MR2 2ZZ Dyno run - Corolla UKDM ECU v Corolla JDM ECU

Started by JB21, November 3, 2022, 08:14

0 Members and 1 Guest are viewing this topic.

JB21

I took my car down to EFI-Parts in Runcorn for a Dyno power run and whilst there Chris done a back-to-back run of both my ECU's.

Some interesting results with power & AFR's.

Red line - UKDM ECU made 185hp @8000rpm (8250RPM limit) with very rich AFR's

Black line - JDM ECU made 190hp @7800rpm (8400rpm limit) with ok AFR's, still a little rich but not crazy rich like the UK ECU

I originally posted this on IG with the results the wrong way around with the UKDM ECU making more power, but on checking the ECU part number the JDM is the better ECU.






shnazzle

I still think vtec and 2zz have the weirdest dyno plots. 
It's be so much nicer if they found a way to bring the 2nd cam lobe in more gradually instead of which this weird artificial bump.
I can see why big power builds do away with it entirely and just opt for a single cam profile.


Nonetheless, very interesting plots confirming that the JDM ecu is better for this setup. 
...neutiquam erro.

JB21

Quote from: shnazzle on November  3, 2022, 08:26I still think vtec and 2zz have the weirdest dyno plots.
It's be so much nicer if they found a way to bring the 2nd cam lobe in more gradually instead of which this weird artificial bump.
I can see why big power builds do away with it entirely and just opt for a single cam profile.

Chris at EFI said exactly the same. He said this engine would really benifit a standalone ECU and remap, to do away with that kick at 6k rpm and smooth it out.

cptspaulding

Quote from: shnazzle on November  3, 2022, 08:26I still think vtec and 2zz have the weirdest dyno plots.
It's be so much nicer if they found a way to bring the 2nd cam lobe in more gradually instead of which this weird artificial bump.
I can see why big power builds do away with it entirely and just opt for a single cam profile.


Nonetheless, very interesting plots confirming that the JDM ecu is better for this setup.

I'm sure it's not news to many but I've read a few times that Toyota deliberately did it this way to appeal to your general petrolhead. "V-Tec kicked in, yo'". It's noticable so it puts a bigger smile on the average petrolhead face?

Interesting to see the plots though. Thanks for posting :)
Former owner 2003, 2zz conversion.

shnazzle

Quote from: JB21 on November  3, 2022, 08:30Chris at EFI said exactly the same. He said this engine would really benifit a standalone ECU and remap, to do away with that kick at 6k rpm and smooth it out.
Indeed. The Stg2 cams on a standalone on a 2zz has always been my favourite setup. Unfortunately not the most reliable. But it had that gradual buildup and no massive bump.

But then standalone brings on a sh1tstorm of issues as well.
The good thing is, by now we've got quite a few ECUMaster maps in the club that have  been slowly calibrated to work well across the year and in different situations.
We had a nightmare making it daily-driveable but then we were the first ones to have it.

I do think we've probably got some of the most mature standalone maps available within the club now. Probably need to start archiving it

...neutiquam erro.

JB21

For me a standalone ECU and remap is just not worth it as my car sits above 6k rpm most of the time on track, and the remap just doesn't add enough gains in that range for the outlay of cash of near £2k. It would be FI or nothing, and as I dont like FI for track it's going to be nothing. I'd rather spend my money on track time to hone my driving skills.

thetyrant

#6
Interesting results and good to see a proper back to back on same dyno, same day etc.  What fuel was in it our of interest

Regards the vtec/vvtli aggressive shift its got to be like this due to physical limitations of the way it switches cam profile its not an electronic thing, if you delve into how the vtec/lift etc works you can see why its like that.

Using an aftermarket ecu wont help this unless you fit a high lift cam and disable the vtec/lift and try and map around the high lift at lower rpm which is what the big power turbo honda vtec guys do, you can of course retain the dual profile lift setup and adjust its engagement point but as im sure honda and toyota have spent millions on finding the sweet spot there wont be much gain without a lot of internal mods or using a power adder like turbo etc.

If you want smooth cam profile switching then look at the BMW valvetronic system on N52 and other engines and way that works, very clever :D

:)
Ex-2005 roadster  owner, i will be back :D

shnazzle

Indeed. I meant a more progressive physical/mechanical profile switch. Not electronic. It's an on/off.
I guess what I had in mind was the valvetronic, which is much smoother. There's also another system but I forgot who made it. Mazda? Basically a continuously variable cam profile. Much like valvetronic but even more control.

