Stealthiskey build thread
03-01-2009, 04:01 PM
#11
0.5 BAR
Thread Starter
Join Date: Jun 2008
Posts: 100
Re: Stealthiskey build thread
Setting Goals
Don't want to spend much time on this section, but I set my goal at 200 hp for a few reasons.
Daily driver, don't often have downtime to build the engine up, and the stock rods pistons are generally considered to safely hold between 200 and 250 hp. Plus oem honda parts are reliable. High hp builds often break.
Already have T25 turbos, which are capable of 200 hp.
Stock map sensor reads to 11 psi. Easy to change, but 10 psi on the stock block should put me close to 200 hp and the t25 can do that.
Don't want to spend much time on this section, but I set my goal at 200 hp for a few reasons.
Daily driver, don't often have downtime to build the engine up, and the stock rods pistons are generally considered to safely hold between 200 and 250 hp. Plus oem honda parts are reliable. High hp builds often break.
Already have T25 turbos, which are capable of 200 hp.
Stock map sensor reads to 11 psi. Easy to change, but 10 psi on the stock block should put me close to 200 hp and the t25 can do that.
03-01-2009, 04:22 PM
#12
0.5 BAR
Thread Starter
Join Date: Jun 2008
Posts: 100
Re: Stealthiskey build thread
ECU Chipping
Woot had the P28, just needed to chip it up. Ordered the kit from Moates along with Ostrich, O-meter, 2-timer, and Hulog. Used some brake cleaner to dissolve the conformal coat on the printed circuit board, a 15w grounded soldering iron and a 25 watt desoldering iron. radio shack special. Also used an ESD vest and wrist strap.
Had some trouble getting to read from the chip at first. Turns out the trace that connects J1 to the MCU is on the top side of the board and is not connected on that through-hole to the bottom of J1. Had to flow in some more solder to make connection. Put a stock p28 bin on there with Crome and drove it for a bit.
Woot had the P28, just needed to chip it up. Ordered the kit from Moates along with Ostrich, O-meter, 2-timer, and Hulog. Used some brake cleaner to dissolve the conformal coat on the printed circuit board, a 15w grounded soldering iron and a 25 watt desoldering iron. radio shack special. Also used an ESD vest and wrist strap.
Had some trouble getting to read from the chip at first. Turns out the trace that connects J1 to the MCU is on the top side of the board and is not connected on that through-hole to the bottom of J1. Had to flow in some more solder to make connection. Put a stock p28 bin on there with Crome and drove it for a bit.
03-01-2009, 04:48 PM
#13
0.5 BAR
Thread Starter
Join Date: Jun 2008
Posts: 100
Re: Stealthiskey build thread
LC-1 and Oil Pan
The next step was to get a wide band in there to get an idea for what the stock air fuel ratios should look like. Scored a second hand but new in box LC-1 and O2 sensor for $125. Turned out to have a defective processor, but Innovate backed up their product and sent me a new one and RMA, no questions asked. I used it to replace the stock narrow band and wired the simulated NB output into the ECU harness. Don't have any interesting pics.
I was also doing an oil change that weekend and figured it would be a good time to tap the oil pan for the return line. Bought a replacement gasket from Pep boys and some pipe fittings from the hardware store. Wasn't sure how everything was going to line up, but picked the most common spot I'd seen in other people's setups.
Not the best day for the welder, but it (eventually) didn't leak so meh.
Cleaned out the debris.
Repainted the outside, not inside.
Spent some time staring up into the crankcase. Saw that junk on the screen of the oil pickup and sucked it out with the shopvac and a small piece of tubing.
The next step was to get a wide band in there to get an idea for what the stock air fuel ratios should look like. Scored a second hand but new in box LC-1 and O2 sensor for $125. Turned out to have a defective processor, but Innovate backed up their product and sent me a new one and RMA, no questions asked. I used it to replace the stock narrow band and wired the simulated NB output into the ECU harness. Don't have any interesting pics.
I was also doing an oil change that weekend and figured it would be a good time to tap the oil pan for the return line. Bought a replacement gasket from Pep boys and some pipe fittings from the hardware store. Wasn't sure how everything was going to line up, but picked the most common spot I'd seen in other people's setups.
Not the best day for the welder, but it (eventually) didn't leak so meh.
Cleaned out the debris.
Repainted the outside, not inside.
Spent some time staring up into the crankcase. Saw that junk on the screen of the oil pickup and sucked it out with the shopvac and a small piece of tubing.
03-02-2009, 12:23 AM
#15
0.5 BAR
Thread Starter
Join Date: Jun 2008
Posts: 100
Re: Stealthiskey build thread
Injectors!
Scored some saturated RC 440's off ebay for around $100. Now that I could change up the tuning and monitor the air fuel ratio it was time to install the larger injectors and get serious. Had some buddies to help out.
Old injectors OUT!
It's not a hobby if you aren't drinking.
Some nasty junk on those old injectors. This is due to the PCV system dumping oily gunk down the intake manifold, but is perfectly normal.
440's in and they look so good! Had to get creative with the O-rings to fit on the top. Apparently RC Engineering also makes "Honda style top" injectors. These were not them, but they fit well enough to not leak. I also jammed them into the stock seals on the bottoms, which are now kind of fucked up, but not causing problems. If/when I could do it over I would get NEW seals and dremel them out so they actually fit.
