4g MAP sensor myth debunked

[cygnus x-1]

So after reading about the supposed benefits of the 4g MAP sensor I decided to do a test of my own to see what the deal really is with these things.

So I took 3 sensors, a 3g (PS10), a 4g (PS20), and another one (PS22) that I think came out of an early '90s Integra or some other OBD1 Honda. The 3 sensors were all connected to a regulated 5V benchtop power supply and were measured with an HP benchtop multimeter. A hand vacuum pump was used to vary the pressure to all the sensors.

Here are the results:



Pressure (inHg) PS10 (V) PS20 (V) PS22 (V)
---------------------------------------------------------
0 2.88 2.87 2.87
5 2.42 2.41 2.41
10 1.92 1.92 1.92
15 1.47 1.46 1.46
20 1.01 1.00 1.00
25 0.52 0.51 0.52
At each pressure the 3 sensors read almost identically with a worst case error of 2% at 25 inHg. Given these results I would have to say there will be no measurable difference in performance when using any of them.

C|

[Ichiban]

I figured as much. As far as people saying that they got mixture enrichment or a 1-2 MPG loss with the 4g map sensor, well, this:

1. The o2 sensor will detect any rich mixture condition, and the ECU will compensate by shortening pulse width

2. A 1-2 MPG loss is virtually impossible to detect without exacting testing conditions and sensitive equipment.

[Rendon LX-i]

wierd....I dont know but i just felt a difference. and yet messy had dyno results that proven that he had hp gains. Well thats good to know bout the map sensor myth

[bobafett]

yep. i have always felt that a 1bar map = 1bar map = 1bar map.

lol... nice looking out though. good to see my suspicions confirmed as well...

[Tdurr]

Yea, me and mushroom told u guys it didnt help. His dyno run he swapped out sensors, and all the other ones did was richen the fuel some. no extra power, tq curve was Super lightly affected, but no real gains.

[Oldblueaccord]

wierd....I dont know but i just felt a difference. and yet messy had dyno results that proven that he had hp gains. Well thats good to know bout the map sensor myth


I'm think you might have caused a code in your ECU and "default rich" which in your case helped your set up. I mentioned this a few times before,way in the past, but was quickly shot down.


cyg thats a great way to test a sensor, great work !


wp

[Rendon LX-i]

ummm i dont thinks so bro...When i put my stock one back it i was like wtf...i notice it on top end. Works even better now when boosted. If i use my stock map i get a code..4gen map no code. Watever it worked for me. im not shooting you down bro. i know what it did and how it performed. Its just wierd. Results are results they dont lie...maybe its justs me.

[lostforawhile]

ummm i dont thinks so bro...When i put my stock one back it i was like wtf...i notice it on top end. Works even better now when boosted. If i use my stock map i get a code..4gen map no code. Watever it worked for me. im not shooting you down bro. i know what it did and how it performed. Its just wierd. Results are results they dont lie...maybe its justs me.

are you sure your stock one wasn't bad? all these do is produce a certain voltage depending on bar, the computer doesn't know the difference, if it has the same voltage to bar curve, it works the same. it would seem like your stock one might have been going thats why it threw a code probably also.

[Oldblueaccord]

You west coast flip flopped there which is it ?--------------------------------------------------------------------------------
wierd....I dont know but i just felt a differance. and yet messy had dyno results that proven that he had hp gains. Well thats good to know bout the map sensor myth


End quote:


so its a myth, but it works for you, but im wrong, it doesn't work?

I'm not buying anything so you can be straight with me.


wp

[lostforawhile]

You west coast flip flopped there which is it ?--------------------------------------------------------------------------------
wierd....I dont know but i just felt a differance. and yet messy had dyno results that proven that he had hp gains. Well thats good to know bout the map sensor myth


End quote:


so its a myth, but it works for you, but im wrong, it doesn't work?

I'm not buying anything so you can be straight with me.


wpi'm thinking if these were swapped for stock sensors, after all these years the stock sensors were probably out of calibration. if you drive a car with one thats going and isn't throwing a code, you get used to it, if you replace it with a new one,of course you would feel the difference.

