MR2 Owners Club Forum banner

2GR Gutting ACIS?

8K views 28 replies 4 participants last post by  Alex W 
#1 · (Edited)
As you might have guessed from my head porting thread, I'm looking for more low cost / low effort ways to squeeze a little more out of a stock 2GR. Not that I'm dis-satisfied with the power, but I just can't leave well enough alone. And once I start wondering if something *might* make a difference, I want to know for sure!

So, any thoughts on what if anything might result from completely gutting the ACIS system and plenum divider?

We know the ACIS system as-is helps torque in the mid range (~2800-4300 rpm):


If I am understanding the system correctly, with the ACIS flapper closed the manifold is divided in half horizontally nearly to the throttle body, with half of the runners taking off from each half of the manifold. When the flapper opens these two cavities are connected, allowing all of the runners to draw from an effectively larger plenum, and with reduced runner length.

But here's the thing. The flap is actually quite small relative to the rest of the manifold. From looking in through the throttle body, I would say maybe an inch wide by a few inches long, and it only opens about 45°. So while opening it might have a decent effect on intake resonance, any flow exchange that might be trying to happen between the two halves of the plenum will be pretty restricted still.

What I'm considering trying is getting another #2 manifold, cutting the ACIS actuator out of it, grinding out the divider to make one open plenum, and then sealing up the ACIS hole with a block-off plate. Might be necessary to cut a hole in the bottom as well for access, but a little sheet metal, some sealant, and a few pop rivets should take care of that and not be visible when done.

Thoughts?
 
See less See more
1
  • Like
Reactions: TEKSMRT
#2 ·
I am really curious to hear what comes of this. I suspect you won't get a noticeable difference but this is one of those things you simply don't know until you try it.

It would also be interesting to play with intake pipe length while talking about cheap gain possibilities.

beyond that it would be neat to design a higher ratio rocker to get more lift without a cam regrind but the cost would probably end up being higher than just getting regrinds.
 
#3 ·
Yeah, worth a try I think. I wouldn't have thought there was a worth while difference between the stock manifolds until I tested it.

Looking at the dyno graph above and doing a little extrapolation, it looks like if the ACIS was held shut all the way to redline there would be a huge loss (looks like it would be down about 10ft-lb at peak torque, and probably a lot more at high RPM). So if opening that small flap causes that sort of gain, it seems plausible that opening it up further could be worth something as well.

New manifold is supposed to arrive on Monday.
 
#5 ·
Any thoughts in trying to 3D print a larger plenum and epoxying/bolting it together (I'd assume it'd be done in like 2-3 pretty big pieces at a minimum)?

Or a fabricated aluminum manifold, but I feel like on an NA application you could make a PETG plenum easily work and be pretty easy to print. Nylon or ABS would be a little stronger at higher temps, but harder to print large stuff at good tolerances.

I would think the worst case loading would be sitting in traffic with everything super heatsoaked, but the plenum should probably still stay below about 160 F, and PETG is strong enough to probably get her done. You could always wrap the finished piece in fiberglass/carbon fiber and you'd easily have enough strength then.

Gutting the stocker is definitely the easy button tho'.
 
#10 ·
A little, and I will try to do a bit more with some sand paper before I close it up. I also think I will try deburring the edges with a torch, I think that should smooth things out nicely (the plastic melts and flows pretty readily).

Oh, and is there room on the plenum close-out to make the plenum volume larger? Basically bump it out to where it just clears the engine? That should shift power up a slight bit.
Maybe a little, but there isn't a whole lot of room to the valve cover. And it would be a lot trickier than just a flat closeout. The plenum volume will be increasing a bit regardless just because of all the stuff that was removed, and the area at the end that was the ACIS actuator adds a bit as well.
 
#12 ·
Well, this turned out to be one of those "Toyota knows what they are doing" things. Preliminary testing based on MAF readings show exactly the same mid range loss as with the ACIS unplugged, and zero top end gain. If anything, it lost a couple of g/s at the top end, but it's hard to pick 1g/s out of the noise on a log of MAF readings.

A little surprising, it looked like it should make a difference! Maybe it would if something else was changed? Could be that the stock manifold is ideal for a motor with stock cams and as you do other things to increase flow the manifold would become a restriction.
 
#13 ·
Alex, I really appreciate you trying this kind of stuff to see if there is anything there. It would really be awesome to get that last 10hp to get to that magic 300rwhp.

According to DDPR, the thing holding us back from that last 10-20hp is longer primary exhaust runners. According to the online calculators the current runners are the right size to make peak power near 7500RPM and if they were to be about 10" longer we'd move that peak power down to 6500RPM or so which would be more appropriate for the current cams. Maybe that's why my headers/tune responds so well to those MWR cams because they move the peak power to that 7500RPM range and everything is in sync then?

