General MAP sensor connection for Speed Density?

[turboglenn]

Man, I think you guys are way over complicating the MAP sensor. The MAP sensor as a whole is much less complicated than trying to calibrate the MAF meter from the DSM in ECMLink.

In the dealer manuals I have, and seminars I have attended to be able to be labeled a "certified" Haltech dealer/installer, the only "rules" for mounting a MAP sensor are as follows. (and I'm grabbing the manual to copy verbatim from)

The MAP sensor is usually mounted high on the engine bay firewall or inner guard using two screws and with the hose nipple facing outwards. Connect the sensor to the inlet manifold via a short length of vacuum hose and fasten with either hose clamps or nylon cable ties. Connect the sensor to the main wiring harness using the appropriate plug. (For 1 Bar sensors the plug is green, for 2 and 3 Bar sensors the plug is orange). Avoid mounting the sensor below the level of the fuel injectors, because fuel may collect in the vacuum hose and run down into the sensor. The sensor assembly is weather-proof but it is good practice to mount the sensor in a protected position away from moisture and heat. 1.3.2

Then notes from the seminars I've attended mention only the following.

1.) Do not share a vacuum line via "T" fitting with anything that has a diaphragm or moving part actuated by vacuum/boost pressure such as a BOV or wastegate. And keep the lines as short as possible.

2.) Keep The vacuum line as short as possible.

3.) If it's not possible to give the sensor a dedicated line from the intake manifold, use a separate vacuum manifold mounted to the firewall that has a feed line with a larger inside diameter than the accumulated diameter of all the hoses attached to it (Avoiding both wastegate control and blow off valve being fed from this manifold)

4.) One of the few places you can reliably "T" the MAP sensor into is the fuel pressure regulator. (Which is often where boost gauges are "T'd from as well )

 

[Calan]

Man, I think you guys are way over complicating the MAP sensor.

Nothing complicated about it. Mounting it directly on the manifold provides a smoother and more accurate signal to the ECU than remotely mounting it with a line, for very logical reasons. The logs speak for themselves.

Avoid mounting the sensor below the level of the fuel injectors, because fuel may collect in the vacuum hose and run down into the sensor.

Maybe it's just me, but that makes no sense at all Glenn. No matter how or where you source the MAP, how is fuel going to work it's way back out of the IM and into a vac hose that is tapped above the injectors (on a stock DSM manifold) and terminated at one end?

As I understand it, the theory behind allowing the MAP sensor to be tee'd into the FPR is that since the FPR is terminated, it "shouldn't" affect the signal seen by the sensor, which makes perfect sense on paper. But we all know that isn't really true, because the FPR diaphragm flexes, and that air in the line is still compressible. Let's not forget that FPR Solenoid (if you still have one) vents to ATM on hot starts.

IMHO, the MAP signal should be isolated and as close to the throttle blade as possible in order to generate the most accurate signal for the ECU with SD. For a boost gauge; no problem. But for the primary tool used to calculate fuel requirements, I want as much accuracy as I can get; especially since my car is evil and hates me.

 

[turboglenn]

Although it doesn't make sense, it wouldn't be in the manual if there hasn't been some instance in which it's happened. Remember this isn't a "DSM only" manual that the quote came from, it's from a Haltech installation manual for a standalone meant for ANY car

I agree there's nothing complicated about it, the only few guidelines in all my books and notes are what's above. I've run mine on it's own signal and as well from a separate vacuum manifold and get a perfectly reliable signal. You can tell when it's hooked into anything with a diaphragm. As for the air compressing, it's only gonna compress to the same pressure as what's in the manifold. I.E. If you have 25 psi in the manifold, you won't have more or less in a line connected to it.

 

[Calan]

As for the air compressing, it's only gonna compress to the same pressure as what's in the manifold. I.E. if you have 25 psi in the manifold, you won't have more or less in a line connected to it

That's not necessarily true though, because you have to account for the expansion of the hose, and the fact that you are compressing air in a "pipe" before the true manifold pressure is felt at the sensor (as well as the reverse affect when the IM is under vacuum). With the sensor mounted directly on the IM, the sensor element feels the true pressure of the manifold almost instantaneously.

Think of it like this (on a larger scale for clarity)...

You turn on your water faucet in the back yard, let it run for 1 sec, then turn it off, back on and then off immediately. What does the person holding the other end of a 50 foot hose see at the same time you do this? Besides a delay, the water pressure at the end of the hose also fluctuates due to friction and expansion in the hose...which doesn't exist at the faucet. The end result is that you aren't seeing a true real-time representation of your water pressure at the end of the hose. (With air the effect would be even greater since it is much more compressible than water).

Granted that's a highly exaggerated example, and with the MAP sensor it's a minor effect... but at the frequency that the ECU is sampling the MAP sensor, it can be noticeable. The log screenshots I posted clearly show that, and I see it in every log I record. Now as to whether or not the effect translates into a change in performance, that depends on too many factors to list. Some people definitely see it, maybe as tip in issues or unpredictable AFRs during spool (which was my case), while others may not notice any difference at all.

Good discussion as usual Glenn.

 

[9!'clipseDOHC]

Think of it like this (on a larger scale for clarity)...

You turn on your water faucet in the back yard, let it run for 1 sec, then turn it back on and then off immediately. What does the person holding the other end of a 50 foot hose see at the same time you do this? Besides a delay, the water pressure at the end of the hose also fluctuates due to friction and expansion in the hose...which doesn't exist at the faucet. The end result is that you aren't seeing a true real-time representation of your water pressure at the end of the hose. (With air the effect would be even greater since it is much more compressible than water).

Good ole Friction Loss. (I used to be a firefighter).

