s5 6 port turbo.. who's done it? questions inside
#79
logical progression
iTrader: (5)
Join Date: Sep 2005
Location: sussex county, delaware
Posts: 406
Likes: 0
Received 0 Likes
on
0 Posts
I built a s4 na turbo a few years ago and then had to abandon it when I just started to add boost. 6port motor with T2 front cover and manifolds. Hks exhaust manifold, and a T04s turbo. Vmount cooling. I got it running and drove it for a few months working out the bugs with no boost. When I started to add boost, I started over heating and decided I need a cooling system rework. I tried using stock radiator. I did it on a budget so I tried to save as much as I could. Let it sit for a year because of finance issues, Lost my Job.... And now back on it. Port matched T2 manifold to housings and removed the aux sleeves. Throttle body mods and 3 inch exhaust turbo back. I used the megasquirt system, and it took me a while to get it figured out, but now it is working well.
In the midst of a custom wire harness job to get rid of the old factory harness issues. And eliminate a million unnecessary wires to things that arent in the car anymore. I ran a turbo setup through inspection with a cat and airpump and passed inspection with flying colors and no emissions crap. I have like 10 wires in my engine bay.... roughly. Of course we dont do visual inspection in delaware, just emmisions.
edit...
I didnt use a baffle in the pan... No issues yet...
In the midst of a custom wire harness job to get rid of the old factory harness issues. And eliminate a million unnecessary wires to things that arent in the car anymore. I ran a turbo setup through inspection with a cat and airpump and passed inspection with flying colors and no emissions crap. I have like 10 wires in my engine bay.... roughly. Of course we dont do visual inspection in delaware, just emmisions.
edit...
I didnt use a baffle in the pan... No issues yet...
#82
Update: Talked to Brian at BNR and he is assembling my turbo and will ship at the beginning of next week, so i decided to start the build.
5-28-10
Took for one last n/a drive. Raised front of the car and removed under belly tray, inner wheel liners, radiator cooling panel, front bumper and supports, battery and tray, filter, afm and snorkel. Thats all i was able to do tonight since i had to replace a a/c compressor on the neighbors car and make a flexible test pipe for a 1990 ka-t 240. I will post picks of stuff when i can get some down time. Will continue to work through this holiday weekend and hopefully get the car to the point where i need the turbo to continue the rest of the build.
5-28-10
Took for one last n/a drive. Raised front of the car and removed under belly tray, inner wheel liners, radiator cooling panel, front bumper and supports, battery and tray, filter, afm and snorkel. Thats all i was able to do tonight since i had to replace a a/c compressor on the neighbors car and make a flexible test pipe for a 1990 ka-t 240. I will post picks of stuff when i can get some down time. Will continue to work through this holiday weekend and hopefully get the car to the point where i need the turbo to continue the rest of the build.
#83
High Comp Booster
iTrader: (4)
Join Date: Apr 2009
Location: Oakville, Ontario
Posts: 1,907
Likes: 0
Received 0 Likes
on
0 Posts
Good to hear man
My 6 port motor Finally let go, I believe the motor is just really really Carboned up, I ran it really rich and Cheap oil i knew it would happen
I just got myself a Low KM jdm T2 motor now Tore it down to replace all the soft seals
Good luck with the Conversion its alot of fun and can be done in a day
S5's dont use a baffle plate because the oil drain in the front cover is in a different position and doesn't drain on to the oil pump drive chain
Just reuse your stock cover and drill out the hole for the oil to go through Works really well just like stock
My 6 port motor Finally let go, I believe the motor is just really really Carboned up, I ran it really rich and Cheap oil i knew it would happen
I just got myself a Low KM jdm T2 motor now Tore it down to replace all the soft seals
Good luck with the Conversion its alot of fun and can be done in a day
S5's dont use a baffle plate because the oil drain in the front cover is in a different position and doesn't drain on to the oil pump drive chain
Just reuse your stock cover and drill out the hole for the oil to go through Works really well just like stock
#90
well turns out that the oil pump that i had purchased that was supposed to be for a t2 was the same as the one on my car. I know my engine had been opened up before so idk if it already has a turbo oil pump of not, but i decided to put the one that was originally on the car back in. hopefully i don't run into problems.
update:
5/29/10-5/31/10
Radiator and a/c condensor removed, Front end cut and welded back up to make room for the corksport fmic. N/a intake and exhaust manifolds removed, p/s set off to the side, rats nest removed, front cover and water pump housing removed. Turbo oil baffle installed, t2 front cover installed. N/a water pump housing reinstalled after i drilled and tapped it for innovate coolant temp sensor. T2 exhaust mani and lim installed. OIl filler neck removed, t2 installed. Removed oil filter pedistal and replaced with banzia racing modified unit.
update:
5/29/10-5/31/10
Radiator and a/c condensor removed, Front end cut and welded back up to make room for the corksport fmic. N/a intake and exhaust manifolds removed, p/s set off to the side, rats nest removed, front cover and water pump housing removed. Turbo oil baffle installed, t2 front cover installed. N/a water pump housing reinstalled after i drilled and tapped it for innovate coolant temp sensor. T2 exhaust mani and lim installed. OIl filler neck removed, t2 installed. Removed oil filter pedistal and replaced with banzia racing modified unit.
#94
rotorhead
iTrader: (3)
Hey I just want to let you guys know what happens when you run a high compression rotary without appropriate spark advance maps. I have been doing some remote tuning on a couple Rx-8's, which as many of you know are 6 port engines with 10:1 compression.