[Rendon LX-i]

You west coast flip flopped there which is it ?--------------------------------------------------------------------------------
wierd....I dont know but i just felt a differance. and yet messy had dyno results that proven that he had hp gains. Well thats good to know bout the map sensor myth


End quote:


so its a myth, but it works for you, but im wrong, it doesn't work?

I'm not buying anything so you can be straight with me.


wp

stock map worked fine or maybe it was bad....wtf is west coast flip........you like my periods lol......................Anyways The only reason i got CEL was under BOOST on my stock my with the 4gen map i DID NOT GET CEL. but it made a difference for me. so :eek:

[3rd Party]

So after reading about the supposed benefits of the 4g MAP sensor I decided to do a test of my own to see what the deal really is with these things.

So I took 3 sensors, a 3g (PS10), a 4g (PS20), and another one (PS22) that I think came out of an early '90s Integra or some other OBD1 Honda. The 3 sensors were all connected to a regulated 5V benchtop power supply and were measured with an HP benchtop multimeter. A hand vacuum pump was used to vary the pressure to all the sensors.

Here are the results:

Pressure (inHg) PS10 (V) PS20 (V) PS22 (V)
---------------------------------------------------------
0 2.88 2.87 2.87
5 2.42 2.41 2.41
10 1.92 1.92 1.92
15 1.47 1.46 1.46
20 1.01 1.00 1.00
25 0.52 0.51 0.52


At each pressure the 3 sensors read almost identically with a worst case error of 2% at 25 inHg. Given these results I would have to say there will be no measurable difference in performance when using any of them.

C|

haha, thank you cygnus! all i have to say is, "BURN!!"

[AccordB20A]

had a feeling they were all the same

[cygnus x-1]

One thing I should point out is that this still doesn't explain why Honda felt it necessary to change the MAP sensors. There is obviously *something* different about them otherwise why would they bother? It could be a number of things though; updated internal circuitry due to manufacturing changes, wider measurement range, better accuracy, reduced temperature sensitivity, better/different dynamic response, reduced electrical noise, etc. All of these things could apply and still not affect the static pressure vs. voltage curve.

As for dyno runs; (chassis) dynos aren't nearly as reliable as most people think they are. There are SO many variables that affect the numbers you will see on a dyno that it's very difficult to get consistent results. To really prove that a particular part has an effect you would need to make several runs with and without that part and average each set of results. You would also need to make sure that on average the engine was operating under as close to the same conditions as possible. Even a few degrees difference in coolant or air temperature could affect the results. And ideally you would want to look at the standard deviation for each set of numbers as well to make sure the data sets have at least similar variability.

Rendon also mentioned boost. Boost is an entirely different game. None of these sensors (the ones I tested anyway) are designed for much more than standard atmospheric pressure (I presume) so once you get into that realm all bets are off. But now that I think about it I suppose I could test them again with positive pressures and see how they do.
In fact I think I'll do just that.

C|

[Rendon LX-i]

Thanks....Why is that with a stock map sensor i get CEL with the 4gen map i get nothing at 10psi. What do you me i need to get into REALM? Im not beening nagative with nothing. I just felt a difference with or without stats. BUT I like how you took your time to see how the differences are. Would like to see how you test em on positive pressure...I assuming should be good for the fact i dont CEL

[3rd Party]

One thing I should point out is that this still doesn't explain why Honda felt it necessary to change the MAP sensors. There is obviously *something* different about them otherwise why would they bother? It could be a number of things though; updated internal circuitry due to manufacturing changes, wider measurement range, better accuracy, reduced temperature sensitivity, better/different dynamic response, reduced electrical noise, etc. All of these things could apply and still not affect the static pressure vs. voltage curve.

As for dyno runs; (chassis) dynos aren't nearly as reliable as most people think they are. There are SO many variables that affect the numbers you will see on a dyno that it's very difficult to get consistent results. To really prove that a particular part has an effect you would need to make several runs with and without that part and average each set of results. You would also need to make sure that on average the engine was operating under as close to the same conditions as possible. Even a few degrees difference in coolant or air temperature could affect the results. And ideally you would want to look at the standard deviation for each set of numbers as well to make sure the data sets have at least similar variability.

