-
Protip: Profile posts are public! Use Conversations to message other members privately. Everyone can see the content of a profile post.
Honcho 3.3L ITB Build Thread
- Thread starter Honcho
- Start date
Thanks for all of that great info! Fortunately, Matt did finish the OBD-I definition for the 91-94 US cars, so I have that. It seems like the trick is then to use tuning software (TunnerPro, for example) to work with the emulator to make real-time changes to the fuel maps on the dyno and while driving and then burn a chip with the tuned values on it. Andreas [MENTION=12723]greenberet[/MENTION] mentioned that he took his NSX to a European tuner, who could read the fuel and timing maps right off the ECU, and then programmed a chip to work with his Comptech cam setup. I wonder how (15+ years ago) that tuner was able to read the ECU EPROM without going through all of the painstaking work Matt did on extracting the NSX definitions? Is there some other kind of system out there?
For budget reasons, Stage 2b of the Path to Imola plan will just be a cam change with new OEM valve springs while the engine is out for the 6-speed conversion. This is because the cams and valve springs need to be changed from the auto-spec to the manual-spec in order to work with the factory ECU (apparently clipping resistor R4 on the ECU board will switch it over to the manual maps). It just happens that, if I'm going to have to replace the cams anyway, might as well go for a hot cam.
Because the Comptech and KSP cams are identical to OEM on the small lobes, no tuning would be necessary if you keep the car out of VTEC. The only thing requiring tuning would be the VTEC fuel and timing curves. It seems in this instance, a simple chip could suffice like it did for Andreas. His NSX has been running reliably for years and years with this setup.Stage 3 will be the full 3.3L ITB build, so I agree full engine management (AEM Infinity 6 with dual widebands) will be necessary at this point. But, in reality, Stage 3 is at least 5 years out on the current budget timelines, so if I can save the dollars with just a quality chip tune, I'd rather go that way.
The kit is still available on the website. Honestly, I was all gung ho for ITB until I read Roman's journey here:
http://www.nsxprime.com/forum/showthread.php/182001-quot-solidol-quot-94-BB-maintenance-build/page5
He went so far as to buy the kit and...still sold it without installing. I'm picturing my Type-S Zero stalling every time it rains, or sputtering on a touge run from 6,000ft to 10,000 ft, hesitating when I go to pass on the highway, or just not starting on a cold morning- all issues with ITB cars that do not come with factory ITB option. If what I am after is just that beautiful NSX induction sound, is the DF intake scoop the real answer? DMS manifold?
More NSX Prime crickets...
Yes, that's what someone mentioned. Still on their website but no longer selling it. What do I know.. I don't care enough to call and confirm with Chris hopefully there are still ITB kits for the NSX for sale.
Also, yes... I've had many discussions with Roman about the pros/cons of ITBs. I've had many more conversations with my tuner and Adnan about it. I believe all the issues can be solved with the proper implementation. Afterall, we know how awesome the BMW OEM M ITBs are.
Also, yes... I've had many discussions with Roman about the pros/cons of ITBs. I've had many more conversations with my tuner and Adnan about it. I believe all the issues can be solved with the proper implementation. Afterall, we know how awesome the BMW OEM M ITBs are.
The ability to see what memory location is being accessed at a particular operating condition (RPM, MAP) has existed for a long time. If you are running the car on a dyno and you have a limited number of memory locations that you are interested in modifying (i.e. only the areas of VTEC operation), recording the location of those cells, making a decision on what to change them to and then modifying the cells is doable. Doing this without a GUI interface such as TunerPro, is more work and difficult. That was the basis for my comment about wanting someone who has done this before. You don't want any " ah, was 0xAAFF00F8 the memory location or was that the memory value?". Fluency in Hex is highly desirable. In the absence of TunerPro (its only been around since 2003) and a matching definition, I am guessing that Greenberet's tuner was fluent in Hex. His tuner likely just did a small subset of the definition, enough to accommodate the cam change.
If the definition is complete and works with TunerPro, then life is much easier. TunerPro appears to play nicely with the Ostrich. That probably leaves you with a lot more tuners to choose from.