Maybe with those the slightly shorter runner setup due to opening the ACIS valve and the plastic around it would make a difference there?
 
#14 ·
Yeah, that's what I'm wondering. And yes, I was really hoping for a cheap way to break that 300 number! I'm not going to throw it away, if I ever get cams I might test it again just in case it helps once other things are flowing better. Longer exhaust runners might be nice, but man that would be a pain in the ass to package.

You say the headers respond well to the cams... have you seen a dyno with the cams yet? I'm really hoping Mike Reed will dyno his car...

I will say one thing, that ACIS module is a brick! For the "every ounce counts" weight savers out there, the ported manifold was 1.8lb lighter than the stock one (most of that being the ACIS).
 
  • Like
Reactions: Drag0nz75
#15 ·
I’m not 100% sure it’s my headers but the whole configuration, changing only the cams responds very well to MWR’s stage 1 cams. essentially the horsepower rolloff moves from 6500 to 7500. it only gives a touch more peak power but makes the power band way wider giving you a bunch more area under the curve when accelerating. I don't have a dyno sheet to point to right now but yeah, Mike should be posting some soon-ish.
 
#16 · (Edited)
We know the ACIS system as-is helps torque in the mid range (~2800-4300 rpm):
...the cams responds very well to MWR’s stage 1 cams.
As you know I have the MWR Stage 1 cams. So this thread gave me some motivation to run out and check the optimum ACIS engagement point with the cams. Is it 4450RPM? Or is it something different?

As background, I had been tweaking a little bit about the shape of my MAF curves and I even went so far as to re-open an ages-old thread about Brake Specific Fuel Consumption on the 2GR-FE, looking for answers. Ha ha.

Anyway I got some VERY unexpected results that are worth reporting here.

The experiment: WOT pulls in third gear, logging MAF versus RPM.

RUN GROUP A: ACIS disconnected from harness (NONE).

RUN GROUP B: ACIS held open by a portable 12v battery (BATT).

RUN GROUP C: ACIS connected normally(NORMAL).

Rectangle Slope Font Plot Parallel



The reactions that I expect:

"Oh wait, you said you had three sets of runs, on the chart, there are only two." Yes, there are three sets of runs. Two of the sets super-impose almost exactly. Set for "ACIS held open by battery, BATT" and set for "ACIS Normal, NORMAL" superimpose nearly exactly. So that's why it looks like there are only two sets on the chart.

"Oh wait, with ACIS open all the way, held open by the battery, you should get more torque (more MAF) at low RPM, and less at high RPM. Your chart shows the opposite." Yup.

"Oh wait then, you must have mixed up your data sets." That's what I thought too, so I double checked and cross-checked everything even comparing with separate runs with ACIS normal taken on different days independent of this experiment. Everything checks out.

In case that you're wondering, the maximum difference in MAF is about 19-20 g/s, which if we multiply by 1.32, is about 30HP at the crank. Like free money, this is something I would never turn down.

Finally, almost miraculously the curves seem to cross almost right about 4450RPM.

Would somebody please help me to understand what's going on here?

PS> I have to consider the possibility that the ECU for some reason is not operating the ACIS the way it's supposed to. I'm attaching the BGB section on the ACIS for how to troubleshoot that.

PPS> I was expecting I would get a DTC for running with ACIS disconnected. There was none, and BGB does not list one.

PPPS. I've found a nice setup with the OBDLink MX bluetooth and Torque Pro that gives decent enough data acquisition rates for tuning. I'll write that up separately elsewhere.

PPPPS. I think I've got it there is a possibility that my harness is shorting the ACIS to ground and not releasing when commanded by the ECU. This would explain what I'm seeing. I believe that the way the ACIS works is it receives 12v on one terminal with ignition, and then the ECU grounds the other terminal to open it. Anyway that's how it works when I test it with my portable battery. So if the ground wire is shorted to ground, then the ACIS remains open all the time. I'm a gonna get down there with my voltmeter and find out if this is it or if the ACIS ground is connected to the wrong terminal on the ECU or what.
 

Attachments

#17 ·
So, i 100% agree that there is something fishy going on here… but not quite in the direction you’re showing.

The unpowered state of the ACIS is open/short runners.
The powered state of the ACIS is closed/longer runners.
http://www.tsienna.net/images/books/424/76/index492.jpg

So applying power i would expect more power down low and the top end to suffer, your batt graph shows this.
removing power would show the opposite, your none graph shows this also.

The only oddity is it looks like your ACIS flap is stuck closed when the ECU is plugged into it. This state is easy to replicate, just check the voltage across the pins at idle. it should be 0V. if it is 12V you’ve got an issue to track down.

