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.
EGT or WB per cylinder is the proper way to go. I can technically differentiate which cylinder knocks first with the MS3x and that would be the lean one most likely. I'm not quite to the point of trusting the OEM sensor and the interpretation by the MS3 to sneak up on it like that. The other method is to keep adding boost and timing and whichever rod exits the block first was the lean one.......
Reality is you use matched injectors and build the intake in the ball park like I have and a mid 11's AFR will be plenty rich enough to compensate for any cylinder to cylinder imbalance.
looks good. what thickness did you use for the plenum? I've been told 1/8" for the longest time, but once becoming a part of the supra crowd, they suggest different requirements since boost is more than 15-20psi you commonly run on a miata. ive been told 1/4" is a good start for a sheetmetal plenum in that case where ballooning leads to weld fatigue/failure. but that's coming from a guy who builds them for 50psi service life which would be like pressure testing with 100-120psi without failure.
I feel like I need to mention: never pressure test with air... its scary if a failure occurs being that air compresses. fill your part with water first if all you have is air pressure to test with.
Yeah I struggled with this, initially I wanted to bend the plenum out of the plate and ordered .090" aluminum. Once I got it in hand I realized that making a smooth bend was going to be impossible so I resorted to splitting the leftover 2" tube into 1/4'ers to make the end caps. It's a compromise and I wish I had used .125" for the plate.
Runners are 2" .125" wall
Plenum floor by runners is .25" because I wanted a heavy base to support the TB flange.
TB flange is .375" because of the threads needed for TB
Plenum plate is .090"
I haven't pressure tested this yet, other than boosting to 14psi. I plan to install bridge pieces through the plenum if I see any movement at all. A 1/4" rod tieing the two halfes together would do wonders for strength and not interrupt airflow much at all. I'll do a leak check / pressure test soon, I normally just use the air compressor set at 30psi for this. I was a pressure testing engineer in the gas industry at one point in my career..... Anything under 100psi doesn't scare me. You want to pucker up, nitrogen test 12" stainless at 1600psi.......I wanted a web cam watching the guages for me while I sat in the hotel.
Yeah, that could work but be ugly as hell. Metal is weak in bending but strong in tension, by putting a tension rod through the center of the plenum it would in effect stop the two half's from pulling apart, now that I think about it a piece of 1/8" by 1" strap would be stronger and disrupt airflow less, heck it could be used to balance airflow if I knew of a lean cylinder.
I've been putting some miles on the new intake and interestingly enough I needed to add ~4 to 6% fuel from 6000 rpm and up, fueling stayed about the same below that in boost. I did need to pull some fuel out in the low end under 100kpa so as expected I traded low end grunt for top end power and more importantly it hasn't cracked yet.
I finally had a breakthrough with the boost control, I found that the seal on my fabricated dual port wastegate actuator was leaking so I made another one that sealed better. That actually made the boost swings worse since the actuator had more power to close. The delay between the ECM seeing the boost swing, the actuator reacting to it, and the turbo reacting was so delayed that the system couldn't keep up with the flow demands needed to maintain a boost pressure. So I yanked the seal and went back to a single port with a heavy spring and the stars aligned. With the spring acting on the wastegate flapper the flapper will open if the pre turbo pressure spikes and close if it decreases all without waiting for the control loop to send a signal. So in essence it's a mechanical closed loop. This let's the MS3 just dictate the boost pressure desired and the spring will take control of the normal little adjustments needed to maintain that pressure. I just don't think the MS3 control logarithm is designed for a dual port arrangement, it works very well with a single port setup though.
I also got to Auto-X this weekend with my old club SWVSCCA. This was our first event in over a year since we lost our last lot so it was good to get back at it with the usual suspects. I ended up 5th out of 76 participants and 3rd in class. Only a viper ACR and a well prepped Z06 beat me and only by a half second at that. I still had a passenger in the car for my fastest run because I missed the announcement that we were only getting 5 instead of 6.......so maybe I could have won the whole thing? Excuses, yeah I got em.
