Honcho's Long Road to Imola Type-S Zero

Honcho · 27 January 2025

mwagner10702 said:
Paul, I don't have an oil temp light on my 1994. Was this a later year enhancement?

Mark

It's a custom mod. I hijacked the seat belt bulb in the gauge cluster, ran it through an Arduino microcontroller and set it to the following conditions:

Light ON steady = Oil Temp < 60C
Light OFF = Oil Temp > 60C but < 120C
Light FLASH = Oil Temp > 120C

It serves two purposes. First, it tells me when the engine oil is still too cold for VTEC and hard operation. I baby the engine while the light is still on. Second, at the track, it tells me when the oil temp is approaching a dangerous level and a cool-down lap is required. Here it is flashing the "HOT" condition during initial testing. Works great and looks OEM.

View attachment 20221130_211152.mp4

TANTO · 29 January 2025

Well done thoughtful engine insurance policy - no "idiot" light on your ride. When young i used to glance at my oil pres/wtr temp gauge now and then. I think we all just don't even bother as the needle range is so small and the oil pressure reads so low it's almost stressing to watch!

R13 · 6 February 2025

Clever! (oil temp warning). Now having my own ways of seeing this, I was surprised at how much the oil temp would change independent of the coolant temp gauge. Also how much of a range of actual coolant temp kept the needle in the usual location. By the time the ECT gauge starts to move toward "hot" your oil is probably well out of pocket.

Honcho · 8 February 2025

R13 said:
Clever! (oil temp warning). Now having my own ways of seeing this, I was surprised at how much the oil temp would change independent of the coolant temp gauge. Also how much of a range of actual coolant temp kept the needle in the usual location. By the time the ECT gauge starts to move toward "hot" your oil is probably well out of pocket.

Thanks! I arrived at this solution after owning my GT-R. On that car, I noticed how much more quickly the coolant warmed up versus the oil. For the NSX, I don't really care what the exact number is, but only is it too cold, too hot, or just right?

Leaving my house in the winter, the coolant gauge is at temp by the exit of my neighborhood, but the oil light doesn't shut off until I'm well along the expressway.

Honcho · 8 February 2025

Winter Days

We had a January thaw after the polar vortex and most of the snow finally melted, so it was time to exercise the car. Last week, I took it to the local cars and coffee. I live in an affluent neighborhood, so always a lot of nice cars there. I'm surprised at how much of a crowd the NSX still draws. I think it's more the rarity factor than anything else these days. You just don't see them anymore. Always fun to eavesdrop on the commentary: "It has a Honda Accord engine", "The entire body is titanium", "It's a Formula One engine", "The wing makes 1,000 pounds of downforce", etc.

Today, I took the NSX for its winter exercise loop: 1st gear to the end of my street (circulates the trans fluid before shifting), 5,000 rpm upshifts on the 1st entrance ramp (lukewarm oil), steady 3,000 rpm cruise for 12 miles, turn around on the underpass, full VTEC on the 2nd entrance ramp, top speed VTEC (limiter in 4th gear), steady 3,000 rpm cruise home, idle for 5 min to let the PCV work. After the exercise, I always inspect the car. Looking for oil leaks, fuel, loose bolts, etc. It's an old car. As always, she looks great- just as clean as the original restoration.

Post-drive idling while the PCV burns off all the water and blowby gas.

Checking the oil condition- golden like honey. It was steaming hot out of the fill hole. It means the engine got good and hot and cleaned itself well. Cam cover metal looks very clean.

I'm very pleased with the Endless MX72+ pads. They've bedded in quite well. Hard to believe these are the original rotors from the donor car!

Plugging in the battery conditioner for another week of rest. I still put a towel even though I have PPF. Exotic car habits...

Cairo94507 · 8 February 2025

I read this entire thread over 2 days...amazing, simply amazing; I can't imagine a finer NSX or someone else with that skill set. I was just blown away by the attention to detail and expertise. Congratulations Honcho on a terrific restoration.