One observation. I read a comment on Prime somewhere that the NSX ECU has multiple fuel maps. That comment mystified me. If your comment about the auto and manual ECUs being the same except for R4 is correct, its likely that 1/2 the maps are for the auto and the other half are for the standard and that selection is made with R4. Don't mix the two when modifying. I still don't know why you would require multiple fuel maps. One fuel map, an EGR map, a VTEC on-off map, a manifold butterfly valve on- off map and an ignition map would seem to do it. Perhaps there are other maps to support the fail safe and limp modes?
I had the M in mind when I was thinking this project up. If BMW can make it work OEM, why can't a well tuned infinity do it? The trick is idle control (the SoS kit integrates the factory EACV into the vacuum block, which I think is genius). I think the VE tuning of the infinity would solve most other ITB issues, as it will adjust based on any environmental change. However, I'm not sure about that. Maybe it is what drove Roman off the idea?
According to Matt, there are the following maps in the ECU:
Fuel Maps
1. Low Cam
2. Low Cam EGR ACTIVE
3. VTEC Hi Cam
4. CTS Failure Backup Map
Ignition Maps
1. Low Cam
2. Low Cam EGR ACTIVE
3. VTEC HI Cam.
Because the low cam lobes are identical between OEM and aftermarket cams, no tuning would be required for these maps, except perhaps for some slight adjustments due to better flow from headers, ported manifolds, etc. Virtually all of the tuning would be on the VTEC fuel and timing maps. It sounds like if I de-cased the Ostrich, I could fit it inside the ECU chassis and wouldn't even need a chip!
[MENTION=18194]Honcho[/MENTION] if you are tuning for ITB's I can probably cook up some custom code to blend AlphaN(tps)/SpeedDensity(MAP) based on your criteria. [MENTION=26435]Old Guy[/MENTION] I shared the definition publicly, on this forum almost 4 years ago. http://www.nsxprime.com/forum/showt...d-91-95-Legend-Complete-ECU-Tuning-Definition
-Matt
-Matt
Last edited:
I remember the posts that you did starting back in 2010. It was interesting; but, having an OBDII car that I did not intend to modify, my interest was more in your process than the end product. My recollection at the time was that the definition was not complete; but, I stand corrected.
Hi Matt-
ITB tuning on the stock ECU makes me nervous, as it is not a VE tuning model. The nice thing about VE (on the AEM, for example), is that in theory it is much more able to adjust for changing environmental conditions. It knows the VE of the engine at every load state, so it really just chases a specified AFR for that state. Thus, if you drive from my house (6,000 ft) to the top of the Vail pass (11,000 ft) and back down, the ECU won't care- it will just adjust fuel and timing on the fly. If I understand it correctly, the NSX ECU is preprogrammed with a set fuel and timing value for every load state. There is a chart that never changes, so at 50% throttle at 5,000 rpm, the chart says open the injectors XX ms and set timing at XX degrees. But, the problem is that at 11,000 ft the car needs less fuel than at 6,000 ft, and now you're running rough over the Vail pass and fouling your cats. Do I have that right? How do you overcome that issue with tuning? Presumably, sometime in 1989, Honda drove a NSX up and down a mountain over and over to set the initial tables. Would I have to do that again?
If the NSX ECU can be made to play nice (e.g., OEM-like) with ITB, that would be AMAZING (and save me $2,000 on the AEM Infinity). But, what I'm really after is using it to work with the first stage of the tune, which is just bigger cams- just like the Honda tuner did for [MENTION=12723]greenberet[/MENTION] apparently 15+ years ago.
I noticed in this build thread that the owner, who is an Acura tech, did not mill the heads or deck the block before installing the cometic MLS head gaskets.
http://www.nsxprime.com/forum/showthread.php/151472-New-owner-Mini-fix-it-build-thread
That car ran fine all the way up to 15 psi of boost. I'd love to replace my 1992 3.0 graphite gaskets with the cometic, but I don't want to tear down the short block and send it out for machining. As many of you know, I have a bad history with head gaskets and I would love to close off that risk with competic MLS and ARP. Is this possible? Is the mill/deck requirement for pure metal gaskets only? (the cometic have a coating on them).
http://www.nsxprime.com/forum/showthread.php/151472-New-owner-Mini-fix-it-build-thread
That car ran fine all the way up to 15 psi of boost.