There's nothing that can be done on the 2zz or vtec to remove the bump, but as cptspaulding said, it's partially artificial as they purposely drop the power a bit just before the profile switch so that it's more of a "kick". A bit daft if you ask me, but obviously it works because people love the "VTEC kicked in yo" feeling
...neutiquam erro.

thetyrant

I dont think they purposely make the power drop before the lift kicks its just the small cam running into its limit, looking at JB graph you can see power doesn't dip before the lift kicks but is just strating to flatten off, however the torque is falling away as the cam is becoming less efficient which i guess you feel more.

Ive no doubt they have played around with this a lot to try and get a good low cam and good high cam for the 2 different jobs they need to do, it still amazes me how relatively simple the cam switch is inside the head and that it works so well at rpm its turning, off course the cams are only doing half the rpm of crank but its still wipping around at a fair pace :D
Ex-2005 roadster  owner, i will be back :D

JB21

Quote from: thetyrant on November  3, 2022, 09:44Interesting results and good to see a proper back to back on same dyno, same day etc.  What fuel was in it our of interest

Regards the vtec/vvtli aggressive shift its got to be like this due to physical limitations of the way it switches cam profile its not an electronic thing, if you delve into how the vtec/lift etc works you can see why its like that.

Using an aftermarket ecu wont help this unless you fit a high lift cam and disable the vtec/lift and try and map around the high lift at lower rpm which is what the big power turbo honda vtec guys do, you can of course retain the dual profile lift setup and adjust its engagement point but as im sure honda and toyota have spent millions on finding the sweet spot there wont be much gain without a lot of internal mods or using a power adder like turbo etc.

If you want smooth cam profile switching then look at the BMW valvetronic system on N52 and other engines and way that works, very clever :D

:)

Fresh Tesco Ultimate 99ron

JB21

So the myth was JDM ECU's are set using 99ron or higher fuel and UKDM set using 95ron, this seems to be true given the 5hp gain.

thetyrant

Indeed typical of JDM ecu to have a slightly more aggressive timing curve or tweaked AFR to take advantage of the higher octane fuel, it would be interesting to pull a datalog from ecu to see what the timing difference is between the two ecus at the top end although obd is so slow might not be accurate, UK ecu really flattens off at 7k but JDM just keeps pulling.

Normally 5hp dyno difference could be put down to a even slight difference in air pressure/ air temp/tyre temp, oil temps etc etc, but the shape of graph shows the bigger picture that something else is going on it seems, either the slightly leaner AFR or more timing advance  or a touch of both :)
Ex-2005 roadster  owner, i will be back :D

Gaz2405

Quote from: JB21 on November  3, 2022, 08:30Chris at EFI said exactly the same. He said this engine would really benifit a standalone ECU and remap, to do away with that kick at 6k rpm and smooth it out.
this is what I've done with mine, smooth cam transition at 5600 by retarding the big cam timing when it comes into lift and then advance and then back down again.
1zz turbo. Home built and home mapped.

Now 2zz turbo. Home built and home mapped

Build thread https://www.mr2roc.org/forum/index.php?topic=67004.0


Zxrob

Adventure before dementia 😁

JB21

Quote from: Zxrob on November  3, 2022, 15:57Phil
Are they crank or at the wheel figures

Rob

I wish it was wheel HP 😆

threepot

This very good to see a back to back dyno of the JDM v's EU spec ECUs.

Also there is a very simple way to overcome the speed limiter on the JDM ecu if you need that.

I've been studying toyota cam profiles quite a bit recently as I'm building a 2AR-FXE engine for my car.
The very weird transition from the low cam to the lift cam is due to quite an odd reason from what I can see.
The cam timing is awkwardly different from low to high - in order to make a good transition Toyota should ground the timing of the lift cam a bit differently.
Literally the ECU/MAP has to retard the cam timing a fair bit before jumping over to the lift cam, otherwise the cam timing would be very wrong. So its not just "engaging the lift cam", it is trying to shift the VVT about to get from one to the other.
It seems a bit of a bizarre choice of cam grind, maybe it was an accident or error but seems unlikely. I would have thought they would grind the lift cam to sit in the right phase angle so they can just jump to the lift cam on the same VVT advance/retard PWM angle.

I've also swapped my MAF housing to a stock one cut out from a 2zz airbox. My AFR's are much better than using the alloy one that cam with some induction kit. Bum dyno feels that it acceleration pulls cleaner.

What MAF housing are you running?