Mintifresh · 18 February 2025

I just read 61 pages of awesome.

Amazing. Truly amazing.

Honcho · 21 February 2025

Mintifresh said:
I just read 61 pages of awesome.

Amazing. Truly amazing.

Cairo94507 said:
I read this entire thread over 2 days...amazing, simply amazing; I can't imagine a finer NSX or someone else with that skill set. I was just blown away by the attention to detail and expertise. Congratulations Honcho on a terrific restoration.

Thank you!

Honcho · 21 February 2025

Emissions

Time for the Zero's biennial smog testing. Took advantage of a sunny day and headed over to the testing facility. After waiting in line for 45 minutes (at least the cats got hot that way), it went on the dyno for the IM240 rolling road test for OBD-I cars:

The car passed in 2023 with the terrible, barely adequate Prospeed tune, so I wasn't worried and then:

FAILED???

Look closer. It passed emissions with flying colors but it failed on the gas cap test?? How can that be? I'm using a new OEM Honda gas cap. In any event, a quick trip to Autozone and I bought a generic Duralast cap. Here they are compared. The best I can figure is that the rubber gasket on the Honda cap dried out and with the 20F cold weather was just too hard to make a good seal. Look at the difference in gaskets between the two caps.

The next day, with my cheap Autozone gas cap in place, I tested again. Easy pass. Look at those numbers. This engine has 141,000 miles on it with original cats. What an incredible piece of engineering!

Honcho · 1 March 2025

Cars and Coffee

Beautiful day today and the first of the month, so it was a good opportunity to exercise the NSX and go to the local cars and coffee. Big turnout today due to the weather. I was late after my exercise loop, so parked near the back. Funny how small the NSX looks next to the Vantage. I experienced a hot start issue when leaving- of course in front of the big crowd LOL. I had to give the car about half throttle when starting before it fired up. The NSX does this about every 100 starts and it's normal behavior, but I may look at the hot start tables in TunerPro to see if I can make it better.

Mintifresh · 1 March 2025

Such a beautiful color , we are having amazing weather here in Houston today however did not get a chance to take her out

Honcho · 1 March 2025

Mintifresh said:
Such a beautiful color , we are having amazing weather here in Houston today however did not get a chance to take her out

Thank you! I just went back out to fill up the car- windows down, gorgeous day. The exhaust sounds so good on a day like this!

Honcho · 1 March 2025

Hot Start Tuning

Starting to investigate the hot start issue with the car. If the car is full hot and is started right after shutting down, it always fires. The problem seems to be when the hot engine is parked for 30-60 minutes and then started again. In that scenario, the engine is still warm, but not fully hot anymore. Prior to the car show today, this event last occurred on the drive to NSXPO where I stopped at McDonalds in Amarillo for lunch. As I have mentioned, it happens every 50-100 starts and I feel more and more like there is a "sweet spot" in terms of ambient temperature, engine temperature and humidity where it just doesn't like it.

This NSX has RDX 410cc fuel injectors, so most of the fuel parameters have been modified to reduce the injector pulsewidths by 25% to as much as 45%. The car runs strong and perfectly, but I never touched the cranking fuel map...mostly because the car typically starts fine in all temperatures and altitudes. However, the hot start behavior suggests that there is too much fuel. To get the car to fire during a hot start failure, I must open the throttle butterfly about 50%. That means that I'm introducing a lot of air into the mixture, e.g., making it leaner. Below are the parameters that control cranking fuel in the OEM NSX ECU:

The NSX cranking fuel strategy depends on coolant temperature and barometric pressure. For baro, the ECU will reduce cranking fuel pulsewidths by 16.47% when the pressure is below 29.76In/hG. Here in Denver, we average about 24.5 In/hG, so there already is some enleanment going on during starts. I'm not sure if the ECU interpolates the reduction in a linear or logarithmic way, or if it's just "on/off". I continue to hope and pray that @MotorMouth93 will eventually release a "cleaned up" version of the ADX file that explains these parameters better!

