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.
IIRC that's the one that feeds the VICS port -- if so, then it'll make a much bigger difference than just the vacuum leak.
--Ian
Yup! That port fed the VICS. Unfortunately, after reconnecting it and fixing the (small) exhaust manifold leak, the car only made marginally more power.
Originally Posted by codrus
Because 14.7 barometric pressure + 22.9 psi boost == 37.6 psi absolute, which is 259 kpa. You're probably a thousand feet above sea level to get it that high.
--Ian
I did some more investigating and it looks like your guess was spot on: the MPX4250 2.5 bar map sensor is only "Good for 21-22psi of boost"
I'm considering picking up a MapDaddy 4 bar sensor, but I'm concerned about being able to remove the MPX4250 without damaging the board. Is this something that's even possible within reason?
Alternatively, I'm also considering just replacing the twisted motion gt2871 knockoff with a genuine Garrett gt2560r or gt2860r, but i'd hate to throw more money at this and only end up with 5 more peak hp. Right now it's taking all of 21 psi to crack 250hp. :-\
I'm also tempted to disconnect the exhaust pre-cat to see if the flyin' miata HF cat is hurting performance, but i'm not hopeful.
That said, i've been playing with timing, fuel and boost quite a bit, and only boost seems to affect power above 5K. Bringing up the timing, hearing knock, and then backing it back down, seems to have little impact on hp. (all as per VD).
I did some more investigating and it looks like your guess was spot on: the MPX4250 2.5 bar map sensor is only "Good for 21-22psi of boost"
I'm considering picking up a MapDaddy 4 bar sensor, but I'm concerned about being able to remove the MPX4250 without damaging the board. Is this something that's even possible within reason?
Alternatively, I'm also considering just replacing the twisted motion gt2871 knockoff with a genuine Garrett gt2560r or gt2860r, but i'd hate to throw more money at this and only end up with 5 more peak hp. Right now it's taking all of 21 psi to crack 250hp. :-\
I'm also tempted to disconnect the exhaust pre-cat to see if the flyin' miata HF cat is hurting performance, but i'm not hopeful.
If it's a ceramic core FM cat, then the answer is yes.
As far as the MAP sensor goes, generally when reworking boards to remove components you have the choice of saving the board or saving the component. So yes, it's possible to remove the on-board MAP sensor without damaging the board, if you have a decent amount of skill in doing solder rework. I'm guessing you don't. (I'm not confident I could do it properly, either).
I like the off-board GM 3 bar MAP sensor from DIY. With that the on-board sensor can become a separate baro sensor.
If it's a ceramic core FM cat, then the answer is yes.
As far as the MAP sensor goes, generally when reworking boards to remove components you have the choice of saving the board or saving the component. So yes, it's possible to remove the on-board MAP sensor without damaging the board, if you have a decent amount of skill in doing solder rework. I'm guessing you don't. (I'm not confident I could do it properly, either).
I like the off-board GM 3 bar MAP sensor from DIY. With that the on-board sensor can become a separate baro sensor.
But yeah, stupid cam timing is stupid.
--Ian
Map Sensor:
Great info! A "decent amount of skill in doing solder rework" is something i certainly don't have. It looks like this is the "off-board GM 3 bar MAP sensor" you mentioned is the way to go. I'm going to order it in case I decide to push more boost through the system - that said, I'm not 100% sure that's my plan forward.
I currently have peak boost sitting just over 22psi to avoid overrunning the current MPX4250.
Cat:
I'm running the 300 cell ceramic cat that comes in FM's 2.5 turbo midpipe. You feel this is restrictive huh? Good to know, but i may still just live with it....b/c environment
General Update:
With the cam gears correct, the tacotaco manifold replaced with a FM cast manifold, and seemingly no boost leaks, here's a Virtual Dyno readout on the smoothest road i can find in my area.
It's currently running a very similar ignition map to Flyin' Miata's hydra map. There's NO knock with this map (the ONE exception, I can hear one count of knock in 5th gear at 6K RPMs ONLY if I'm running the car hard through 4th and into 5th).
Strange Voltage on Datalog:
As you can see in the screen shot about, my Voltage is showing >17. I can't explain this and need to look into it.
Next Steps:
I'm torn between: 1. Spend the time/money on dyno time to try and get the car dialed in better myself. 2. Replacing parts to unlock more potential (e.g. bigger/authentic Garrett turbo, different cat, etc). 3. Finding a reputable MS tuner familiar with Miata's and have them finish the tune. (Or some combination of the three).
In any event, i'll likely bring it to the track next week and to see what mph it can trap.
Garrett GTX2867R (a/r 86) to be delivered by end of week.
Also did some exhaust testing. Fully removing my muffler (axle back) and it made no difference in performance (according to VD). So if there are any exhaust restrictions, it's not that. The exhaust is now 2.5" all the way through (though I'm still running a FM cat). More to come.
I think I just hit 200ish HP and that was at about 10-12 PSI. This is on what is the equivalent of a 2560.
Hmmm, interesting. That sounds roughly around where I'm at with the same psi.
At this point I should have the new turbo on by Sunday. If the car is still down on power relative to PSI, I'll pull the cat off - though I don't have high expectations with removing the cat.
If the cat and the turbo are still not effective, I'll be taking the head and revisiting the valves. More to come.
I'd like to see at least 270whp with 20psi based on my other variables.
Garrett GTX2867R (a/r 86) to be delivered by end of week.