I'd love to replace my 1992 3.0 graphite gaskets with the cometic, but I don't want to tear down the short block and send it out for machining. As many of you know, I have a bad history with head gaskets and I would love to close off that risk with competic MLS and ARP. Is this possible? Is the mill/deck requirement for pure metal gaskets only? (the cometic have a coating on them).
about 5yrs ago my stock HG started weaping in Cyl #5 . time for a new HG. I went cometic and the head was simply trued to flat and the block was "cleaned up" (no milling or blocking of the sort). Car still feels strong but it's actually getting needed maintenance as I type and i'll know the leakdown numbers soon.I noticed in this build thread that the owner, who is an Acura tech, did not mill the heads or deck the block before installing the cometic MLS head gaskets.
http://www.nsxprime.com/forum/showthread.php/151472-New-owner-Mini-fix-it-build-thread
That car ran fine all the way up to 15 psi of boost.
You don't want to mill or deck too much because timing chain issues could be a problem then you have to do the timing gear and a lot of that rigmarole.
Your assumptions are unfortunately misguided. A fuel map is, in essence, a map of volumetric efficiency.
I can tell you haven't taken a look at the definition
. There are far more adjustments with even this 26 year old ECU than with any aftermarket platform, including barometric pressure compensation (built in pressure sensor on ECU board).-Matt
Your assumptions are unfortunately misguided. A fuel map is, in essence, a map of volumetric efficiency.
I can tell you haven't taken a look at the definition
-Matt
I guess my ignorance bled through that entire post. :smile: The way AEM sells the infinity is by stating that factory ecus have fixed fuel and timing tables that get thrown out of whack when you change components or environmental conditions, but the infinity starts from the other end, which is to specify a AFR and then adjust fuel and timing to hit it. Guess I fell for the marketing. If the ECU can be tuned for ITB (or cams), that is HUGE. It seems that TunerPro and the Ostrich gets you there, provided you have a good tuner. I knew the ECU had a baro sensor on the board, but did not realize there was such sophisticated adjustment available. Looks like I need to download TunerPro and take a look at the definition. I'm just a dumb lawyer, but hopefully I can educate myself enough about tuning to understand what I am looking at!
The governing equation for a MAP based fuel injection system is the ideal gas law
n = PV/RT
As applied in the fuel injection system, P is the manifold pressure, V is the volume of the cylinder and T is the temperature of the air entering the cylinder, R is a constant and n is the number of moles of air. Multiply the right side by the mole mass (Mm) of air which is treated as a constant and you convert moles into the mass of air in the cylinder
air mass = Mm (PV/RT)
Once you have the calculated mass of air in the cylinder, divide this by the air fuel ratio to get the mass of fuel that must be injected for a stoichiometric mix.
fuel mass = Mm (PV/RT) (1/14.7) - for gasoline
The above formula assumes that we actually know the pressure and temperature of the filled air in the cylinder. We don't, so we add in a fudge factor that we have the hubris to call the volumetric efficiency to adjust for this. In practice Ve corrects for the pumping losses (efficiency) of the intake and exhaust system plus all the other non linarites.
fuel mass = (Ve)(Mm)(PV/RT)(1/14.7)
You will know your injector's flow rate (in grams/ sec), so you can directly calculate the injector pulse width required for each combustion event
pulse width = (1/injector flow rate)(Ve)(Mm)(PV/RT)(1/AFR) - I modified the equation to allow for operation at something other than 14.7 operation.
This is the general form of the fuel equation for a MAP based engine. It does not deal with warm up and all the other gory details. Depending on the speed of the ECU, it may recalculate this value every combustion cycle. Once the engine is configured, injector flow rate, Mm, V and R are all constants. The formula then simplifies to:
pulse width = K (Ve)(P/T)(1/AFR) - K is the combined constant of injector flow rate, Mm, V and R
The AEM and other aftermarket systems allow you to separately enter table values for Ve which you can treat as an engine configuration value and AFRs which you can treat as a tuning value. This is a conceptual 'thing' which in part reflects the fact that these ECUs are being applied to multiple engines. The Honda ECU is only being applied to one engine. Honda knows the VEs and AFRs they want, so they can combine the two tables to create a pulse width table based on something like this
pulse width = K'/T
K' is an even more 'combined constant' which includes the Ve and AFR for the operating point. Because P is automatically defined by the vertices of MAP and RPM in the 'combined' look up table, the P value can be pre calculated into into the value stored in the table and the only real-time adjustment that has to be made is to correct for temperature.
