LOST JEEPS :: View topic

I don't know what you are seeing when you are checking MAF signal, but the MAF only samples a portion of incoming air. That portion is calibrated to represent the total volume of intake air.
The calibration factor is programmed into the ECU.
All the MAF does is send a voltage signal to the ECU.

Have you taken your altitude? That factor could be what's messing you up.

The higher you go the less dense the air is,regardless of turbo or not your always sucking in(or boosting less) air at 6500' then sea level and the MAF(MAP for us gasser KJ's) will reflect that in what they are sending the PCM.

If you use the DRBIII to read the MAF it only generates a voltage reading, and never reports what the flow actually is.

This is in contrast to nearly ever other sensor, the analog (typically 0-5v) voltage is displayed, along with the engineering units. The ECU analog input has a calibration table in it relative to the sensor type, however since the DRBIII cannot display the engineer units it makes me wonder if the ECU actually has this table, and if it just looks at a raw voltage level to determine EGR control.

I think you should look at the MAP value, which is used for engine control, and then calculate flow off of that, IAT is an available data point(twice actually). If you can grab the data for CSV and do the math you should be good.

Regardless of what the flow is some map in the ECU(boost limiter?) should contain the surge map, the ecu should be keeping the turbo out of the surge region.

What I posted on another thread:
I think you are correct and it (MAP) uses IAT to determine the volume of air entering the motor.
Found this in the FSM:
OPERATION
The intake air temperature sensor is a negative temperature coefficient (NTC) thermistor (resistance varies inversely with temperature). This means at cold air temperature its resistance is high, so the voltage signal will be high. As intake air temperature increases, sensor resistance decreases and the signal voltage will be low. This allows the sensor to provide an analog voltage signal (0.2-4.8 volts) to the ECM.

When the intake manifold pressure is low sensor voltage output is 0.25-1.8 volts at the ECM. When the intake manifold pressure is high due to turbo boost, sensor voltage output is 2.0-4.7 volts. The sensor receives a 5-volts reference from the ECM. Sensor ground is also provided by the ECM. The ECM uses boost pressure combined with intake air temperature to determine the volume of air entering the engine.

WWDiesel wrote:

Correct, you need the intake air temp to calculate flow, there is an IAT sensor on the airbox, the outside air temp(used for EVIC display and some other stuff I think) and then a "manifold air temp" that I think also from the MAP.

[
Would IAT and boost tell you flow though?

Sir Sam wrote:

WWDiesel wrote:

I can see the manifold temps with torque pro but not the air box temps. I really need to just buy a drb.

Sent from my Nexus 6 using Tapatalk

mass-hole wrote:

Would IAT and boost tell you flow though?

YES.
These two are absolutely critical for flow calculation.
So yes, suction temperature(IAT) is very important.

Wish we had a way to capture and chart all the data points and learn the relationships, bet we could learn a lot!!!!

WWDiesel wrote:

Wish we had a way to capture and chart all the data points and learn the relationships, bet we could learn a lot!!!!

I bet you could rig up some data IO with a raspberry pi.

Right now we have compressor suction temp, and manifold intake temp, but we do not have compressor discharge temp, or compressor discharge pressure.

Heck if you wanted to frankenstein something together get a couple of extra sensors and an ecu and shove them on the engine and read it off a second obdii port.......

Which actually gives me an idea, I have several SI ECUs that my company makes that have a ton of IO on them.

Get a couple of cheap GM map sensors to monitor compressor inlet pressure/temp, discharge pressure/temp, intercooler discharge pressure/temp, and then I can pull the data onto a laptop serially, or off the ECU with an OBDII port.

Would make an interesting bit of data.

WWDiesel wrote:

Wish we had a way to capture and chart all the data points and learn the relationships, bet we could learn a lot!!!!

I thought i remember kieth telling me it was a pressure sensor to sense when the air filter was plugged. It never got programmed so to speak so it just sits there pretty much doing nothing. That was 2 years ago tho so i wouldnt rely on that being 100% accurate...

Author:	mass-hole [ Mon Aug 01, 2016 12:16 pm ]
Post subject:	MAF accuracy?
Does anyone know how accurate our MAF sensors really are? I was trying to use it to help me tune a little bit and when I plotted the data on a compressor map, the flow rates were quite a bit lower than I expected. Peak flow rates are 20 lbs/min with a pressure ratio of 3:1. Granted, I am running more boost and am at 6500', but the turbo spent a good amount of its time to the left of the surge line in the map below. Is this map a close match to our compressor? It appears to be based on what I know about the turbo. The blue and orange lines are two datalogs i took and the red dots are my theoretical values for the stock tune. either the Volumetric efficiency is much worse than I thought or the MAF reads low. I would think if the VE was poor it would only show at high rpms. One thing I plan on doing is bypassing the turbo to see if I can determine the VE for the engine through the entire rpm range.

