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.
If you want 18 psi (225 kpa) of boost, you do not have to put 225 in the corresponding cells.
The engine may react differently to boost under different rpm and load conditions, and trying to come up with a universal set of PDI values for every single condition will be an unpleasant experience.
What my friend did was come up with PID values that would work with general conditions, and then modify some of the target values on the map to achieve the desired results.
In other words, use whatever deviation the system may have to your own advantage.
With this setup, boost quickly climbs to 18 psi, and stays there.
It appears to me, on the surface, that you are using the throttle to control MAP, rather than wastegate. That the wastegate(s) would not be told to open until the MAP sensor at the manifold sees 18psi, and that until then, the boost upstream of the throttle could be much higher. Up to whatever pressure the WG springs allow with no diaphragm pressure.
Of course, if not much air is passing the throttle, then not much exhaust is trying to spin the turbine.
EDIT: what does boost duty look like on a log? Say 4000 RPM at 10% throttle, and then rolling it in from there.
It appears to me, on the surface, that you are using the throttle to control MAP, rather than wastegate. That the wastegate(s) would not be told to open until the MAP sensor at the manifold sees 18psi, and that until then, the boost upstream of the throttle could be much higher. Up to whatever pressure the WG springs allow with no diaphragm pressure.
Of course, if not much air is passing the throttle, then not much exhaust is trying to spin the turbine.
EDIT: what does boost duty look like on a log? Say 4000 RPM at 10% throttle, and then rolling it in from there.
My boost reference is on the IC outlet, a couple inches before the TB.
As for the actual pressure upstream of the throttle, that is rather irrelevant since I am only concerned with the pressure inside the IM.
Also, my WG pressure is 7 or 8 psi, depending on ambient temps.
I will log various boost conditions as soon as I have time. I am curious to see those logs, too. I realize I have much to learn.
I keep doing thought experiments and going round and round. Not sure yet if I think you are doing something that is not advisable, or if the MS FW writers have massively complicated EBC.
I can't speak to this application, but for motion control applications in precision machines with fixed control architecture (e.g. only PID or Open Loop) input shaping is a very effective means of achieving better performance than something typical like stage acceleration feedforwad. Granted this is all usually done with a non-trivial algorithm to turn it into a convex optimization problem, but if you're getting results for transient response without winding on enough integral gain to cause issues given the limited actuation bandwidth, more power to ya.
Maybe I'm confused, but he doesn't do anything different than anyone else for EBC. He sets actual boost targets at high RPMs, and he sets an artificially lower duty at lower RPMs when you'd typically see overshoot and boost cut.
So it's still overshooting, but it's fake overshoot to an actual target you are after.
While a neat idea, I don't think it would fix it completely or 100%. Pretty sure most of us bump overboost because the conditions to make that boost level are never the same. So either the tuning was done in some way to take that into account, which I don't think possible on limited MS3 boost control, it'll still be able to run over target on occasion (cold days) and probably lag under target on much hotter days. Tuning it to work great one day is pretty easy, getting it to work religiously all year, nah. Not as it is in code now.
Maybe I'm confused, but he doesn't do anything different than anyone else for EBC. He sets actual boost targets at high RPMs, and he sets an artificially lower duty at lower RPMs when you'd typically see overshoot and boost cut.
So it's still overshooting, but it's fake overshoot to an actual target you are after.
While a neat idea, I don't think it would fix it completely or 100%. Pretty sure most of us bump overboost because the conditions to make that boost level are never the same. So either the tuning was done in some way to take that into account, which I don't think possible on limited MS3 boost control, it'll still be able to run over target on occasion (cold days) and probably lag under target on much hotter days. Tuning it to work great one day is pretty easy, getting it to work religiously all year, nah. Not as it is in code now.
Doesn’t look like MS3 to me.
I’m referring to the situation at 50% throttle and 4K RPM when full boost is requested. I’m not sure what the turbos can output under that condition, but the EBC will be dumping fully, I would suspect. If not immediately, then shortly, as, again, the table is calling for 18 psi in the manifold. My concern is that the turbos are capable of say, 25psi, so they will drive to that until the IM sense line sees 18.
What I’m ignorant about is where the turbines can drive the compressors under that part throttle condition. It may not be an issue. In a sense, it is like the a turbo with an 18psi wastegate spring, and one is at partial throttle, only desiring 10 psi. The turbo naturally (turbine and compressor matched) does not drive to a high pressure differential across the throttle. However, it could be worse than a 18psi wastegate if the EBC is dumping to atmosphere.
We we typically use something like a 9 psi wastegate, with 9psi being called for at about 50% throttle, and then dialing up to 18psi at 100% throttle.
Hakan, when I say sense, I mean the line going from the IM to the MS. I think you are referring to the source of the air pressure that drives the wastegate actuator, which is located between the outlet of the Intercooler, and the throttle body.
Not too long ago Car passion Greg unhooked the boost pipes on his 6258 and made some pulls with the wastegate closed. If I remember right it never cracked a higher RPM on the wheel speed sensor than it did at 21PSI. Didn't really make any sense to me at first, but I guess the turbo needs the turbo to spin to full speed.