The AEM operating in MAP mode and the Honda ECU are effectively identical in the fuel equation. Anything that the AEM can do in MAP mode the Honda ECU should be capable of doing. Storing the calculated pulse widths directly and then adjusting for temperature is more computationally efficient than storing AFRs and Ve values and having to fetch those values and do the extra calculations. However, this is no longer 1991 and processors have changed so those advantages may be less of an issue. If you were perverse, you could take a crack at back calculating an equivalent Ve map and AFR target map for the Honda ECU based upon the pulse widths; but, only if you seriously lack for things to do.
If a MAP based system is tuned at sea level with the top row of the tables configured at 100 kPa, that engine should be just fine operating at increased elevation. When somebody drives from Seattle to Denver, the only thing that happens is that the engine MAP probably never gets much above 80 kPa. The ECU and the fuel system operation is indifferent to whether the 80 kPa is caused by the fact that the throttle plate is causing a manifold pressure drop or there just isn't any air pressure above 80 kPa period. A MAP based NA system tuned in Denver can have problems at sea level. The maximum MAP value available for real time tuning will only be 80 kPa. If the ECU is only configured to a maximum of 80 kPa and then driven at lower altitudes, it will fall off the upper row of the fuel map likely running lean. A savvy tuner could fill in the table values above 80 kPa to allow for safe operation at sea level. Operation could be safe; but, likely suboptimal.
I inferred that on a given engine, the Ve values were fixed. This is mostly true. Sr5guy refers to barometric pressure compensation. OEM ECUs do this, probably to optimize emissions compliance. Some of the aftermarket ECUs that I have seen have this capability. It comes in two forms. They grab the baro off the MAP sensor just before start up and save it or they have a totally separate atmospheric MAP sensor which allows constant monitoring of the barometric pressure. Barometric pressure is a factor because it affects the volumetric efficiency of the engine on the exhaust cycle. Under identical input operating conditions (MAP, RPM, T), an engine operating in Denver will actually have a slightly higher Ve value than it will at sea level because there is less back pressure on the exhaust system. The effect is small which allows most tuners to get away without configuring it. However, the bigger factor is that it would be exceedingly difficult to configure on a one off engine requiring testing over a range of back pressures. Also, change you exhaust system and chances are that you screw up the correction.
So, the one sentence answer is that you don't need to repeatedly drive from sea level to Vail Pass to configure a MAP based ECU (Honda OEM, AEM or whoever). If you want to have a full range of altitude driveability, you would dyno tune it at sea level or someplace where you can get true 100 kPa barometric pressure (don't confuse station barro pressure with the actual barro pressure). You could dyno tune in Denver; but, then you would have to fake the tables above 80 kPa to allow for safe; but, suboptimal operation at sea level.
I have no direct experience with alpha X N systems. My gut says a pure alpha X N system would be altitude sensitive; but you may be able to use a barro correction to scale the whole alpha X N fuel table up or down based on barro pressure. I know that people have discussed blended systems alpha X density X N systems on ITBs; but, I again have no experience. If you can get a MAP system to work with ITBs, that should be capable of dealing with changes in altitude. However, I conceptually struggle with the idea of getting a useable MAP signal from ITBs. Signal filtering and blending the individual ports in a manifold may help / work. Again no direct experience to comment on; however, I see that Honda and others have applied EFI to single cylinder engines as small as 250 cc. The EFI that Honda used on its 1982 CX 500 turbo motorcycle was a MAF based system so clearly doable with MAF. I don't know whether the new systems are MAP, MAF or alpha X N with barro; but, they clearly work and work well. In the longer term if you want to go with ITBs, you might be better served checking out the EFI systems that are being applied for the two wheel set.
Last edited:
[MENTION=26435]Old Guy[/MENTION] I haven't seen a post that long in years.