At this point I also had some other parts stocked up. The gauges and a DSM BOV. Hooked in the boost/vac gauge so I could do some tuning.
Wasn't too familiar with Crome and wanted to keep things simple, so I copied the Crome fuel values into Microsoft Excel. From there I multiplied them all by just bit more than 240/440 to get a starting point. Copied those values back into Crome and Put them to the ECU. Fired up the engine and got started.
First things first, I started tuning the idle with the car in the driveway. By watching rpms and the boost/vac gauge I could tell what cells the ECU was reading from. By watching the AFR I could tell whether to add or subtract fuel. Given that the car was stationary the engine load or manifold absolute pressure (MAP or boost/vac) stays at -20 inHg regardless of throttle position. What does change is rpm. So I was able to get good AFRs for a range of RPMs around -20 inHg.
To tune up other MAP settings I had to load the engine by driving the car. It was important to keep the engine in low vacuum and make changes slowly. Turned out the values were pretty close to what they should be, but I made a few tweaks. Brought the car back to the garage and sat down with the laptop for a bit.
Larger injectors have higher flow rates (440 cc/min) but also heavier internals. The added mass means the injectors take longer to open. For this reason a direct multiplier of 240/440 is not entirely accurate, a constant or offset must also be added. I found the offset in excel by plotting the stock values in one column and the corresponding tuned values in the next column. By treating the first column as x-coordinates and the second as y-coordinates a linear correlation with a slope and intercept could be found. In my case the correlation that fit best was New_Fuel_Val = Old_Fuel_Val * 0.60 + 18. I applied this formula to all the stock values to calculate what values to use with 440 cc injectors. I copied the calculated values into Crome for both low cam and high cam tables and put them to the ECU. This proved to be a very accurate method, and AFRs were dead on under WOT and in VTEC and required no further tweaking.
Scored some saturated RC 440's off ebay for around $100. Now that I could change up the tuning and monitor the air fuel ratio it was time to install the larger injectors and get serious. Had some buddies to help out.
Old injectors OUT!
It's not a hobby if you aren't drinking.
Some nasty junk on those old injectors. This is due to the PCV system dumping oily gunk down the intake manifold, but is perfectly normal.
440's in and they look so good! Had to get creative with the O-rings to fit on the top. Apparently RC Engineering also makes "Honda style top" injectors. These were not them, but they fit well enough to not leak. I also jammed them into the stock seals on the bottoms, which are now kind of fucked up, but not causing problems. If/when I could do it over I would get NEW seals and dremel them out so they actually fit.
At this point I also had some other parts stocked up. The gauges and a DSM BOV. Hooked in the boost/vac gauge so I could do some tuning.
Wasn't too familiar with Crome and wanted to keep things simple, so I copied the Crome fuel values into Microsoft Excel. From there I multiplied them all by just bit more than 240/440 to get a starting point. Copied those values back into Crome and Put them to the ECU. Fired up the engine and got started.
First things first, I started tuning the idle with the car in the driveway. By watching rpms and the boost/vac gauge I could tell what cells the ECU was reading from. By watching the AFR I could tell whether to add or subtract fuel. Given that the car was stationary the engine load or manifold absolute pressure (MAP or boost/vac) stays at -20 inHg regardless of throttle position. What does change is rpm. So I was able to get good AFRs for a range of RPMs around -20 inHg.
To tune up other MAP settings I had to load the engine by driving the car. It was important to keep the engine in low vacuum and make changes slowly. Turned out the values were pretty close to what they should be, but I made a few tweaks. Brought the car back to the garage and sat down with the laptop for a bit.
Larger injectors have higher flow rates (440 cc/min) but also heavier internals. The added mass means the injectors take longer to open. For this reason a direct multiplier of 240/440 is not entirely accurate, a constant or offset must also be added. I found the offset in excel by plotting the stock values in one column and the corresponding tuned values in the next column. By treating the first column as x-coordinates and the second as y-coordinates a linear correlation with a slope and intercept could be found. In my case the correlation that fit best was New_Fuel_Val = Old_Fuel_Val * 0.60 + 18. I applied this formula to all the stock values to calculate what values to use with 440 cc injectors. I copied the calculated values into Crome for both low cam and high cam tables and put them to the ECU. This proved to be a very accurate method, and AFRs were dead on under WOT and in VTEC and required no further tweaking.
03-02-2009, 03:09 AM
#16
1.0 BAR
Join Date: Sep 2006
Location: Jacksonville, FL
Posts: 236
Re: Stealthiskey build thread
Thats a really interesting method for tuning. Im sure theres still some slight modifications you'll have to make but that sounds like a really good way to make a basemap. Keep up the good work.
03-05-2009, 09:52 PM
#20
1.0 BAR
Join Date: May 2006
Posts: 381
Where did you get the .6 and 18 from? I guess you exstrapolated that from the side by side comparisons of old injects vs new injectors for a given constant (afr)? Tuning is the unknown frontier that I have not explored although I am eager to learn.
Thread
Thread Starter
Forum
Replies
Last Post
Gold DA9
General Discussion
1
08-15-2008 11:58 AM
catch.can
General Discussion
87
05-22-2006 11:52 AM
