What you've been waiting for results inside!
10-17-2010, 03:17 PM
#101
Senior Member
Thread Starter
iTrader: (7)
Join Date: Dec 2006
Location: Chicagoland
Posts: 1,376
Total Cats: 4
but like others have said, power under the curve in YOUR power range wouldn't help him at the drag track at all. if he runs a 8500 rpm redline, spending time and $$$ making power from 4-6k would be stupid as he would never be under 6k down the 1/4 mile or launching.
Reply
0
0
10-17-2010, 07:51 PM
#105
Boost Pope
iTrader: (8)
Join Date: Sep 2005
Location: Chicago. (The less-murder part.)
Posts: 33,017
Total Cats: 6,587
I'm torn here.
I respect the **** out of what jtothawhat has done here. Being able to say the words "425 HP" and "1.6 Miata" in the same sentence is nearly unfathomable.
But as a daily driver? I honestly think I'd go for Sav's setup. Calculate the area under the curve- it's much larger.
You'd also need a twelve speed gearbox, if we're doing an apples-to-apples comparison.
Is it really necessary even to do this?
Comparing straight gasoline to straight ethanol, there's a fixed multiplier with which you can compare and convert any of the important properties between them, be it energy per unit volume, volumes of air per volumes of fuel required to achieve ideal lambda, etc. So once you know (thanks to the sensor) what the ratio of gasoline to ethanol is in the tank, you should then simply be able to apply that multiplier to the output of the fuel pusewidth calculation, as is done on most ECUs with compensations for battery voltage, coolant and air temperature, air density, etc. In the Megasquirt world, this is the Gammma variable.
What I am not 100% certain of is whether there exists a linear relationship between fuel composition and ideal spark advance as the ratio of gasoline to ethanol varies. So far as I know, the MS's FlexFuel code does make this assumption.
That was pretty much my reaction when I read that as well. If you have a 128x128 table in which there is significant inter-cell variation which is not able to be approximated by inter-cell interpolation on a 16x16 table, then something is very wrong.
I mean, let's say that you're running an 8,000 RPM redline and 30 PSI of boost. Do you really need to have individual cells every 62.5 RPM and 0.35 PSI?
I can't imagine having to deal with a table that large.
I respect the **** out of what jtothawhat has done here. Being able to say the words "425 HP" and "1.6 Miata" in the same sentence is nearly unfathomable.
But as a daily driver? I honestly think I'd go for Sav's setup. Calculate the area under the curve- it's much larger.
Comparing straight gasoline to straight ethanol, there's a fixed multiplier with which you can compare and convert any of the important properties between them, be it energy per unit volume, volumes of air per volumes of fuel required to achieve ideal lambda, etc. So once you know (thanks to the sensor) what the ratio of gasoline to ethanol is in the tank, you should then simply be able to apply that multiplier to the output of the fuel pusewidth calculation, as is done on most ECUs with compensations for battery voltage, coolant and air temperature, air density, etc. In the Megasquirt world, this is the Gammma variable.
What I am not 100% certain of is whether there exists a linear relationship between fuel composition and ideal spark advance as the ratio of gasoline to ethanol varies. So far as I know, the MS's FlexFuel code does make this assumption.
128x128 maps are asinine and excessive, but a great way for tuners to rip people off I guess.
I mean, let's say that you're running an 8,000 RPM redline and 30 PSI of boost. Do you really need to have individual cells every 62.5 RPM and 0.35 PSI?
I can't imagine having to deal with a table that large.
Reply
0
0
10-17-2010, 09:14 PM
#110
Former Vendor
iTrader: (31)
Join Date: Nov 2006
Location: Sunnyvale, CA
Posts: 15,442
Total Cats: 2,099
Is it really necessary even to do this?
Comparing straight gasoline to straight ethanol, there's a fixed multiplier with which you can compare and convert any of the important properties between them, be it energy per unit volume, volumes of air per volumes of fuel required to achieve ideal lambda, etc. So once you know (thanks to the sensor) what the ratio of gasoline to ethanol is in the tank, you should then simply be able to apply that multiplier to the output of the fuel pusewidth calculation, as is done on most ECUs with compensations for battery voltage, coolant and air temperature, air density, etc. In the Megasquirt world, this is the Gammma variable.