I'd try it against a partial throttle but my turbosmart dual port gate is binding garbage and anything I do would probably vent my block. Super stoked they came out with a can that doesn't suck ******* *** yet of course they won't just sell the swivel joint rod. Another $250, alright...
I have decided to never worry about my 5 speed transmission, so I have started looking at 6 speed BMW gearboxes and BMW or Mercedes Benz diffs.
The DCT option was on the table for a while, but raising the engine for clearance is almost impossible, plus it will be VERY expensive. (Very few DCT equipped Bimmers over here, and whatever gearbox is available is a fortune.) 6 speed manuals are dime a dozen, and are pretty much in the $200 range. So are the diffs. Rebuilding a BMW diff is also very easy over here.
I will not buy a conversion kit.
8 to 1 currency rate, shipping halfway around the world plus iffy customs rules and taxes makes buying a kit prohibitive.
I will not use a BMW clutch. The setup will have a custom twin plate (one sprung hub, one solid plate) Miata clutch. 225 mm vs 240 means a lighter, more compact setup. Stock pedal effort and engagement range.
Adapter plate will be hand made by yours truly out of 20 mm 6061 T6.
Flywheel will be a hella custom CNC unit, out of 6061 T6 aluminum with a steel face, and will sport a Miata ring gear.
I have not decided on flywheel weight yet. Any input on weight will be appreciated.
A custom driveshaft with heavy duty u-joints is no big deal, so are the hybrid axles.
VSS will be provided by a Crank sensor mounted on the aluminum driveshaft adapter just behind the gearbox. Building the necessary circuitry is pretty simple. 4 equally spaced M6 bolt heads will provide the reference points for the repurposed Crank sensor.
I will make a quick and dirty adjustable shifter and linkage to find the best throw/feel, and then final configuration will be cnc.
There are plenty 19 mm bore clutch master cylinders available, and I can get them for next to nothing.
I will definitely incorporate the PPF in the new setup. The car is subjected to separate and distinct stresses, resonances if the diff is secured to the rear subframe. Please tell me if I am wrong.
And, just selling my 4.1 Torsen diff and axles will almost pay for the whole project... It just might cover it. They are very rare here.
Of course I am going to start a new thread on this adventure... I just hope Dave Calzada does not get upset since I am not buying his swap kit. Paging @KMiata!
I am neck deep in rpm-speed-shiftpoint-cruise rpm etc graphs at the moment. I want to do my homework right.
This topic is relevant to my interests... More than happy to help proto/measure, though I suspect you have what you need available to you
That would be great. I will be posting all measurements, cnc files, whatever info I can share.
It's just a matter of deciding on a gearbox-diff combo right now. Only then I can measure release bearing depth, make allowances for the adapter plate thickness and Miata twin plate clutch stack height, to come up for actual flywheel depth.
Maybe I should give you a call - in a couple weeks, after the honeymoon.
Had some Iskender and kazandibi with pistachio crumbles from the local kebab place today. Not quite the same without tea and someone smoking a camel, but still very crazy how much taste/flavors can bring back memories
Feel free to call whenever We're taking a vacation(honeymoon?) from the 19th-2nd, so we'll be extra free then.
Gents, I need your opinion on a curious/unpleasant phenomenon.
My engine shuts off unceremoniously at full load, full throttle, nearing redline.
I then have to cycle the key to fire it back up. Trying to bump start while the car is rolling without cycling the engine does not help.
This happened 3 times over the past 2 years, and the latest one was about a week ago.
I have no logs, neither can I log it, because trying to duplicate this condition produces no results.
My ECU must be resetting or completely shutting off under some extremely rare circumstances. That's all I can think of.
It's an "Enhanced" MS2 by Rev. The car has been running great with it since 2012.
I mean, I torture tested the car at the F1 track and this never happened there (OK, I did lose a cam sensor once on the track).
What do you think?
Where do I start chasing such an "ethereal" problem?
msq attached...
This problem came back big time.
I was able to log the engine shutdown on three different occasions.
I am attaching all three logs. They are short, with unhappy endings.
My old DIYPNP MS2 used to do that, but it would automatically bump start itself back going again. It was an issue with the power to the board in my case, I retouched all my jumper wires and it quit doing it. If I wasn't on the laptop logging you never would have known the MS2 reset, it was just 1-2 seconds of dead.
I guess it could be flipping to the bootloader mode, you could try pulling the boot signal up to 5V/VREF. Too bad it's not repeatable, otherwise checking with port monitor like JSM suggests would be a good tell.
Hakan, the logs look like a power loss on the ms board. It can be internal to the MS board or on the board power/ground supply.
As it doesn't come back alive while a complete electrical shutdown is made, If I had to point a culprit, i'd say it's a relay.
Keep a multimeter on the car and when the problem happens again, in a safe place to do so, read the voltage to the MS.
Had an intermittent problem similar to yours on my MS3 accord. It was a defective ignition switch - vibration would turn it off for long enough to stall the car, and would turn on before beeing able to find out what happened. MAJOR PITA to diagnose.
10-29-2020, 12:46 PM
0
We're taking a vacation(honeymoon?) from the 19th-2nd, so we'll be extra free then.
