

20221223_202841_resized.jpg

[sailorbob]

OK, a bit more information on the eec-v EEPROMs:

The 56K, 88K, 112K, 200K and 216K EEC variants use 87F66/88FF80/88F112 and later derivative EEPROMs (not sure why the 88F80 is given as '80' at the end instead of '88' unless it's a typo).

The 220K EEC uses a AMD-29F040 EEPROM.

[jsa]

Sailorbob,

Do you know of any catchcodes or hardware codes for the 200k and 220k boxes?

[sailorbob]

I unfortunately do not, the 'catch' code list I have is very spotty on the last of the eec-v. I am not even sure the 200k memory made it into a production vehicle as I've only seen reference made to it in conjunction with development tools.