The other thing to consider is that the MAF curve should look similar to the horsepower curve (except that the pumping losses are not removed and a few other things like AFR not accounted for. but your “none” curve looks much more like the stock HP curve than your “normal/batt” curve. 



Is it possible your ACIS is wired incorrectly and is constantly powered instead of ECU switched?
 
#19 · (Edited)
check the voltage across the pins at idle. it should be 0V. if it is 12V you’ve got an issue to track down.


What Marc said!
Let's get on the same page.


Rectangle Slope Font Parallel Diagram



This is an excerpt from the pdf attachment that I added previously.

If I read this right, the ACIS is powered with 12v on ignition, and grounded/ungrounded by the ECU for operation.

Diagnostics - for clarity this includes the connector and pin numbers:


Font Parallel Rectangle Document Number



1. ACIS harness connector pin 2 - ground: 12v with ignition - CHECK.
2. ACIS harness connector pin 1 - C55 pin 107: 0 Ohm - CHECK.
3. ACIS harness connector pin 1 - ground: 50 kOhm - CHECK.

So unless I can come up with a way to find a bad wire or short to ground, I have to conclude the problem is with the ECU.

Thoughts?
 
#18 ·
What Marc said!

Regardless of what the problem ends up being, you have found an issue that was costing you a bunch of power, so that's good!
 
#20 ·
In theory yes, that points to the ECU but i would confirm the voltage during actual operation.

In my experience when people have issues like this it ends up being some goofy undocumented wire that is getting ground or ignition incorrectly causing some behavior different than expected.

"I had been tweaking a little bit about the shape of my MAF curves". You may also want to try putting the tune back and seeing if the issue goes away. There are a few spots while trying to change the tune that i got some really odd behaviour that i really did not think had anything to do with where i was changing things.
 
#21 · (Edited)
Definitely I have some work to do to figure out where this
odd behaviour
is coming from.

Meanwhile I have some good data that is useful in estimating the effect of the ACIS, See the graph below. The ACIS contribution seems to start about 3200RPM and makes a maximum of about 8 g/s of MAF or about 10BHP at 4200RPM. My cross-over point seems to be at about 4600RPM which is just slightly more than the OEM spec 4450RPM. This could be due to the MWR Stage 1 cams, or the exhaust, or other factors, hard to know for sure at this point. Ha ha.


Rectangle Slope Plot Font Parallel



PS. Someone in the Technical Writing departement at Toyota decided to make a better representation of the ACIS system than what we find in the Repair Manual. This is from the Toyota Venza Owner's Manual.

Slope Font Parallel Circle Pattern
 
#22 ·
That's good data and i appreciate you posting it. I'd really love to get an actual dyno chart with your cams to see what the gain is.

I suspect your cams are causing some loss down low because i can't see Toyota putting that whole unit in there for the ~10hp in the mid range. I know Alex looked this all up on regular cams so hopefully he pitches in with the normal cam differences.
 
#25 ·
So... Alex did post a dyno chart with two lines, that I take to be "ACIS plugged in" and "ACIS not plugged in," at the top of this thread. The HP difference between the two lines on the chart looks consistent with what I measured. At least that's what I thought the two lines meant...

If you can gain 10HP at 150HP which is where most people drive most the time, that's significant and will add some liveliness to the car.

Anyway I'd be interested to do a MAF-to-MAF comparison with Alex. With correction for temperature, humidity, and pressure, we should be able to compare meaningfully between volumetric flow rates.
 
#23 · (Edited)
Dyno with ACIS on vs off is in the first post.

I don't even think it's 10hp. Maybe 10-12 ft-lb of torque, which at that RPM is less in power. It does seem like a lot of effort on Toyota's part for relatively small gain.

I never investigated holding it shut to see how much power would be lost, but it looks like a lot. Interestingly, it looks like almost zero loss at ~7700rpm (based on the MAF data), but a big loss between 4500 and 7500. Seems odd that the long runners would become optimal again at very high RPM.
 
#26 ·
I never investigated holding it shut to see how much power would be lost, but it looks like a lot. Interestingly, it looks like almost zero loss at ~7700rpm (based on the MAF data), but a big loss between 4500 and 7500. Seems odd that the long runners would become optimal again at very high RPM.
I was thinking about this too and there looks like there is possibly a higher crossover point at 7700RPM.

Personally I don't think it's a cross-over point I think it's where other factors are choking the engine and limiting mass flow.

Maybe next weekend I'll go and do some runs to 8100RPM which I can hit currently if I want to and we'll see how that looks up there.
 
#29 ·
I wasn't planning on dynoing it, at least not as a stand alone trip to the dyno. Maybe if I am there in the future to test other things. But the MAF numbers (back to back, same day, same stretch of road) appeared to show a power loss with the gutted manifold, or at minimum no gain.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top