maxcooper

Executive summary:
1. don't use 4mm vacuum hose to connect to the GM 3-bar MAP sensor
2. Use Option2, Scale 43500, Offset 0 as the PowerFC settings for the GM MAP sensor
3. The GM MAP sensor tops out at about 31 psi
4. I'm sure glad I didn't break any seals!

Long story:
I recently had my engine rebuilt and have been doing the break-in and not really boosting much, but I have been playing with the boost up to 4500-5000 RPM under part throttle sometimes. I have been doing it enough to open the 15 psi spring on my wastegate a few times (very easy to tell since it dumps to atmosphere). I recently installed a GM 3-bar MAP sensor since I plan to run ~15 psi boost and I wanted to have a little headroom in case of an overboost, or at least be prepared in the event that I decide to run a little more boost.

I have noticed a few things that concerned me, so I have been letting off the gas pretty quick as soon as the boost builds. My wideband starts to show pretty lean (up to 13.5:1) mixtures when the boost climbs. However, I haven't heard any detonation (and I'm not sure I would since the car is so loud) and the idle is still smooth, with constant but somewhat low vacuum (~11-12 inHg at 950 RPM, 1-piece ceramic seals and a decent street port, so I am not surprised). The idle quality and vacuum haven't changed much since I first got the car back, so I am pretty sure my apex seals are still intact. I popped my last motor with the wideband still reading in the 10s (I later decided it must have been the advanced timing that I was running rather than the mixture), so I have been a bit suspicious of it's accuracy. I also have the sensor mounted further down the pipe (right before the midpipe) than I did before, so I though that perhaps the charge was still burning in the pipe giving false lean readings.

My AVC-R has only been reading up to about 0.35 kg/cm^2 boost, too, despite the fact that I have been boosting more than that. I had accidentally pulled one of the wires out of the AVC-R MAP sensor connector once when the wiring wrapped around the steering column (first evidence that I am dumb ). I reattached it and it had been working fine since then, however. I also did some datalogging and found that my peak boost reading was only "16", which is only ~2 psi.

Today I investigated the problem and found that the little 4mm hose that I stretched over the GM MAP sensor nipple had a little hole in it. That explains the low readings on both of my MAP sensors, since I feed both the GM and AVC-R MAP sensors from the stock location (with a T in the hose). So I replaced all the hoses for my MAP sensor connections, and used some 4mm-to-6mm adapters so that I could run a larger hose to fit the GM MAP sensor. This was a dumb mistake, and I am sure glad that my engine seems to have survived my carelessness. I don't know why I was so lucky, but perhaps the 1-piece seals were strong enough to take the abuse, the ceramic rotor face coating helped fight detonation, the FMIC was keeping the air cool, or some combination of very good luck and a strong motor.

After that I decided to check the PowerFC settings for the GM sensor. I had previously set it to Option2, Scale 41800, Offset 0, which are the settings that you commonly see recommended for the GM sensor. Or so I thought anyway -- when I looked at the settings in FC-Edit, I noticed that I had selected Option1 by mistake, though I did set the scale and offset to the right values for Option1. That got me wondering if the options really matter (is option1 really different from option2 if you set the scale and offset correctly?). I didn't investigate that question, however, but rather just switched to option 2 and set it up correctly.

I wanted to verify that the MAP sensors were now reading correctly, so I got out my brake PowerBleeder cannister, which is a bottle with a pump and a pressure guage that you can use to pressure bleed your brakes. I made sure the bottle was empty and attached a new hose to avoid getting any brake fluid into the MAP sensor plumbing and started checking things out. I started by making sure the AVC-R MAP sensor reading matched the gauge on the PowerBleeder. They seemed to match up well at 5, 10, 15, and 20 psi (doing the psi-to-kg/cm^2 conversions in my head as I went along), so I moved on to setting up the PowerFC to read the GM MAP sensor. The sensor was reading about -30 at atmospheric pressure, where the AVC-R read 0. I played with the offset and scale for a while and ended up keeping the offset at 0 and changing the scale to 43500. Now all the gauges read 0 at atmospheric pressure like I wanted. Then I started pumping up the PowerBleeder and checked for consistency in all 3 pressure readings. They matched up very well at 5 psi increments up to 20 psi. The PowerBleeder gauge goes up to 30 psi, so I decided to see what the limit was for the GM sensor. I pumped up the PowerBleeder past the 30 mark on it's gauge to the point that should be about 35 psi. The PowerFC/GM sensor was reading 220, which is ~2.2 kg/cm^2 or just a tad over 31 psi. I backed down a bit and pumped the PowerBleeder up again, while watching the boost reading with FC-Edit. It topped out at what looked to be about 31-32 psi on the PowerBleeder gauge, which is what I was expecting (hoping for). I wasn't paying attention to the AVC-R at this point, so I don't know how high it will read.