JB21

Our American friends aren't having this back to back comparison as a fair test, as they're saying the JDM ECU which ran last with zero road miles on it to learn the sensors parameters is not showing a correct dyno plot 🤷

It mirrors the UKDM ECU for the most part, peaking higher mostly at the top end.

https://www.spyderchat.com/threads/corolla-fielder-2zz-jdm-ecu-v-corolla-t-sport-ukdm-2zz-ecu-dyno-results.160204/#post-2188913

shnazzle

Quote from: JB21 on November  4, 2022, 15:59Our American friends aren't having this back to back comparison as a fair test, as they're saying the JDM ECU which ran last with zero road miles on it to learn the sensors parameters is not showing a correct dyno plot 🤷

It mirrors the UKDM ECU for the most part, peaking higher mostly at the top end.

https://www.spyderchat.com/threads/corolla-fielder-2zz-jdm-ecu-v-corolla-t-sport-ukdm-2zz-ecu-dyno-results.160204/#post-2188913
I've seen this argument before on DannyDC2's posts wrt his intake tests. This was thoroughly punched in the gut by him promptly replying with "we reset the ECU via the battery removal before every test". 

There's something to be said for that of course but it doesn't stop the fact that the power keeps climbing when the UK ecu falls flat due to its much lower rev limit. 
...neutiquam erro.

threepot

I've looked at the eeprom data from these ECU's and there is no stored learn/calibration/trim data.
They do not even store the fault codes between key on / off cycles.
Literally the only thing possibly stored in this generation of Toyota ECU is the immobiliser code which typically sits in the first 6 bytes of the eeprom, and the VIN number which can be programmed into the next long block, but most are not programmed in.
Most of these ECUs which are from non immobiliser bases cars e.g. JDM cars do not even have the EEPROM on the PCB, so have zero save space even possible.
But they do have the complete the readyness test for one full cycle before lift will engage, and from the dyno plot it is obvious we certainly have lift!




If there were some LTFT/STFT values in operation then they would get wrote back to the eeprom like in other Denso ecu platforms, likewise for TPS high/low points etc.

Calling it invalid is fricking forum know it all mob idiocy.

I posted something on spyderchat last week, and basically got told what I was doing was impossible. I can't be arsed with the fight. Forums are bordering on toxic nowadays.

I do think the MAF housing and exhaust manifold will have the greatest effect on the ARF's though, after all the calibration was created for the stock celica/corolla manifolds and induction pipework.

JB21

Quote from: threepot on November  4, 2022, 16:38I've looked at the eeprom data from these ECU's and there is no stored learn/calibration/trim data.
They do not even store the fault codes between key on / off cycles.
Literally the only thing possibly stored in this generation of Toyota ECU is the immobiliser code which typically sits in the first 6 bytes of the eeprom, and the VIN number which can be programmed into the next long block, but most are not programmed in.
Most of these ECUs which are from non immobiliser bases cars e.g. JDM cars do not even have the EEPROM on the PCB, so have zero save space even possible.
But they do have the complete the readyness test for one full cycle before lift will engage, and from the dyno plot it is obvious we certainly have lift!




If there were some LTFT/STFT values in operation then they would get wrote back to the eeprom like in other Denso ecu platforms, likewise for TPS high/low points etc.

Calling it invalid is fricking forum know it all mob idiocy.

I posted something on spyderchat last week, and basically got told what I was doing was impossible. I can't be arsed with the fight. Forums are bordering on toxic nowadays.

I do think the MAF housing and exhaust manifold will have the greatest effect on the ARF's though, after all the calibration was created for the stock celica/corolla manifolds and induction pipework.

I'm using the 2zz manifold and a rogue custom MAF pipe. Reckon it would benefit from using the 2ZZ OE MAF housing?

Gaz2405

I take everything I read on spyderchat with a bucket of salt. Everyone is a tuner....
1zz turbo. Home built and home mapped.

Now 2zz turbo. Home built and home mapped

Build thread https://www.mr2roc.org/forum/index.php?topic=67004.0

threepot

Benefit... erm maybe?! I'm not one for hollow promises.

If your rogue one has a difference in diameter, nozzle placement and vane confutation I'm 100% sure it will have an effect on measure air mass thus in turn will effect your AFR.
After all, the toyota ECU fuels entirely on MAF & RPM's, and if you change the perceived MAF sensors nozzle velocity then you will shift the bit/column of the map table you are fuelling for.

At the same time if the MAF is too close to the air filter or the throttle body, it will skew the measurements as the airflow will be turbulent... this is the purpose of the vanes in the MAF I suspect, to try to kill the turbulence from the OEM airbox so the MAF gets a more reliable measurement. Keeping the pipe joints tidy and some sort of velocity stack on the intake will change the turbulence too.

For better or for worse, well mine certainly runs cleaner, and the fuel trims are smaller. And I feel it isn't as boggy or hesitant at certain load point. Just feels like a cleaner burn.

But its not going to help me keep up with a 911 GT3 RS