If you were able to crack open a disassembler and peek at the code running the stock NSX, I bet you'd change your mind quickly about aftermarket ECU's. I have successfully reverse engineered almost 2 dozen unique platforms from foreign, domestic and motorcycle manufactures and am intimately familiar with their control strategies. No code I have encountered has been as tight, modular, logical and visually pleasing as the 90's Legends and NSX's.
Unfortunately I don't have time for a long response, but I'll let the picture do some speaking.
This photo shows a subroutine extracted from a Legend ECU (the NSX is very similar) that builds a base fuel multiplier (one of many).
If you look closely you can see the basic math being done on the individual multipliers to consolidate them into one - "FuelMultBase".
If you were able to crack open a disassembler and peek at the code running the stock NSX, I bet you'd change your mind quickly about aftermarket ECU's. I have successfully reverse engineered almost 2 dozen unique platforms from foreign, domestic and motorcycle manufactures and am intimately familiar with their control strategies. No code I have encountered has been as tight, modular, logical and visually pleasing as the 90's Legends and NSX's.
Unfortunately I don't have time for a long response, but I'll let the picture do some speaking.
This photo shows a subroutine extracted from a Legend ECU (the NSX is very similar) that builds a base fuel multiplier (one of many).
If you look closely you can see the basic math being done on the individual multipliers to consolidate them into one - "FuelMultBase".
- The first line is a multiplier that is based on an onboard resistor value. There is a resistor you could change (or install a potentiometer!) to globally adjust fueling.
- The second is an intake air temperature compensation.
- The fourth line is barometric pressure compensation.
- References to the IMA (intake mixture adjuster) sensor are for "gulf spec" cars sold in the middle east without cats or oxygen sensors. They have an adjustment pot installed under the dash that behaves much like the screws on an old carburetor. You can see the condition for it's computed fuel multiplier to be active - the car must be in active "idling" mode and an internal firmware option needs to be set to turn the whole shebang on (and disable the oxygen sensors).
- Towards the bottom you can see that there is even a barometric compensation applied to correct for the EGR valve opening.
If my long winded reply came across as favouring aftermarket ECUs, then I screwed up in my communication. Being application specific, I have no doubt that the code for the NSX will be more elegant and execute faster and deal with more things (aftermarket ECUs not being emission compliant and generally not having to manage annoying things like EGR valves). I think individuals who have fitted aftermarket ECUs and struggle with the whole start / warm up / idle / operate at anything other than WOT should probably have an appreciation for the level of sophistication of the code and the amount of work put into configuring the OEM ECU. My only comment about 20 year newer aftermarket ECUs was that hardware has changed which perhaps lessened the need to write compact elegant code that executes fast.
My primary intent in dredging through a greatly simplified version of a MAP based fueling equation was to try and address Honcho's question about whether the NSX ECU fuel map could be modified like it can be on an AEM. I was primarily attempting to show that at the fundamental level, a map of fuel pulse widths is the same as a map of AFR multiplied by a map of Ve. If you have the hardware and knowledge to modify the fuel map locations on the PROM on the NSX ECU, then its 'effectively' as tunable as the AEM ECU in MAP mode.
As to whether you can get the MAP based ECU to work nicely with ITBs, that is to be discovered by somebody else.
Edit:
This is a bit off topic since it is not directly related to Honcho's ITB build; but, while I have your attention and since you are familiar with the firmware, does the NSX ECU manage short term exhaust gas correction separately on the front and rear cylinder banks (i.e you can be running separate pulse widths on the front and back cylinders) ? Also, is the long term fuel trim a single global value applied to all the cells in the fuel map or does the ECU calculate and store a long term trim value for each cell in the fuel map (with separate values for front and back)?
I believe that some ECUs in the '90s implemented long term fuel trim using a single calculated global value applied to all cells in the fuel map. My understanding is that strategy is pretty much history with modern ECUs calculating long term trim on a per cell basis. The description of the long and short term fuel trims in the NSX service manual is a little hazy. My reading between the lines suggests that long term trim is a global value which would be consistent with what was common practise in the '90s.