Then I hooked everything back up to start the car and see how closely the vacuum readings matched between the PowerFC and AVC-R. I did find that the AVC-R seems to read higher (more) vacuum than the PowerFC, but I figured that wouldn't be much of a problem so long as the fuel maps are adjusted accordingly. I'd rather have the positive readings be accurate, and I was very pleased with how that part of the calibration had gone, so I left everything alone at that point.

-Max

DaleClark

iTrader: (55)

The 4mm-6mm adapter you've got running to the sensor - make SURE it's a brass adapter, and it's clamped down well on both sides. I've seen more than one engine pop due to a failure in a plastic tee or some such. If at all possible, run a dedicated single vacuum line to the boost sensor with no tees in it and use another nipple for the AVC-R's input. It's one of those things where it pays to play it safe!

Dale

maxcooper

I've had a plastic T in that line for years and it looked as good as the day I installed it when I swapped it for a plastic 6mm T today. I will consider some alternate plumbing, but I think what I've got is pretty solid now. I used a zip tie on the GM sensor nipple since it was short and has only one barb, but the rest of the connections are all pretty long with many barbs. The little plastic air filter for the AVC-R MAP sensor line did split in half today as I was removing the old hoses. , er

-Max

DaleClark

iTrader: (55)

It's also probably the location of the tee. If they're near the turbo, they'll get mighty brittle over time. Brass is just a safe bet.

Yeah, I've seen a number of those inline filters for boost controllers and such fail like that.

Dale

maxcooper

I've got both the MAP sensors right next to each other where the stock sensor used to be. I think it is pretty cool there.

-Max

rzograbian

so after the installation of the gm 3 bar map sensor, you have to adjust it accordingly to match with the computer?? and if i use the datalogit so set it to the setting mentioned above, will it be correct??? or is it different from car to car??

also, isnt 10-11 inches of vacuume low??? how big of a port did u do !!

maxcooper

You adjust the PowerFC settings to read the manifold signal from the new MAP sensor properly. I just wanted to check to make sure that it was indeed reading the signal accurately. You should be able to just install the sensor and use my settings to get yours reading properly as well. It shouldn't vary significantly from car to car.

It doesn't really matter that it reads the manifold pressure signal accurately, so long as the reading (however mis-scaled or shifted) changes in proportion with the real manifold pressure. Once you tune the car, any discrepancies would be accounted for in your maps. However, it is easier to read the logs and compare your maps with other people's maps if your logged manifold pressure values are accurate. I think it is worth making it read the pressure accurately for these reasons.

It's a decent sized street port, but it isn't too-the-limits huge. I think the 1-piece apex seals also contribute to my "low" (11-12 or 13) idle vacuum. The car idles pretty smooth and makes good power. The vacuum is about what I expected anyway, based on the port and apex seal choices I made.

-Max

twokrx7

iTrader: (1)

Max,

Little humor, now that you adjusted your map sensor settings you might have more issues than before comparing datalogit maps as most don't change these map sensor settings.

maxcooper

Kyle,

It was the pre-tuning prep doc that you and Chuck put together (stickied in this section of the forum) that inspired me to perform the test/calibration. Realistically, I don't think I will be comparing much to other people's maps, but I will enjoy being able to read the log entries and know what my boost pressure really was.

-Max

pluto

Hey Max,

Try 44000 with 357 offset. That seems to be the most accurate for the 3 bar sensor.

try it out and let me know if you like that better or not.

CCarlisi

Steve where did you come up with those numbers????

RX794

This is why I use a 3bar MAP sensor from APEXi directly.

CCarlisi

The GM sensor is fine when setup correctly. After getting my car tuned by someone who shall remain nameless (it was not Steve Kan) 15psi didn't feel like 15 psi. Sure enough my boost gauge and boost controller were giving me different readings than the PFC. Despite this the tuner told me to assume they were BOTH wrong and to go by the pfc. I didn't buy this and decided to purchase an $80 Mity Vac Silverline to get to the bottom of the problem. Sure enough the 'tuner's' calibration was off by 1 1/2 psi at 13psi! Ultimately, I arrived at the value of 44000 with an offset of 320ish. Using this value the GM sensor was dead on at 0, 5, 10, 15, and off just a tad at 20. I have tested 3 different GM sensors with the Mityvac using this calibration and all were pretty close. The main point of variation is in the amount of offset required. Unless all the apexi sensors are dead on at the factory value I don't see the benefit of spending the extra money for one -aside from the fact that it is physically easier to install.

pluto

Obviously from you.

maxcooper

Steve, I will try those settings when I get my car running again. Thanks. I am happy with the positive pressure readings I get with my current settings, but the vacuum seems to be a little bit off.