As the engine gets colder, the ECU will add more fuel during the starting cycle. We can see in the top left chart that at a coolant temp of -2C (below freezing), the car will add 50 ms to the injectors (that's a lot!) Considering that I have larger fuel injectors, I was thinking about reducing each cell in this map by 20%. But, my NSX starts every time in very cold weather. Thus, I'm reluctant to adjust the cold starting fuel. Instead, this NSX occasionally fails to start when warm unless I feed it considerably more air. As such, I reduced the 18C, 78C and 205C cells to 26.000, 8.000 and 8.000 respectively. Just taking out about 10% of the fuel. Given that this issue only occasionally pops up, I'll monitor to see if it gets better with these changes.

alum04org · 2 March 2025

LOVE that new blue A-M Vantage beside your NSX !

bartman1980 · 2 March 2025

WOW, I have read this start to finish over my night shifts at work, what an amazing build.
Your attention to detail and documentation is unreal, you should have been an aircraft engineer. Your project and thread has given me some great inspiration for me to do my manual conversion. Awesome work and what an amazing NSX

docjohn · 2 March 2025

...and for all you prime newbies he is "just" a lowly corporate's legal Eagle...afaik all of his technical skill is self taught...

MotorMouth93 · 2 March 2025

Honcho said:
Hot Start Tuning

Starting to investigate the hot start issue with the car. If the car is full hot and is started right after shutting down, it always fires. The problem seems to be when the hot engine is parked for 30-60 minutes and then started again. In that scenario, the engine is still warm, but not fully hot anymore. Prior to the car show today, this event last occurred on the drive to NSXPO where I stopped at McDonalds in Amarillo for lunch. As I have mentioned, it happens every 50-100 starts and I feel more and more like there is a "sweet spot" in terms of ambient temperature, engine temperature and humidity where it just doesn't like it.

This NSX has RDX 410cc fuel injectors, so most of the fuel parameters have been modified to reduce the injector pulsewidths by 25% to as much as 45%. The car runs strong and perfectly, but I never touched the cranking fuel map...mostly because the car typically starts fine in all temperatures and altitudes. However, the hot start behavior suggests that there is too much fuel. To get the car to fire during a hot start failure, I must open the throttle butterfly about 50%. That means that I'm introducing a lot of air into the mixture, e.g., making it leaner. Below are the parameters that control cranking fuel in the OEM NSX ECU:

View attachment 198015

The NSX cranking fuel strategy depends on coolant temperature and barometric pressure. For baro, the ECU will reduce cranking fuel pulsewidths by 16.47% when the pressure is below 29.76In/hG. Here in Denver, we average about 24.5 In/hG, so there already is some enleanment going on during starts. I'm not sure if the ECU interpolates the reduction in a linear or logarithmic way, or if it's just "on/off". I continue to hope and pray that @MotorMouth93 will eventually release a "cleaned up" version of the ADX file that explains these parameters better!

As the engine gets colder, the ECU will add more fuel during the starting cycle. We can see in the top left chart that at a coolant temp of -2C (below freezing), the car will add 50 ms to the injectors (that's a lot!) Considering that I have larger fuel injectors, I was thinking about reducing each cell in this map by 20%. But, my NSX starts every time in very cold weather. Thus, I'm reluctant to adjust the cold starting fuel. Instead, this NSX occasionally fails to start when warm unless I feed it considerably more air. As such, I reduced the 18C, 78C and 205C cells to 26.000, 8.000 and 8.000 respectively. Just taking out about 10% of the fuel. Given that this issue only occasionally pops up, I'll monitor to see if it gets better with these changes.

The ECU uses linear interpolation on nearly everything, so that table isn't an on/off switch. So if you are at 24.5inhg, the actual multiplier used can be calculated with the following, where X is the table input (in this case barometric pressure), VL is the low output value (-16.47%), VH is the high output value (0%), BPL is the low breakpoint (19.47inhg), and BPH is the high breakpoint (29.76inhg).