The dual-port actuators that I've seen have all still had a mechanical spring in them, were you running yours with no spring at all?.
AIUI, the point to the dual port is to allow you to supplement that spring with boost pressure and use a lighter spring that would get blown open if it was being used in a single port actuator, to give a large range of controllable boost. The MS3 algorithm works great with my dual-port EWG actuator.
I was using a light spring, I had the EBC regulating one side of the controller so I could have a rather wide range of control like you mentioned.
Maybe a dual port EBC would have worked better, maybe it was just my tuning skills but I put a ton of time into it and could never get it to be consistent. I would tune for one set of conditions and the ocilations would come back under another set of conditions.
I was reading on his build thread and he had an external turbosmart actuator and a dual (4) port MAC valve for an EBC. Maybe he had a dual port internal WG actuator as well on his original 2560 build.
I was using a light spring, I had the EBC regulating one side of the controller so I could have a rather wide range of control like you mentioned.
Maybe a dual port EBC would have worked better, maybe it was just my tuning skills but I put a ton of time into it and could never get it to be consistent. I would tune for one set of conditions and the ocilations would come back under another set of conditions.
What brand of dual port actuator did you use?
I have an external wastegate setup. It's a Turbosmart Hypergate (45mm, I think), with the standard Turbosmart actuator on it. I'm driving it with a MAC 4-port solenoid that I bought from Amazon.
My 2560 was a traditional single-port actuator. I believe E02K and Aidan are using dual-port actuators with 4-port solenoids on their internally-gated EFR turbos, though.
Dual port ebc fixed all my problems back in 2012. The exhaust pressure was forcing open my internal WG at higher RPM and no amount of spring pressure could fix it. I think it was partly the fault of running a log manifold and an internal WG with a lot of boost, but switching to dual port was like night and day. The wiring and the logic for the four port solenoid was identical to a three port but it worked a million times better.
Seeing a trend, dual port actuator needs a 4 port solenoid. I really wanted the wide range of boost control because I was going to implement traction control and wanted the MS3 to have quite a bit of control of the power available. I'm still going to do that some day but at this power level I don't see the need, the 245's do a darn good job getting off the corner.
Unless you're running a preset boost for each gear/driving condition, it won't react fast enough. Can't you momentarily retard timing or something to control power? That would give you a very short feedback loop.
You have a few different options for traction control, but yes I've seen spark is the quick feedback loop. Boost would be slower and you could certainly use the boost by gear feature to get you close.
Spark retard didn't work for me. If you break loose while spooling boost is increasing so far that spark retard wasn't cutting enough power, just spiralled out of control. Spark cut seemed to work though.
A wide range of EBC is desirable because you can program it into your TPS map and get a more progressive throttle.
For TC you really need to do individual cylinder fuel cuts. I have a RaceLogic unit that does this very well, but AFAIK nobody has set up an MS to do this yet.
Were you trying to use the dual-port actuator with a three-port solenoid? If so then I'm not surprised that didn't work very well because you need a way for pressure to escape after switching the solenoid. With a 3 port, switching the solenoid leaves the singleton port closed after switching, so it has to bleed past the seals which is going to be slow. With a 4 port solenoid it switches the other port to atmosphere, allowing that pressure to vent quickly.
The traction control stuff is really just a feature I'll play with later, traction is a non issue past 1st gear unless you're turning and I wasn't going to try to set this up so I could mat it at Apex and ride the TC out of the corner.
So I had my ebc setup to regulate only the open the wastegate side of the controller, the close the wastegate side got full boost all the time. So 0% boost ment you were on spring only and I had very little preload, so 130 kpa was possible. 100% ebc ment you had spring and and boost pressure keeping the wastegate closed and the top of the controller was vented to atmosophere. I set it up this way by advise of one of the old guard on here, need to look back and see who suggested it.
09-04-2017, 05:59 PM
0
