When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
3/Noticed that all of a sudden I was getting some knock at 12psi/4500rpm. Tried upping the W/M, adjusting timing, changing the stage 2 w/m timing and was ready to give up and just turn down the boost. Then I finally made the connection .... about four months ago I changed from Gull98 (E10) to BP 98 Ultimate. Reason for the change was that Gull stopped making 98E10 and now only sell 95! It would seem the Gull product was a much more knock resistant fuel than the BP one with no Ethanol.
Good part of that learning is that I didn't blow the engine finding this out - luck was on my side............. for once!
Once I realised that, I reset the w/m such that the 2nd stage came in much earlier and this seems to have solved it. Really lowered my confidence in the W/M setup though. I never had this when I ran E22 (or when I ran Gull98 + W/M)
.
Found another issue that was exacerbating this. The way I had the W/M controller setup was to increase flow as boost increased. This worked well most of the time BUT, at higher rpm when mashing the throttle .... the spoolup of the turbo is too quick for the controller/pump. Bringing the second stage in earlier was like a sledgehammer to crack a walnut approach. I've now found a way to accelerate the pump faster via the settings - working well so far. It means it's getting near full flow at much lower boost but at least I'm not drowning the engine with the 2nd stage now.
The problem with pressure regulated flow is that at low flow, you have terrible atomization, thus knock suppression is much lower than if the same volume would be very fine. And also, flow changes roughly with square rot of pressure difference.. ie, even if you say you want to cover only 4 to 8k rpm, and only the max load, that is still a factor of 2 in required optimal flow... that requires Pmax/Pmin=2^2 ie, the max pressure should be 4 times the min pressure... The red FI of the s1 are water/ethanol tolerant (except the mesh filter on the inlet which can be changed) ... I plan to utilize them with a microcontroller and try to aim at ~% of fuel injection quantiy..
The problem with pressure regulated flow is that at low flow, you have terrible atomization, thus knock suppression is much lower than if the same volume would be very fine. And also, flow changes roughly with square rot of pressure difference.. ie, even if you say you want to cover only 4 to 8k rpm, and only the max load, that is still a factor of 2 in required optimal flow... that requires Pmax/Pmin=2^2 ie, the max pressure should be 4 times the min pressure... The red FI of the s1 are water/ethanol tolerant (except the mesh filter on the inlet which can be changed) ... I plan to utilize them with a microcontroller and try to aim at ~% of fuel injection quantiy..
I do realise that ... which is why I went for two stage setup. But don't forget that this turbo spools to 10psi by 3000rpm! That makes it even worse. So the controller does still help at low rpm.
I have the pump maxed out at 300psi ATM and as much as 650cc going into it when the 2nd stage kicks in yet the engine barely drops any power at all.
I somehow doubt that the pressure keeps constant when you open the 2nd stage. The pressure drops of as you increase nozzle flow capacity.
Yes I do remember you're on target boost at 3k rpm but for simplicity i wrote with "4".
It would be great if we could come by bmw water injection nozzles for reasonable $$..
Also would be worth cracking the ecu strategy used as they for sure have measured the impact on cylinder pressure and how much timing advance is needed to maintain performance.
Update:
..
3/Noticed that all of a sudden I was getting some knock at 12psi/4500rpm. Tried upping the W/M, adjusting timing, changing the stage 2 w/m timing and was ready to give up and just turn down the boost. Then I finally made the connection .... about four months ago I changed from Gull98 (E10) to BP 98 Ultimate. Reason for the change was that Gull stopped making 98E10 and now only sell 95! It would seem the Gull product was a much more knock resistant fuel than the BP one with no Ethanol.
Good part of that learning is that I didn't blow the engine finding this out - luck was on my side............. for once!
Once I realised that, I reset the w/m such that the 2nd stage came in much earlier and this seems to have solved it. Really lowered my confidence in the W/M setup though. I never had this when I ran E22 (or when I ran Gull98 + W/M)
...
Something similar on 100RON/E10 - 4500rpm towards 5000, 75kpa of boost, 9 degrees of timing, ECU pulls 2 degrees(from a base of 11 degrees). Then goes away pretty soon after.
Put another 3300 kms on 'Red' as a test of the manifold and system. A few pics of the trip:
Viaduct near Ruapehu on way down the North island Waiting for the Cook Straight ferry Into the south island. Lake on west coast This bars does great smoothies if you are ever in franz Joseph The famous Crown Range road where Mad Mike did an awesome short Drift film Layby overlooking Lake Hawera
more pics: Furthest South I got : Lake Manapouri My favourite pic from the trip. Taken at sunrise, overlooking Lake Wakatipu In front of The Remarkables. Same spot by Remarkables from other direction Lake just Noth of Queenstown Mt Cook in background Final leg of the trip at Kaikoura on way North
Car went great the whole trip ...no issues with turbo setup ..some takeaways from the trip:
*Used up most of my W/M solution early on and had to convert to water only for last part of trip. Need to put smaller nozzle back in first stage to address this.