Last edited:
I've done some more research here. The Viton coating on the Cometic gaskets is specifically there for rougher surface compatibility (50 Ra), so assuming the heads are flat and in spec, you only need to properly clean the mating surfaces to get a good seal. As with everything, this is where issues arise. If you do it like Mile High Acura and use a power tool with a spinning abrasive disc, you ruin it. If you use a good gasket remover (CRC), a plastic scraper, isopropyl alcohol and some patience, you get a good seal. That said, I'm still going to use ARP head studs to be sure.
Thank you for that incredible explanation. Yes, that is what I am after. AEM sells the Infinity by pointing out that, once tuned, if you make a change to the engine, on a MAP-based ECU, you need to spend a huge amount of time re-programming each cell in all of the tables to properly adjust the tune. On the Infinity, the computer adjusts the tune for you based on its calculated improvements in Ve and uses processing horsepower to muscle through all of the back-end calculations. Believe me, I would rather just clip resistor R4 and tune the existing ECU when I put the cams in during the manual conversion. But, I'm concerned that any more aggressive mod like ITBs down the road will be constrained by the ECU. Matt seems to be saying that with the NSX definition and TunerPro, you can get to the same place as the Infinity, but with a much more elegant tune. Still, I'm not convinced it is an accident that all of the SoS ITB cars out there that I'm aware of are running Infinity engine management. It seems more and more, that a good tune with TunerPro will get me where I need to go with the cams and we can table the ITB issue until the time comes.
This is just my opinion; but, don't get too caught up in the 'self tuning' / 'self configuring' / 'self anything' function. I am not directly familiar with the AEM; but, the self tuning function generally takes two forms.
The first is where you attach a lap top to the ECU and operate the engine over a range of conditions. The lap top software monitors the exhaust gas correction and will then display a table with 'suggested' changes to the Ve values that will more accurately hit the target AFR values for each cell in the Ve map (if changes are required). You have the option of deciding whether you want to accept those changes. This can be a useful tool for doing a final refinement on Ve map - my objective being to try and minimize the amount of exhaust gas correction that is being applied in any cell. You do need to treat this software with care as I have had it provide contradictor changes to Ve cells on subsequent test runs. You need to understand how the software is calculating the recommended change to VE. Is it a weighted value of all the 'hits' on the cell? How long does the engine have to be running in that cell before it counts as a hit (EGO corrections at wide open throttle are not reliable because of the transit time through the engine).
The second 'tuning' feature is where the ECU continually monitors exhaust correct and calculates a 'fuel trim' value, either globally or on an individual cell basis, which attempts to minimize the exhaust gas correction required to hit the target AFR. The NSX ECU does this and I expect that all modern emission compliant cars have this feature. The objective on emission compliant cars is to make sure that the AFR remains spot on at 14.7 which was the most efficient operating point for the 3 way catalysts. The fuel trim feature helped these cars to remain emission compliant as they aged. This is a great feature on an emission compliant car. I don't like it on a performance engine with an aftermarket ECU. If something goes 'off', the fuel trim correction can mask the problem which can then become a bigger problem (things tend to go south on performance engines faster). On the NSX ECU, the ECU calls a halt to the whole fuel trim thing if it gets somewhere around 20% and then throws a CEL telling you that you have a problem.
As an observation, on an aftermarket ECU which separates Ve from AFR, both of these features adjust the Ve values to hit the target AFR value that someone has entered into the AFR table. To me, tuning for torque / power is about finding the AFR that gives you the maximum value. AEM is correct that after a change to the engine, the software could help you adjust the Ve to get the engine back to the original target AFR. However, that kind of misses the point that if you make a significant change to an engine, you typically need to change the AFR to maximize torque / power. You need a some dyno time for that. None of this software will tell you what AFR you should be running at 80 kPa/6000 RPM to maximize torque / power at that point.
The software tools that the aftermarket vendors provide can be useful 'tools'. Just be informed about what they actually do so that you can make an informed choice on OEM versus aftermarket or which of the aftermarket ECUs you want to go with.
Last edited:
Coz also used copper spray on that to help seal. The C30 engines do not come from the factory with a 50ra or smoother surface that Cometic recommends. Some good reading begins at posts 15 and below here:
http://www.nsxprime.com/forum/showthread.php/197835-Engine-need-advice?p=1913142&viewfull=1#post1913142
Personally, I would want to have the correct Cometic-recommended surface finish if I we're doing an expensive engine rebuild, but the copper spray seems to work. Definitely don't want to just slap a Cometic HG on a C30 and call it a day with just minimal cleaning.