Interpolated value = VL + (((X - BPL) / (BPH - BPL)) * (VH - VL))

-16.47 + (((24.5 - 19.47) / (29.76 - 19.47)) * (0 - (-16.47))) = -8.5 %

This adjustment though actually results in a net enrichment in cranking fuel though, as 29.76 to 24.5inhg is a roughly 18% drop in air pressure, which results in an 8.5% reduction in fuel based on that table. So for whatever reason Honda wanted it to crank slightly richer at low barometic pressure. I'm not sure if this is ideal or not...my car doesn't exactly warm start perfectly either LOL.

Also, your base cranking pulsewidth table doesn't seem like it has been adjusted from the factory defaults for 240cc injectors so you'll probably want to multiply all of those by 0.59 to account for this before you do anything else, you'll be cranking ultra rich from that alone.

Realistically I don't think there will ever be an updated XDF unless I am unemployed for a long period of time. I don't think I'll ever have COVID WFH and rarely leave the house for 2 years straight levels of free time in my life again until I retire in 20+ years.

Zoom · 3 March 2025

MotorMouth93 said:
my car doesn't exactly warm start perfectly either LOL

Seems like it's univeral on 90's Honda sports cars at least

Mine has the exact same issue as narrowed by Honcho and it's still has OEM ECU and fuel system. I recall my former cars (2 Preludes, 1 Accord) had similar issues. All had a new main relay which is the first thing you'd think of. It seems we now have the proper answer ! Puzzling as to why Honda decided to map their ECUs this way...

BBVNSX · 3 March 2025

Yep! Mine is OEM and does that too… it happens very rarely… but it doesn’t need to be parked for 30 min, some times just after I stop to put gas it will happen… but the car will always start, crank will take an extra second or two and start.

MotorMouth93 · 3 March 2025

I believe the cranking time with these cars is dependent on how long it takes for the #1 cylinder to reach TDC when you engage the starter. The ECU doesn't know what the crankshaft is doing until it receives a pulse from the TDC sensor, so until then it has no idea when to fire the injectors or coils.

Honcho · 3 March 2025

BBVNSX said:
Yep! Mine is OEM and does that too… it happens very rarely… but it doesn’t need to be parked for 30 min, some times just after I stop to put gas it will happen… but the car will always start, crank will take an extra second or two and start.

Yeah my 91 did it occasionally, as well as my Accord. I do think it's a Honda thing, but I want to see if I can make it better.

alum04org said:
LOVE that new blue A-M Vantage beside your NSX !

I really liked that color too- especially in the sunlight. I'm seriously considering a DB11 as my next daily (or DB12 if they ever offer the V12) and I would love that color!

bartman1980 said:
WOW, I have read this start to finish over my night shifts at work, what an amazing build.
Your attention to detail and documentation is unreal, you should have been an aircraft engineer. Your project and thread has given me some great inspiration for me to do my manual conversion. Awesome work and what an amazing NSX

Thank you!! The best part is I get to drive it now.

docjohn said:
...and for all you prime newbies he is "just" a lowly corporate's legal Eagle...afaik all of his technical skill is self taught...

I'm back in private practice now. If anyone's company needs a good employment lawyer, let me know!

MotorMouth93 said:
The ECU uses linear interpolation on nearly everything, so that table isn't an on/off switch. So if you are at 24.5inhg, the actual multiplier used can be calculated with the following, where X is the table input (in this case barometric pressure), VL is the low output value (-16.47%), VH is the high output value (0%), BPL is the low breakpoint (19.47inhg), and BPH is the high breakpoint (29.76inhg).

Interpolated value = VL + (((X - BPL) / (BPH - BPL)) * (VH - VL))

-16.47 + (((24.5 - 19.47) / (29.76 - 19.47)) * (0 - (-16.47))) = -8.5 %

This adjustment though actually results in a net enrichment in cranking fuel though, as 29.76 to 24.5inhg is a roughly 18% drop in air pressure, which results in an 8.5% reduction in fuel based on that table. So for whatever reason Honda wanted it to crank slightly richer at low barometic pressure. I'm not sure if this is ideal or not...my car doesn't exactly warm start perfectly either LOL.