Also did some exhaust testing. Fully removing my muffler (axle back) and it made no difference in performance (according to VD). So if there are any exhaust restrictions, it's not that. The exhaust is now 2.5" all the way through (though I'm still running a FM cat). More to come.
Virtual dyno with no retuning isn't necessarily the best way to check for exhaust restrictions, because there's too much noise in the signal. A better method is this: run VEAL a few times to get the fueling dialed in, then turn off EGO correction. Do a few low-RPM to redline pulls in 3rd using the same piece of road and datalogging them. Change exhaust, then repeat the pulls and compare the logs. A freer-flowing exhaust will manifest as quicker spool, a leaner air/fuel ratio, and will also affect the boost. If you're running open loop EBC you'll get higher boost, if closed loop then you'll get lower duty cycle for same boost. (mechanical wastegate will also have lower duty cycle, but since it's mechanical it's not logged so you can't tell).
The best way of doing it, though, is to actually instrument the backpressure and monitor it. To do this, you need a pressure sensor from mouser, similar to the one used inside the ECU as a MAP sensor, and you need a way to hook it to the exhaust. I've done this by getting an O2 sensor bung plug, drilling a hole in it, tapping that hole for 1/8 NPT, then getting a 1/8 NPT compression fitting from the hardware store and a coil of copper tubing. Use that to go 4 or 5 feet from the exhaust and by that point it will have cooled down to the point that you can stick a rubber or silicone vacuum line onto it. Run that to the pressure sensor, then wire the sensor to a spare analog input on the ECU and datalog that value. You can then do pulls and actually measure the backpressure in the system. If you've got a tapped hole in the exhaust manifold for an EGT probe, then you can pop that out temporarily and add the sensor there too, to get exhaust manifold pressure.
When I did this on my car, I discovered that the "high flow" cat that I had was a much greater restriction than advertised. Swapping it out picked up over 40 hp on the dyno (with retuning).
Thanks Ian!
I should have been more clear: When i removed the axle back, I did some fuel optimization (via VEAL) and took multiple pulls on a (nearly) flat road from <3K to read line. (I did NOT turn off EGO correction though).
If there was some change in spool or performance, it was canceled out by noise/other variables. I'm not saying the axle back ISN'T a restriction, but it's certainly doesn't seem to be the biggest one currently in my setup. In other words, it could become an issue after i remove the cat, for example.
Regarding back pressure monitoring, I totally agree. If some of the easier solutions don't pan out, I'm definitely going to exactly what you outlined above.
In the meantime, here's the new genuine GTX2867R next to the Twisted Motion (eBay) GT2871.
Interesting development:
The GT2871 from eBay was advertised as an .64A/R turbine housing. I found a ".86A/R" embossed on the housing today.
THUS! I have a hunch this GTX2867R is going to make nearly the exact same HP/PSI i'm currently at!
I'm pretty sure that the 2560 runs out of steam around 15-16psi and makes more heat than anything else past that, but I've seen dynos of it (JasonSBB?) at about 270hp. And he had a 2.5 exhaust at the time afaik.
I think Vlad (18psi) does some road tuning. Maybe talk to him as a first step? I think his prices are pretty reasonable and it might be helpful to have him run through your tune to see if anything jumps out at him. I think PatMX5 does this as well. Worth a shot before you throw more money at parts...given that the limiting factor here is your understanding of tuning ish, right?
I'm pretty sure that the 2560 runs out of steam around 15-16psi and makes more heat than anything else past that, but I've seen dynos of it (JasonSBB?) at about 270hp. And he had a 2.5 exhaust at the time afaik.
I think Vlad (18psi) does some road tuning. Maybe talk to him as a first step? I think his prices are pretty reasonable and it might be helpful to have him run through your tune to see if anything jumps out at him. I think PatMX5 does this as well. Worth a shot before you throw more money at parts...given that the limiting factor here is your understanding of tuning ish, right?
Well, 22+ lbs on a ebay GT2871 is ~ 270whp. :-\ But again, that's with 8.4:1 pistons and the FM cat (all through the 2.5" exhaust).
I'm really hoping that the eBay turbo was greatly holding me back, but i'm not optimistic (especially if the ebay turbo had a .86 turbo housing and not the .64 advertised).
If the new turbo + dropping the cat doesn't EASILY make over 300whp at <24psi, I WAS going to pull the head; it's the only item left in the system that could be greatly holding me back.
All that said, thank you very much for mentioning Vlad and PatMX5! I'll certainly be reaching out to them if things don't greatly improve with the turbo swap. I probably should have already .
Well, without touching ANYTHING except fuel and boost, here's a ROUGH comparison between the Twisted Motion (ebay) GT2871, and the Garrett GTX2867R.
As you can see, the GTX blows the eBay turbo away with efficiency, making 30+whp with 3+ (15%) less psi! (Obviously the car became very rich up top).
Much more tuning to come this weekend, but I'm very happy with the rough/preliminary drive.
This is likely the cleanest/flattest pull yet. Power is ok I guess. I'll likely be bringing the car to the track tonight to see what MPH it turns (weather permitting).
I'm just kind of lost as to why your spool is so seemingly slow. Though that might just be the bigger turbo. My tune isn't great and I'm making more torque than you till a bit past 4k. That's on a JB 2560
06-18-2017, 08:49 PM
0
(I'm not confident I could do it properly, either).