Rendon also mentioned boost. Boost is an entirely different game. None of these sensors (the ones I tested anyway) are designed for much more than standard atmospheric pressure (I presume) so once you get into that realm all bets are off. But now that I think about it I suppose I could test them again with positive pressures and see how they do.
In fact I think I'll do just that.

C|

i'm assuming they changed MAP sensors 'cause the motor was obviously changed, along with the ignition system. as we all know, 4th gen Accords are OBD1, and we're OBD0, so i'm assuming the MAP sensors were changed for the new OBD shindig?

[cygnus x-1]

Thanks....Why is that with a stock map sensor i get CEL with the 4gen map i get nothing at 10psi. What do you me i need to get into REALM? Im not beening nagative with nothing. I just felt a difference with or without stats. BUT I like how you took your time to see how the differences are. Would like to see how you test em on positive pressure...I assuming should be good for the fact i dont CEL


That's exactly what I'm saying. The test I just did was only with vacuum and doesn't apply under boost. Given what you've experienced I'm guessing that the sensors do act differently under boost. 10PSI happens to be +20inHg, which if the sensor curves are linear would be right at the edge of their maximum pressure capability.

It will be interesting to see what happens. I'll try it tonight.

C|

[mushroom_toy]

I already had it posted that these dont do anything in my dyno thread. lol.

[bobafett]

I don't recall the details, but that does correspond with real world experience and everything I have read up at hmt, if you run more than 10-11psi, you must use either check valves to limit that amount of boost that your map sensor sees, or you need to upgrade to a 2,3, or 4 bar map sensor so that it can read boost and still give an accurate reading.

[Nafs Asdf]

Wouldn't a 1 bar sensor read up to 14.5 psi?

[Hauntd ca3]

a 1 bar sensor will read up to 1 bar absolute which is atmo
need a 2 bar sensor to read up to 14.5 psi boost which is 2 bar absolute

[AccordB20A]

the model changes could be cause of anything, the biggest thing i can think of is one has wires coming out, one had a plug directly on it and the other mounts to the TB but there could be a lot of other revisions too.

[lostforawhile]

i'm assuming they changed MAP sensors 'cause the motor was obviously changed, along with the ignition system. as we all know, 4th gen Accords are OBD1, and we're OBD0, so i'm assuming the MAP sensors were changed for the new OBD shindig?we are actually obd1, except for the carbs, the FI had diagnostics and trouble codes and a check engine light. even the prelude dual carb models were obd1, they had onboard diagnostics, they even had a mass airflow sensor on some of them and a map sensor.

[stat1K]

it may have had the diagnostics but it's not obd1 lost, not at all.

['A20A3']

hmmm. maybe i'm crazy. but i know it made my top-end better...

[labeledsk8r]

its good to see this myth finly put to bed, i know mush-toy did it on the dyno and no one seemed to listen to his results, but these ones are for sure proven under the same conditions, nice job. mayby now the "4th gen map sensor" mod will be removed from alot of peoples mod lists. will be interesting to see what happens under boost though





we are actually obd1, except for the carbs, the FI had diagnostics and trouble codes and a check engine light. even the prelude dual carb models were obd1, they had onboard diagnostics, they even had a mass airflow sensor on some of them and a map sensor.

LAWL

['A20A3']

we are actually obd1, except for the carbs, the FI had diagnostics and trouble codes and a check engine light.


uhhh. no. sorry.

k thanx bye.

[lostforawhile]

uhhh. no. sorry.

k thanx bye.well oficially thats what it's considered. anything from the early days of trouble codes all the way to obd2 is actually considered obd1, i guess people here just call it a different name. I don't have a computer anyway so it really doesn't matter to me. the EPA considers it on board diagnostics and on board diagnostics 2

[lostforawhile]

there was actually a primitive code reader available to the dealers that plugged in somewhere in the harness. not sure where but i remember seeing them in use. it does exist. or did.

[Rendon LX-i]

shit if we were obd1 bro we would be tunning like a mofo. we cant chip the pjo for shit.

[lostforawhile]

shit if we were obd1 bro we would be tunning like a mofo. we cant chip the pjo for shit.

early obd1, they got better the first stuff was useless. not trying to start an arguement on here and derail the thread, it's all considered the first generation of on board diagnostics, but it was really primitive in the old days,i guess what everyone means is they are upgrading to later obd1.I should have clarified.

