Late series engine compatibility with earlier series engines?

I have never heard of a Z32 32bit ecu. 8bit and 16bit only.

 

Yep, what he said.

MoulaZX

 

Could be mistaken on that Vipec quote, it was a loooong time ago. Could have been just for installation. So maybe tack on another $500-800 for a Tune. But yea, if you really wanna go nuts with your dosh... there are plenty of ways my friend, lol.

MoulaZX

 

Haha, that'd be right

I'm sure I could spend 2-3k elsewhere haha

 

Clarification time guys...there is no "32-bit" ecu...the ecu's are.

8-bit OBDI pretty common, it's what most of the early to mid 90's zeds have
16-bit OBDI not as common, but out there, can be found on some 95 cars, but all 96 onwards. Outwardly look identical, the same plug and pin-outs however the injector connectors on the loom are different as well as a few other minor changes.
OBDII USDM zeds only, and for the state of california's emmissions laws. Nil, Zip, Zero, Nadda, None, saw a right hand drive car unless you've imported a motor, loom and ecu from the states.

Consult will work fine on either 8-bit or 16-bit ecu as it talks through the OBDI (on board diagnostic) interface. 16-bit or 8-bit ecu's are interchangable with no other mods. ie, turn car off, unplug 8-bit ecu, plug in 16-bit ecu and start car.

Nistune boards use a 28-bit flash-prom.

I will safely say that there is no Nissan OBDII z32 ecu's in Australia

 

Quote from Ashspec, that seems to detail the 2 types of 16 bit ECU, but says the "early" 16 Bit are not interchangable with 8 bit cars.

http://www.ashspecz.com/shop/prodInfo2.php?prodID=14

 

I don't think that the USofA ever got a 16-bit OBDI ECU, if this is the case then it explains Ash's answer.

The maps on the 16-bit ECU are slightly different to take advantage of 16-bit over 8-bit, it's true. But pin-outs are the same.

Proof? My car had a 16-bit ecu in it...I fried it by shorting out an injector driver and replaced it with an 8-bit ecu. Granted both were chipped with my tune, so I never ran it with the stock rom. I still have that ECU in the garage, so I will photograph it for you.

 

Cheers mate, I'm sure this should be thrown in the tech section!

On any account, I'm in the process of getting another 16 bit ecu (cheers Ryan), but on account of my lazy-ness and being so busy, haven't had a chance to tee it up properly.

Not to mention the huge financial wall I'm still contemplating weither or not to climb/attemp just yet!

 

The 16 bit ROM is 64k, how did you chip it to the 8 bit (32k ROM) ECU ?

 

I used tunerpro and an emulator to write my tunes and there is a configuration file for the 16-bit ecu. It was easy enough to rewrite the data into the 8-bit config.

Sorry if I mislead, it wasn't a case of a chip swap, but I didn't need to change the loom.

 

No worries Andy thanks for all the info.

Did you alter the "closed loop area" on the main fuel map, on 16 bit ROM's it looks like the entire main fuel map is closed loop in Nistune.

I assumed the 16 bit processor had more capacity to allow this.

Did running full closed loop on a 8 bit ECU cause a problem.

Care to comment if indeed 16 bit ECU's run full main fuel map closed loop ?

 

I'm a little unique in this regard, because I was running E85 I disabled my closed loop functionality as I didn't know at the time whether how the narrow band O2's would work with the AFR's I was aiming to achieve in the map.

The 16-bit rom does not have a seperate portion for closed and open loop running, which is interesting, it uses a value table instead which the AFR table refers to for it's correction factor back to 14.7. Stock tune has the entire map in closed loop. But the TP and rpm corner of the map has a correction value of almost nil. So if you can imagine the afr map, as it richens it's mixture, it simularly decreases it's correction value.

So with 16-bit ecu's it is critical on increased boost applications to not only alter the AFR's, but to reduce the correction values as well. The last thing you want to do is have the ecu in a closed loop scenario with the factory type O2 sensors at 20psi of boost. If i was writing it for petrol, I would zero out the correction factor at full boost. The only thing I want my ecu to use for AFR's is my map.

If you wish to transplant a tune from an 8-bit ecu to a 16-bit ecu or vice versa then you need to do a little data entry The cell references between the 2 don't line up nicely.

 

Nice work Andy, really appreciate it !

I cant see any advantage of running a 16 bit ECU over a 8 bit.

The extended closed loop area can actually be a disadvantage as you mention above in non stock situations, stock its effect is minimised any way by small correction factors.

I imagine most of it was emissions based changes.

Is there any benefit or disadvantage in running a 16 bit ECU ?

Other then less availability of stock tuned maps, must run Nistune type 3, or lift a SMD chip, burn and replace to "chip".

Any other differences ?

 

Yeah, thats why the US got them in 93, and then the OBDII units.

Not really...hence I wasn't concerned about replacing mine with an 8-bit.

The only reason that there is less availability of maps is because 99% of the off the shelf chips out there are written with software from the 90's written around the cell references from the 8-bit unit. I use tunerproRT with an emulator. Beauty of tunerpro is that it isn't ECU specific. It allows you to write your own references, or use any already written.

One benefit of 16-bit, theoretically, is it's resolution. The 8-bit ecu has a table resolution of 16x16 cells. 16 TP and 16 rpm.

8-bit will allow you to increase your TP and RPM range. So stock is 16 - 88TP. If you up the boost to say 18psi on stock blowers, you are reaching your max TP at around 4500rpm or something similar. So at full noise from 4500 rpm the ecu is reading from the last few cells. So with 8-bit you can increase this from 88 to say 144 but you still have only 16 cells per rpm range, so your resolution drops. Savvy? Same with the RPM scale

Theoretically...(I haven't done this, but I might play around with it over the next few days and come back to you, my 16-bit ecu still works enough that I can see if it accepts the maps)...you may be able to double the table to 32x32 without compromising the efficiency of the ecu. If this is the case, then there is an added benefit for running the 16-bit on high hp engines. But usually if your doing that, your running an aftermarket piece of kit.

 

Yep, resolution a potential benefit.

Thanks again Andy, let us know what happens.

 

confused...

so if the whole map is in closed loop, which is normally a 14.5:1 ratio, do these correction values allow for the richer air fuel ratios at higher rpms?? Or am I understanding it incorrectly?

 

Yep they do.

Soooo, what happens is seeee...

In the corrections table there are values that will range from 128 to 192. Anything over 192 is open loop. The closer to the value of 192 is in the cell (disregarding above 192) the less the ecu will pull it to stoich (14.7) and the closer it will stay to it's requested AFR from the map. The table is populated by 128 in all the areas that would typically be closed loop in an 8-bit ecu. From outside this it scales quite rapidly, and smoothly to 190 in the stock unit. At 190, there is very little change from the posted value to stoich, it's effectively open loop.

As was mentioned this was done purely for that extra little reduction of emissions

edit: I'm going to take a screen shot of the stock table and post it...it might make more sense.

 

Gees, it does get pretty rich at high RPMs doesn't it? I think my 93 jap ecu has an A/F ratio of 10.2:1 at the far end of the chart.