I've been helping a couple guys with Greddy 18G kits. The factory Greddy emanage base map is a piece of ****. The timing maps make no sense and basically do nothing as far as retarding timing to compensate for the boost. So here is what happened to one guy who used the default Greddy map:
At 5500rpm the engine detonated pretty bad. The factory timing maps called for about 26 degrees leading with 15 split, which is just too much for 10:1 compression and a boosted setup. Luckily the Rx-8 PCM is smart enough to severely pull timing and the engine survived. In this case the engine was running about 7psi boost (spring pressure) on a Greddy TD06-18G with I believe the 10cm^2 hotside.
Because the emanage is a piggyback, I had to build correction maps relative to the factory commanded spark advance curve. I had to take the stock timing map (where it maxes out, b/c Rx-8 isn't designed for boost) and then use a spreadsheet to figure out the correction I wanted. So check this out:
On the very top line is the Rx-8 commanded spark table when the engine gets into boost. The factory timing maps are not scaled for the additional loads of a turbocharged engine. The middle part of the spreadsheet is the E-Manage Ultimate correction map I built based on rpm and boost. The bottom part shows what the new timing values should be (might vary a bit) at these rpm and boost levels. The EMU maxes out at 20 degrees spark retard. Note that I am using 15 degrees split just like the factory Rx-8 timing maps.
If you look across the spark correction map, you'll see that I progressively ramped up the rate of spark retard over the rpm range. The result is a much "flatter" timing curve. The reason for this is because the engine doesn't need as much timing correction in the low rpm range, and pulling too much timing will hurt spool. I also ramped up the rate of spark retard relative to boost. So at low boost levels, timing is pulled slowly, and as boost increases the rate of spark retard increases. This is also to preserve spool.
This Emanage timing correction map fixed the knocking problem on the Rx-8 in question. On an FC with an Rtek 2.1 or a standalone the timing maps are way less bullshit... that E-Manage Ultimate correction map was a lot of work. I had to locate the factory load-based timing map, create a boost-based ignition correction map, and then blend it together to figure out a resulting boost-based timing map. Controlling timing with a piggyback is a pain, avoid it if you can. And to make matters worse, Greddy reversed the leading and trailing wiring and their respective maps. So the leading correction map actually controlled the trailing and the trailing controlled the leading. If someone doesn't pick up on that the results could be disastrous.
I've been helping a couple guys with Greddy 18G kits. The factory Greddy emanage base map is a piece of ****. The timing maps make no sense and basically do nothing as far as retarding timing to compensate for the boost. So here is what happened to one guy who used the default Greddy map:
At 5500rpm the engine detonated pretty bad. The factory timing maps called for about 26 degrees leading with 15 split, which is just too much for 10:1 compression and a boosted setup. Luckily the Rx-8 PCM is smart enough to severely pull timing and the engine survived. In this case the engine was running about 7psi boost (spring pressure) on a Greddy TD06-18G with I believe the 10cm^2 hotside.
Because the emanage is a piggyback, I had to build correction maps relative to the factory commanded spark advance curve. I had to take the stock timing map (where it maxes out, b/c Rx-8 isn't designed for boost) and then use a spreadsheet to figure out the correction I wanted. So check this out:
On the very top line is the Rx-8 commanded spark table when the engine gets into boost. The factory timing maps are not scaled for the additional loads of a turbocharged engine. The middle part of the spreadsheet is the E-Manage Ultimate correction map I built based on rpm and boost. The bottom part shows what the new timing values should be (might vary a bit) at these rpm and boost levels. The EMU maxes out at 20 degrees spark retard. Note that I am using 15 degrees split just like the factory Rx-8 timing maps.
If you look across the spark correction map, you'll see that I progressively ramped up the rate of spark retard over the rpm range. The result is a much "flatter" timing curve. The reason for this is because the engine doesn't need as much timing correction in the low rpm range, and pulling too much timing will hurt spool. I also ramped up the rate of spark retard relative to boost. So at low boost levels, timing is pulled slowly, and as boost increases the rate of spark retard increases. This is also to preserve spool.
This Emanage timing correction map fixed the knocking problem on the Rx-8 in question. On an FC with an Rtek 2.1 or a standalone the timing maps are way less bullshit... that E-Manage Ultimate correction map was a lot of work. I had to locate the factory load-based timing map, create a boost-based ignition correction map, and then blend it together to figure out a resulting boost-based timing map. Controlling timing with a piggyback is a pain, avoid it if you can. And to make matters worse, Greddy reversed the leading and trailing wiring and their respective maps. So the leading correction map actually controlled the trailing and the trailing controlled the leading. If someone doesn't pick up on that the results could be disastrous.
#96
rotorhead
iTrader: (3)
I have put together a new Rtek base timing map for 9.7:1 rotors as a more conservative starting point than the 9.4:1 map I have supplied elsewhere. Over 5.5psi I pulled an additional 2 degrees of timing in most areas compared to the 9.4:1 map. I also ramped up to 15 split at lower boost.
leading:
split:
There is noticeably more timing here than the Greddy EMU maps I posted for the Rx-8 (which has 10:1 compression). The Rx-8 Greddy 18G kit is really only designed for about 7psi so that Rx-8 map I posted pulls a lot of timing as boost climbs over 7psi.
leading:
split:
There is noticeably more timing here than the Greddy EMU maps I posted for the Rx-8 (which has 10:1 compression). The Rx-8 Greddy 18G kit is really only designed for about 7psi so that Rx-8 map I posted pulls a lot of timing as boost climbs over 7psi.
#100
update, car is all put back together besides hood and inner fenders. Started the car, timing is off and i got a pretty big oil leak, looks like the leak is from the turbo drain right at the drain, i'd spend more time with it but i gotta get up early and go to work. i'll post vids of it running soon. O and i think that my tps might be **** or the tb is hanging up and not letting the tps to read closed.