[Rendon LX-i]

Now your catching on. Reason we want obd1 for tuning and just better overall performance not only engine idle, When you just get in a obd1 car its so smooth.

[lostforawhile]

Now your catching on. Reason we want obd1 for tuning and just better overall performance not only engine idle, When you just get in a obd1 car its so smooth.
well when honda was building a prelude with dual carbs,500 vaccume lines, a mass airflow sensor, and a map sensor all combined together,with a computer with a check engine light and trouble codes, you know their early shit was fucked up. they couldn't fix the problems on those when they were new.

[Civic Accord Honda]

good job!
u need a myth busters busted sign now




we are actually obd1, except for the carbs, the FI had diagnostics and trouble codes and a check engine light. even the prelude dual carb models were obd1, they had onboard diagnostics, they even had a mass airflow sensor on some of them and a map sensor.
LOL!!! its OBD-0....:flash:

[lostforawhile]

good job!
u need a myth busters busted sign now




LOL!!! its OBD-0....:flash:no obd 0 is the carbs, if it has trouble codes it's all obd1 just a very early version of it. just like a modern pc compared to a radio shack trs80, they are both computers,one is useful,but one is good for nothing now but a bookend.

[3rd Party]

no obd 0 is the carbs, if it has trouble codes it's all obd1 just a very early version of it. just like a modern pc compared to a radio shack trs80, they are both computers,one is useful,but one is good for nothing now but a bookend.

the PGM-FI Accords are also OBD-0 for the pre-'91 you can't hook up any sort of diagnostic equipment to the OBD-0 cars...all you can do is watch the blinking lights, ;)

[labeledsk8r]

the PGM-FI Accords are also OBD-0 for the pre-'91 you can't hook up any sort of diagnostic equipment to the OBD-0 cars...all you can do is watch the blinking lights, ;)

ding ding ding we have a winner, lost is sorta right it is a on board diagnostic, but theres no interface, obd-1 had/have a way to comunicate with other computers obd-0 only had the blinky code thrower. now lets get off this topic becuse were horeing up a good thread with good info

were are those numbers on boosted maps

[Rendon LX-i]

I just gave up....LOST where obd0 lol....Im not beening a dick dude but listen lol. AND SORRY FOR |WHORING UP THE THREAD.

[3rd Party]

I just gave up....LOST where obd0 lol....Im not beening a dick dude but listen lol. AND SORRY FOR |WHORING UP THE THREAD.

i was merely saying that the cars restricted to OBD-0 weren't just carb'd motors, but also PGM-FI motors as well.

[cygnus x-1]

Part 2.

Here are the measurements when done with positive pressure instead of vacuum:



Pressure (inHg) PS10 (V) PS20 (V) PS22 (V)
---------------------------------------------------------
0 2.84 2.83 2.83
5.9 3.74 3.74 3.74
11.8 4.24 4.23 4.24
17.7 4.70 4.74 4.70
The pressure values are odd because the gauge I was using was marked in bars so I converted to inHg to keep things consistent. The highest pressure is equivalent to 0.6 bars or 8.7 PSI. Interestingly, all 3 sensors topped out at 0.6 bar, and would not read any higher than about 4.7V.

So I have no explanation for what Rendon reported about the CEL coming on with the 3g sensor and not the 4g. There is one other thing I can check but it's a long shot.

EDIT: The other thing I wanted to check was the output impedance of the sensors. You could think of it kind of like the ohm range that an amplifier is rated for. If you connect a speaker with too low of an impedance to an amp that isn't designed for it, the amp can become unstable and the sound from the speakers will be distorted. So I checked the voltage output of three sensors with a few different resistor values connected across the outputs and ground. They all behaved the same so they have similar output impedances.


C|

[A18A]

mind over matter?

[cygnus x-1]

mind over matter?


The only thing I can figure is that for whatever reason the two sensors that Rendon has, have maximum output voltages that are different enough that one fools the ECU and the other doesn't. Without looking at the voltages of his sensors I can't really say anything else.

C|

[thegreatdane]

At long last! That stuff was rediculous.