*Front wheel scraping the gaurd every time I do a hard right turn.
*Vibration in exhaust at low rpms
*Sorted issue with car tracking poorly in a straight line early in the trip (thankfully) by increasing tyre pressure in front to 36psi.
*Gas consumption over 300 miles on a tank on the cruise back home from CHCH. No issue leaving SSV open for the trip.
*Nordlock washers held up on turbo flange - no leaks this time.
*Catch can had approx. 50ml of brownish liquid gunk in it after the trip.
*BOV sticks closed after a few days without use - have had this problem for years on this BOV.
*Had some epic drives on near empty roads down the west coast - having a turbo with low end response making the drive a delight.
Put another 3300 kms on 'Red' as a test of the manifold and system. A few pics of the trip:
Great pics 👌🏻 This is my plan as well when the kids can look after themselves. Just me and the wife on some scenic trips around Europe in the awesome 8.
Did an oil change today and checked a few things
*spoke too soon re the Nordlocks - they were all a bit loose . Bolt not backing off , but needing 1/4 turn to tighten.
*Scraping noise turned out to be the brake line bracket that I had modified to clear the turbo exit coupler.
Brett, what are you running for a radiator right now? After swapping around REW swap build concepts I have gone full circle and ended up in the stock location. I am curious how your temps are if on a stock size radiator?
Last I remember you only had one oil cooler on the driver side as well, but that may be old info.
I was thinking of putting together a simple list of how to do a bare necessity rew swap as I go.
I'm running an aftermarket Aluminium Rad. that is a bit thicker than stock. Nothing special. Plus, I turned a stock oil cooler into a water radiator on one side and run the stock single oil cooler that came with the car on the other. The oil does get hot during a track session but I'm not doing that much anymore so haven't bothered upgrading the oil cooler.
With Rad. in stock location you can't completely block airflow to it with an IC but it works just fine if you leave a 40mm gap(min.) under the IC .
I got one of those mazdaspeed bumper remakes and noticed it had MASSIVE inlets for the oil coolers so I thought it may be viable to run one larger oil cooler on just the driver side and simplify the system.
I swapped to a 3in, but stock or mishimoto is only 1.5in. I had figured if someone stuck with that size they could just use the passenger oil cooler side as an extra water if needed.
In regards to REW swaps, admittedly I'm not monitoring oil temps. But my factory cooling system with factory oil coolers (dual oil coolers and mazdaspeed bumper) with a FMIC (and likely a 40mm gap as Brett mentions) is not even getting close to overheating, I'm seeing between 175F and 185F water temps during street driving even on the hotter summer days (80-90F). So I can't comment on track days, but as far as bare minimum for REW swaps go, I think it's the factory Rx8 setup for a street application (climate dependent).
I got one of those mazdaspeed bumper remakes and noticed it had MASSIVE inlets for the oil coolers so I thought it may be viable to run one larger oil cooler on just the driver side and simplify the system.
I swapped to a 3in, but stock or mishimoto is only 1.5in. I had figured if someone stuck with that size they could just use the passenger oil cooler side as an extra water if needed.
mishimoto is a GPI made radiator.. no need to pay the brand tax (for their painting) - and it is almost double the thickness of the oem.. the oem indicates the max thickness not the thickness of the core.
*spoke too soon re the Nordlocks - they were all a bit loose . Bolt not backing off , but needing 1/4 turn to tighten.
Nordlocks are picky about being torqued to spec. I believe the factory manifold torque spec is 42-60 N-m (31-44 lb-ft). The dry torque spec for an M10 Nordlock is 56N-m (41lb-ft), so you would have to be right at the top of the torque spec to properly engage them.
Nordlocks are picky about being torqued to spec. I believe the factory manifold torque spec is 42-60 N-m (31-44 lb-ft). The dry torque spec for an M10 Nordlock is 56N-m (41lb-ft), so you would have to be right at the top of the torque spec to properly engage them.
I think the issue is more related to the 8-900C heat cycles they are subjected to. I'm hoping that retorquing them after several cycles will do the trick.
if the stud is to stiff, than it will always plastically (permanently) deform due to heat expansion of the manifold, and thus reduce the clamping load, no matter the initial torque copper nuts have a tendency to "flow"...
this seems to have solved it. Really lowered my confidence in the W/M setup though. I never had this when I ran E22 (or when I ran Gull98 + W/M)
but for simplicity i wrote with "4".