Notice I just say C30. The C32 has mls-style HGs from the factory and their surface finishes may be smoother.
Rob from DAL also said it here:
http://www.nsxprime.com/forum/showthread.php/89427-what-head-gasket-are-you-guys-using?p=829018&viewfull=1#post829018
My $0.02.
Last edited:
Imola painted coupe for sale. Needs love:
https://newyork.craigslist.org/jsy/ctd/d/1991-acura-nsx-slick-top-5/6658544352.html
https://newyork.craigslist.org/jsy/ctd/d/1991-acura-nsx-slick-top-5/6658544352.html
It seems that the NSX ECU can do all of the above with TunerPro RT. With the starting definition that [MENTION=16606]sr5guy[/MENTION] provided, isn't it really just a matter of adjusting the pulsewidth and timing cells in each map to develop maximum power at a safe AFR (instead of the emmissions-ideal 14.7)for each load state? If the ECU also has a baro correction curve built into it, then perhaps your 80 kPa tuning concern is not as much of an issue. Asuuming you get an ideal map at 80 kPa, Matt seems to be implying that the ECU can self-adjust from there as atmospheric pressure changes. I might be getting this part wrong, but I don't think Comptech sold any kind of tune or engine management change with the original IEM package back in the 90's. If that's true, perhaps they thought the factory maps were still safe with the bigger VTEC lobes, even if not ideal. From [MENTION=12723]greenberet[/MENTION] experience, it seems even the stock 240cc injectors are up to the task for the cams and enlarged TB/manifold. I think Matt and you have me convinced to just tune the car on TunerPro RT with the Ostrich 2 and/or a burned chip. It will definitely work for just the cam change and likely will work even with ITBs. Matt also suggested using a 27SF512 chip instead of the 27C256 stock chip for compatability with the Moates chip burner. An added benefit is that the chip has more memory than the factory chip, so you could theoretically put two maps on it- a "stock" map for emmissions checks, and a "performance" map with the fully tuned values. One would have to figure out a way to switch between maps though- maybe one of the clock button switches? Or the TCS switch? (I'm deleting TCS)
Fel-Pro sells a MLS gasket that is rated for 60 Ra surfaces, specifically to avoid expensive machining of the head and deck surfaces. Part of me wonders if it is just better to replace the OEM graphite gasket and use ARP studs at 85 lb/ft and avoid the whole MLS nonsense. Another option is just too check the torque of the head bolts when I open them up for the cam change and only replace the gasket if the bolts are loose. I'm wary of the copper seal stuff- it is what Mile High Acura used on my new head gaskets and those let go shortly thereafter. Cometic specifically says NOT to use any spray sealer on their gaskets.
Already bought a car. You're right though- that one needs love and is about $10k overpriced.
You're right - Cometic explicitly says not to use any sealant... but it appears some have had success doing so here. I don't condone it for the record.
I wasn't aware Fel-Pro sells a HG now. Sounds like it might be a good option for some.
Last edited:
IEM - I missed that reference and focused on the comment about cam lobes. Did the CT package include cams with revised profiles? If it just included Intake and exhaust modifications, that would probably work with an OEM fuel map. If it included cams with a significantly more aggressive profile, it would be hard for those to work well with an OEM map.
The package was called "Internal Engine Modification." If you ordered the whole thing you got:
1. Big cams (hardwelded- not regrinds)
2. 36mm intake valves
3. New valve springs and retainers
4. Ported intake runners
5. Valve job
6. Milled heads (slight compression increase)
You could order it by piece. [MENTION=4282]docjohn[/MENTION] bought the whole package. @greennberet, if I recall correctly, just got the cams and intake mods. There was, the doc just confirmed, no ECU tuning associated with this package. His car put down almost 300 whp on the stock ECU with this package (and I'm assuming CT headers) and then a valve spring broke and destoryed his engine.
Andreas also got close to that with this modifications, but he had a tune done. Andreas kept his Honda valve springs and has been driving on them problem free for over 15 years. What this means for my build is that the stock ECU can adjust enough for these mods even without a tune.Similar threads
- Locked