Comparing straight gasoline to straight ethanol, there's a fixed multiplier with which you can compare and convert any of the important properties between them, be it energy per unit volume, volumes of air per volumes of fuel required to achieve ideal lambda, etc. So once you know (thanks to the sensor) what the ratio of gasoline to ethanol is in the tank, you should then simply be able to apply that multiplier to the output of the fuel pusewidth calculation, as is done on most ECUs with compensations for battery voltage, coolant and air temperature, air density, etc. In the Megasquirt world, this is the Gammma variable.
The other question is whether you would need more fuel at idle vs. WOT to compensate for stuff like wall wetness, latent heat of vaporization under boost, etc. A quick glance at my 100 octane maps vs. E85 maps shows that I'm either losing fuel pressure at high RPM and high boost (walbro 255HP, unlikely but possible I suppose), or the motor wants a lot more fuel up top after the switch.
A quick google search turned up folks that said the setup is as simple as doing a gas tune and an E85 tune and clicking a button to let the ECU interpolate between the maps based on ethanol content in the tank. This would lead me to believe that a shop charging $3000 for setting it up is less than scrupulous, but I'm not a proEFI dealer so I have no idea how difficult the setup actually is.
As convenient as flex fuel is, I cannot ever imagine paying $3000 for the pleasure. If you must, have two separate AEM tunes and a test port/drain valve on the fuel feed line so you can drain the tank when necessary.
Reply
0
0
10-17-2010, 11:53 PM
#111
Senior Member
Thread Starter
iTrader: (7)
Join Date: Dec 2006
Location: Chicagoland
Posts: 1,376
Total Cats: 4
Sav I get what you're saying, I could do something like that--but it isn't very street friendly. I may just run E85 on this ECU and carry an extra 5 gallons in the trunk just in case, and just spend that extra money on other things such as wheels and body work. Either way, next on the list is a 949 twin disk, Quaife Trans, ATI Damp, 6UL wheels, Carbon Mirrors and a OEM lip...maybe a respray. I really need some sponsorship lol
Reply
0
0
10-18-2010, 12:06 AM
#112
Dude how much/far do you drive to be that scared of running out of e85?
I don't understand it when people say that.
If the nearest station is 20 miles away and its the only one then ok, I get it, but if not, then what is the problem?
I may be spoiled having 3 within 5 miles of my house and 2 within 5 miles of my work, but even if that wasn't the case its not that bad.
Unless you take long *** drives for hours and go 100 miles away I don't see it being a problem big enough to warrant a 6k ecu/tune.
You fill up, drive car, get to quarter tank start thinking of filling up (if its that hard for you to find e85). Quarter tank gives you at least a good 60 miles or more of distance to find e85 assuming you don't pound on it all the way there.
again, I might be spoiled having it easily available. I dunno
I don't understand it when people say that.
If the nearest station is 20 miles away and its the only one then ok, I get it, but if not, then what is the problem?
I may be spoiled having 3 within 5 miles of my house and 2 within 5 miles of my work, but even if that wasn't the case its not that bad.
Unless you take long *** drives for hours and go 100 miles away I don't see it being a problem big enough to warrant a 6k ecu/tune.
You fill up, drive car, get to quarter tank start thinking of filling up (if its that hard for you to find e85). Quarter tank gives you at least a good 60 miles or more of distance to find e85 assuming you don't pound on it all the way there.
again, I might be spoiled having it easily available. I dunno
Reply
0
0
10-18-2010, 12:19 AM
#113
Senior Member
Thread Starter
iTrader: (7)
Join Date: Dec 2006
Location: Chicagoland
Posts: 1,376
Total Cats: 4
Well actually I have a few E85 stations within 10 mins from my house, the closest is 2 miles away. lol But then you have to worry when they switch to winter blend E70, and it's never consistent at all etc. Just a lot of variables it seems like to me, but maybe I am just over thinking the whole thing.