I'm on a marathon rush to get it ready for the tuning session and have it apart currently. I'm having some stuff ceramic coated now. Embee has a "turbo coating" they say is good to 2400F (they also have 1300F and 2000F coatings) and comes in black or silver. While I am very skeptical it will last, Embee seems to be pretty on the ball so I'm going to try it on the turbine housing. Rust is ugly.

I did some further confirmation of my apex seals being intact -- the turbine looks perfect, and I actually felt my apex seals through the exhaust ports to make sure they weren't broken. I only checked 5, but they were all intact -- I felt silly so I stopped before checking the last one.

Today the car got fresh oil (thin 10W-30 Mobil 1 ) & filter, new fuel filter, and a bunch of suspension bushings. I am focusing on stuff that matters for the dyno tune, but I might as well get the rest of the car in shape while I'm waiting for other stuff. I measured the clearance for a muffler in the midpipe, too, and it looks like most of the 14" oval ones should fit (5x8, 4x9, 4x9.5) in center/center or offset/center configs. I'm going to make a midpipe with both a Random metallic cat (less smell!) and a muffler (less noise!) in it, hopefully in time for the dyno tuning session.

-Max

ghostrx7

Max what do you need that silly muffler for?... Half of the fun of owning an rx7 is pissing off the neighbors when you come home at 4 am from whatever bar/club you were at. ( Sole reason I like my turbo timer)

CCarlisi

Max:

I started with a metal substrate cat positioned at the beginning of the midpipe section and then added a round 14"magnaflow resonator a couple inches behind it. The clearance on the sides is a little tight, but there is still enough room for the muffler to move around without making contact with the chassis. I am using a 14" oval Magnaflow resonator for the catback muffler. For what it's worth, I don't believe it would fit in the midpipe without hanging down below the chassis.

maxcooper

You must live somewhere that the cops don't harass you for having a loud muffler. You can't enjoy your car much when you are paranoid about getting pulled over for an arbitrary assessment that your muffler is too loud. Loud mufflers are often attached to cars with emissions stuff removed, too, so they might want to take a look under the hood. Hot-rodding is as American as apple pie, and yet California has made it pretty much illegal. It sucks man, it really sucks.

Also, I seem to have become a "cranky old neighbor"-type at some point myself, so that "damn noisy car" that I wish would quiet down happens to belong to me.

-Max

maxcooper

That's what I plan to do, except I think I will use a 5x8", 14" long Magnaflow, perhaps a center/offset one. It will be tight, but I think we can make it fit without banging the underbody or sticking down too much. I know it will be somewhat tight, but I think the combo I am looking at should fit. I am thinking:

front of mp -- bend -- cat -- very short bend -- muffler -- end of mp

I have a few questions:
1. What made you choose the round versus an oval muffler? I basically want the quietest stright-thru muffler that will fit, but I am unsure if, say, a 5x8" oval would be quieter than a 6" round.

2. What part sticks down on your setup? Is it the round muffler?

Thanks,
-Max

CCarlisi

That sounds similar to my current setup. However, it has gone through so many iterations that I'm going to have to get under it to confirm verify where the bends are.
I'm due for an oil change in about 2-3 weeks. When I get it done I'll snap a couple pictures for you.

We measured the oval catback muffler and concluded it would not fit in the midpipe location without the widest axis being positioned vertically. In that orientation, with my downpipe, it extended below the car by a couple inches. The exterior of the Magnaflow canister does not taper down into the piping, so it is a very nice 'hook' to catch on things. I do a lot of city driving and figured it would only be a matter of time before I clipped something and it wreaked havoc on everything attached to both sides of the muffler. Bear in mind that I’m using the long muffler. If you went with a shorter canister the oval muffler could probably be positioned horizontally.

I spoke to the tech department at Borla and Magnaflow about the sound tone and suppression of oval vs. round mufflers. They both said that there is no different between the two shapes assuming the exhaust pipe is straight.

Although this doesn't really fit into the discussion, I want to mention that extending the length of the catback tip SIGNIFICANTLY decreased interior sound volume in my car. If all else fails it’s worth giving that a shot.

-Nothing hangs down with the round muffler; I was referring to the oval muffler.