 

[turboglenn]

Calan, with that analogy I can see your point, and although I do agree that such a phenomena could and probably does take place. The only time I've had real issues with the MAP sensor line/mounting is when connecting it to the BOV line with a T one time when I redid some lines and didn't have a spare nipple or vac manifold to install. The car would falter under large on to off transitions, but other than the the signal from the GM sensor to the Haltech has always been really reliable and accurate.

Sorry if I was a little wrong with "over complicating" I forget that other people often like to think a little deeper than normal too, but I just feel like I'm ranting when I do it.

 

[blazinglaser_90]

FWIW I run a MegaSquirt with a 4 bar MAP sensor that is installed behind my center console. That makes my vacuum line around 4 feet in length. On my data logs I have extremely smooth boost curves, I have great throttle response and the car went 10.4 off the trailer on street tires and a street tune so maybe the distance away from the vacuum source isn't a huge deal?

 

[97gst_spyder]

For anyone that runs a stock 2g IM, the OmniPower 4bar is great! Stock connector, and no extra drilling/tapping.

 

[DashLaflash]

For the Evo 3 intake manifold I see the brake booster being the best spot to thread in a MAP sensor. This leaves me with the issue of having to drill and tap another line for the brake booster though. Is there anyplace that I can tee into for the brake booster line? PCV line maybe? I just don't want to loose braking ability with a poor vacuum source.

 

[Calan]

For the Evo 3 intake manifold I see the brake booster being the best spot to thread in a map sensor. This leaves me with the issue of having to drill and tap another line for the brake booster though. Is there anyplace that I can tee into for the brake booster line? PCV line maybe? I just don't want to loose braking ability with a poor vacuum source.

I'm not familiar with the Evo 3 IM, but I wouldn't recommend tee'ing the brake booster line in any case. It should have it's own dedicated port with a solid fitting.

Loosing a BOV on a curve at 60mph is one thing; losing brake pressure is another.

 

[DashLaflash]

Ya I hear what you're saying Craig. I guess what I would really like to do is take the port on the manifold where the PCV dumps into and use that for my brake booster. Then just leave the PCV port open (Keeping the check valve of course) and letting it dump in my engine bay. Is that a bad idea or will it cause a fire or something? I know it would be best to run it to a catch can but I don't want to have to buy another one.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

 

[snowborder714]

I'm sure Craig will tell you all the reasons why leaving the PCV open is bad, so I'll let him do that

Why not just get a fitting/nipple/bung welded on the manifold for the PCV?
http://www.dsmtuners.com/forums/152320673-post602.html

 

[Calan]

I'm sure Craig will tell you all the reasons why leaving the PCV open is bad, so I'll let him do that

Why not just get a fitting/nipple/bung welded on the manifold for the PCV?
http://www.dsmtuners.com/forums/152320673-post602.html

Yep...with it left like that, it will be permanently open since you don't have IM vacuum going to it. In other words, it's not doing anything at all.

I agree with Brian... just weld a bung on and call it done. For that matter, you can usually just drill and tap the plenum and thread a barb nipple directly into it, as long as you don't over-tighten it and strip the threads.

Welded bungs are better though. Something like this:

You must be logged in to view this image or video.

 

[snowborder714]

That is also assuming that the PCV is in good working condition, correct? If the internal check valve isn't working anymore, would your statement still be true?

I'm not sure if you can just drill and tap into the Evo plenum. I don't know how thick it is, so this is something that would need to be looked into. I don't remember it being very thick though.

 

[Calan]

...take the port on the manifold where the PCV dumps into and use that for my brake booster.

For some reason, I just now got this and saw the pic.

I don't think that is the best place to source the brake booster. If that IM is the same as the DSM's, that PCV port ties into a galley that runs the length of the manifold, which is in turn tied into the base of each runner through a very small hole. I would much rather see the brake booster being fed from a large port directly off the plenum, instead of a small secondary source coming off the runners.

Brakes just aren't something to take a chance on with a performance car IMHO.

That is also assuming that the PCV is in good working condition, correct? If the internal check valve isn't working anymore, would your statement still be true?

It's basically just a ball and spring inside the valve that gets "sucked" closed proportionally with vacuum (although it never fully closes in this direction as far as I know). If it's stuck open and allowing air to pass through (or working properly), then I would expect a nice oil mist to start covering the IM area. If it's stuck closed and clogged, then it would be the same as a weak plug in the VC port. Either way it's not the best idea.

I'm not sure if you can just drill and tap into the Evo plenum. I don't know how thick it is, so this is something that would need to be looked into. I don't remember it being very thick though.

If it's the same as that ^^ 1G IM, it is pretty thin. I was able to thread a fitting in and it held, but I temporarily JB welded it just to be safe until the IM came back off and I could do it the "right" way. You do have to be careful to not tap the hole too large though.

Bungs are much better.

 

[DashLaflash]

I would love to do what turbosax2 did with the Golden Eagle vacuum block but I have the coil pack relocation bracket from JMF and I don't think the two will work together. I was already running the brake booster from the source in the picture and it seemed ok for the most part. If I pumped the brakes at idle though the pedal would start to get a little hard. Maybe if I turned the vacuum block around the other way it would fit? I will probably just put my factory 2g intake manifold back on until I can get whatever fittings welded on.

 

[91stocker]

Just another example for the 1g's. Hope it helps.

 

[93_tsi_fwd]

Just an update:

I'm running my MAP sensor on the BOV port on the IM and the BOV is now on the PCV port. Will post pics when I get a camera. Works great so far.

 

[DashLaflash]

I never posted what I ended up doing so here it is. I welded an aluminum bung onto one of the runners on the intake manifold and used the factory brake booster spot to thread in my map sensor.