Also, your base cranking pulsewidth table doesn't seem like it has been adjusted from the factory defaults for 240cc injectors so you'll probably want to multiply all of those by 0.59 to account for this before you do anything else, you'll be cranking ultra rich from that alone.

Realistically I don't think there will ever be an updated XDF unless I am unemployed for a long period of time. I don't think I'll ever have COVID WFH and rarely leave the house for 2 years straight levels of free time in my life again until I retire in 20+ years.

Thank you for this, John. I didn't adjust the table since the car nearly always starts perfectly, but yeah it's got to be super rich. I'm going to start with the 10% reduction on the top three temp cells and see if it helps.

Zoom said:
Seems like it's univeral on 90's Honda sports cars at least Mine has the exact same issue as narrowed by Honcho and it's still has OEM ECU and fuel system. I recall my former cars (2 Preludes, 1 Accord) had similar issues. All had a new main relay which is the first thing you'd think of. It seems we now have the proper answer ! Puzzling as to why Honda decided to map their ECUs this way...

It may be because they wanted to err on the rich side vs lean.

aeam1990 · 8 March 2025

Honcho said:
Weight Reduction - S Zero Conversion 05

More EPS

Completed the removal of the EPS power unit from the passenger footwell. Quite cramped, as usual, but fairly straightforward. Worst part by far was the wire harness plug on the unit. Absolutely brutal and took almost an hour to remove. That is even after I cut the harness (I am going to trim off the pigtail that goes to the EPs power unit) That plug felt like it was welded together.

More work on the EPS rack. Removing tie rod ends. Fairly simple tools needed, but you need a large flat head screwdriver or chisel to bend back the locking washer tabs.

Full Gen 1 EPS Auto trans kit ready for packaging and then on its way to Matt and Charles at NSX Rack Repair.

Hi, Is the electronic steering rack different in the auto trans cars versus the manual trans cars.

Honcho · 10 March 2025

aeam1990 said:
Hi, Is the electronic steering rack different in the auto trans cars versus the manual trans cars.

The rack itself is not different between transmissions, but it is different based on the year.

Honcho · 10 March 2025

ECU Tuning

Absolutely stunning weather today and dry roads, so I decided to take the opportunity to do some logging. It's been almost a year since any real tuning on the car simply because it runs so well already. Still, there are a few nits here and there that I'd like to dial in before burning a chip. First is the AC clutch EACV pulse. At times, the car will briefly rev to almost 1,500 rpm when the AC comes on. This means the EACV is opening way too much and raising the idle. It's really just supposed to "bump" the idle slightly so that the car doesn't stall. The EACV has two AC clutch settings, one for the "high" state of the triple pressure switch, and one for the "low." The table covers two functions: first, when the AC request signal kicks on, the EACV will open to the "Max Init" value and then decay to the "Min Dcyd" value, where it will remain open as long as the AC clutch is engaged. Second is when the AC clutch request turns off, the EACV will decay by the "Off" value every 10ms until it zeros out.

Since my base idle with the AC on is a steady 850 rpm, I didn't think there was anything wrong with the "Min Dcyd" values. Instead, I elected to reduce the "Max Init" value by 10% in each table to take some of the "punch" out of the initial opening. I wasn't sure what pressure state the AC switch is when the spike happens, so I adjusted both tables.

It was a good day to test, as the temperature was nearly 70F and it was bright sun. As such, the AC was cycling on and off for the entire drive. I'm happy to report that the AC engage event is a lot less dramatic now and the car happily bumps the idle slightly now. I'm going to call this solved, but the real test will be this summer.

MotorMouth93 · 10 March 2025

These are probably values that should be scaled with the injector change as well. I think these are additive values (multiplicative tables do not need to be scaled) but I haven't dug around in the relevant assembly code in a while.

Honcho's Long Road to Imola Type-S Zero

Similar threads