Reply
0
0
10-18-2010, 02:13 AM
#118
it seems like the smarter way to go (and it seems like everyone is starting to convince you as well) to just keep the aem and spend the money on other important parts to make the 600 hp 1.6 a reality. The bottom line is that you in no way need that heavy duty of an ecu, and for long trips your options would be, extra gas can in trunk, load a pump gas map, or plan to have e85 stations. That just sounds like the smart way to go when you have your current ecu almost dialed in, and you're a few parts away from the most badass 1.6 miata in probably the world.
Reply
0
0
10-18-2010, 02:38 AM
#119
Former Vendor
iTrader: (31)
Join Date: Nov 2006
Location: Sunnyvale, CA
Posts: 15,442
Total Cats: 2,099
OK. No pissing match, just straight comparisons to help you improve your setup.
Here's a comparison from your setup to Eliminator's setup. IIRC, this chart was done on an RM 2 liter, 50trim T04E, .63 stg3 hotside. A little more displacement, but not nearly enough to justify the difference. He made 508whp at 23psi, ~576 at 30psi, and was still able to obtain excellent under-the-curve torque. I put my torque curve in as a comparison - the difference between green and blue is the difference I expect to see when you go from a ~400whp capable turbo to a ~600whp capable turbo. The red line is what I would expect to see on an 800whp capable turbo, not a PTE5857.
This is a great thread, full of same-day same dyno comparisons of a bunch of different turbos. 30R, 35R, HTA-wheeled 30 and 35Rs, and the PTE5857 is in there as well. Looks like it spools about as fast as a 30R, but delivers much better flow up top.
Page 4 has the PTE5857 chart. http://motorgeek.com/viewtopic.php?f=14&t=34852
What I'm trying to say is that your setup could be a lot better - you should be able to deliver the same power you have at least a thousand RPM earlier without changing the top-end you have now. Something in your setup isn't cutting it.
Here's a comparison from your setup to Eliminator's setup. IIRC, this chart was done on an RM 2 liter, 50trim T04E, .63 stg3 hotside. A little more displacement, but not nearly enough to justify the difference. He made 508whp at 23psi, ~576 at 30psi, and was still able to obtain excellent under-the-curve torque. I put my torque curve in as a comparison - the difference between green and blue is the difference I expect to see when you go from a ~400whp capable turbo to a ~600whp capable turbo. The red line is what I would expect to see on an 800whp capable turbo, not a PTE5857.
This is a great thread, full of same-day same dyno comparisons of a bunch of different turbos. 30R, 35R, HTA-wheeled 30 and 35Rs, and the PTE5857 is in there as well. Looks like it spools about as fast as a 30R, but delivers much better flow up top.
Page 4 has the PTE5857 chart. http://motorgeek.com/viewtopic.php?f=14&t=34852
What I'm trying to say is that your setup could be a lot better - you should be able to deliver the same power you have at least a thousand RPM earlier without changing the top-end you have now. Something in your setup isn't cutting it.
Reply
0
0
10-18-2010, 07:54 AM
#120
When I see the dyno graphs, it does look to me like jtothawhat's timing and or fuel settings need work on the lower end.
jtothawhat..Shim under lifters are very light weight lifters with shims that go between a post and the valvestem underneath the lifter. On a 99 solid lifter, you can see how the shim sits on top of the lifter. The cam comes in contact with the shim first. 1.6 hydralic lifters would look like a shim under lifter if you just pull the valve cover and look in. You would have to pull the cam and lifter and flip the lifter upside down. It would be much lighter too.
Also, I am suprised that you do not have a damper yet. Do you have a good oil pump? What port work have you done to the head? Be careful and make sure you have all the safegaurds before you crank up the RPMs.
jtothawhat..Shim under lifters are very light weight lifters with shims that go between a post and the valvestem underneath the lifter. On a 99 solid lifter, you can see how the shim sits on top of the lifter. The cam comes in contact with the shim first. 1.6 hydralic lifters would look like a shim under lifter if you just pull the valve cover and look in. You would have to pull the cam and lifter and flip the lifter upside down. It would be much lighter too.
Also, I am suprised that you do not have a damper yet. Do you have a good oil pump? What port work have you done to the head? Be careful and make sure you have all the safegaurds before you crank up the RPMs.
Reply
0
0