Author:	flash7210 [ Mon Aug 01, 2016 6:57 pm ]
Post subject:	Re: MAF accuracy?
I don't know what you are seeing when you are checking MAF signal, but the MAF only samples a portion of incoming air. That portion is calibrated to represent the total volume of intake air. The calibration factor is programmed into the ECU. All the MAF does is send a voltage signal to the ECU.

Author:	mass-hole [ Mon Aug 01, 2016 7:34 pm ]
Post subject:	Re: MAF accuracy?
flash7210 wrote: I don't know what you are seeing when you are checking MAF signal, but the MAF only samples a portion of incoming air. That portion is calibrated to represent the total volume of intake air. The calibration factor is programmed into the ECU. All the MAF does is send a voltage signal to the ECU. In torque the maf is output in grams per second so it was a fairly easy conversion to lbs/min. The values I get are not ridiculously far-fetched but seem a bit low. Sent from my Nexus 6 using Tapatalk

Author:	tjkj2002 [ Mon Aug 01, 2016 8:58 pm ]
Post subject:	Re: MAF accuracy?
Have you taken your altitude? That factor could be what's messing you up. The higher you go the less dense the air is,regardless of turbo or not your always sucking in(or boosting less) air at 6500' then sea level and the MAF(MAP for us gasser KJ's) will reflect that in what they are sending the PCM.

Author:	Sir Sam [ Mon Aug 01, 2016 9:17 pm ]
Post subject:	Re: MAF accuracy?
If you use the DRBIII to read the MAF it only generates a voltage reading, and never reports what the flow actually is. This is in contrast to nearly ever other sensor, the analog (typically 0-5v) voltage is displayed, along with the engineering units. The ECU analog input has a calibration table in it relative to the sensor type, however since the DRBIII cannot display the engineer units it makes me wonder if the ECU actually has this table, and if it just looks at a raw voltage level to determine EGR control. I think you should look at the MAP value, which is used for engine control, and then calculate flow off of that, IAT is an available data point(twice actually). If you can grab the data for CSV and do the math you should be good. Regardless of what the flow is some map in the ECU(boost limiter?) should contain the surge map, the ecu should be keeping the turbo out of the surge region.

LOST JEEPS http://www.lostjeeps.com/forum/phpBB3/

MAF accuracy? http://www.lostjeeps.com/forum/phpBB3/viewtopic.php?f=5&t=85046	Page 1 of 2

Author:	Sir Sam [ Mon Aug 01, 2016 9:28 pm ]
Post subject:	Re: MAF accuracy?
What about some other ways to verify the flow data? old school inches of water reading? some other calibrate flow measurement device?

Author:	mass-hole [ Tue Aug 02, 2016 7:45 pm ]
Post subject:	Re: MAF accuracy?
tjkj2002 wrote: Have you taken your altitude? That factor could be what's messing you up. The higher you go the less dense the air is,regardless of turbo or not your always sucking in(or boosting less) air at 6500' then sea level and the MAF(MAP for us gasser KJ's) will reflect that in what they are sending the PCM. This could definitely be it but I am running more boost than the stock tune does at sea level. Sir Sam wrote: If you use the DRBIII to read the MAF it only generates a voltage reading, and never reports what the flow actually is. This is in contrast to nearly ever other sensor, the analog (typically 0-5v) voltage is displayed, along with the engineering units. The ECU analog input has a calibration table in it relative to the sensor type, however since the DRBIII cannot display the engineer units it makes me wonder if the ECU actually has this table, and if it just looks at a raw voltage level to determine EGR control. I think you should look at the MAP value, which is used for engine control, and then calculate flow off of that, IAT is an available data point(twice actually). If you can grab the data for CSV and do the math you should be good. Regardless of what the flow is some map in the ECU(boost limiter?) should contain the surge map, the ecu should be keeping the turbo out of the surge region. I believe the ecu does use flow rate in mg/second based on the maps that control the egr/fcv. Would IAT and boost tell you flow though? I mean you could estimate but it doesn't really tell you how fast it's moving through the motor. If you had one motor that flows well and one that doesn't, you could have the same boost but one would move more air. Sent from my Nexus 6 using Tapatalk

Author:	WWDiesel [ Tue Aug 02, 2016 8:47 pm ]
Post subject:	Re: MAF accuracy?
What I posted on another thread: I think you are correct and it (MAP) uses IAT to determine the volume of air entering the motor. Found this in the FSM: OPERATION The intake air temperature sensor is a negative temperature coefficient (NTC) thermistor (resistance varies inversely with temperature). This means at cold air temperature its resistance is high, so the voltage signal will be high. As intake air temperature increases, sensor resistance decreases and the signal voltage will be low. This allows the sensor to provide an analog voltage signal (0.2-4.8 volts) to the ECM. When the intake manifold pressure is low sensor voltage output is 0.25-1.8 volts at the ECM. When the intake manifold pressure is high due to turbo boost, sensor voltage output is 2.0-4.7 volts. The sensor receives a 5-volts reference from the ECM. Sensor ground is also provided by the ECM. The ECM uses boost pressure combined with intake air temperature to determine the volume of air entering the engine.

Author:	Sir Sam [ Tue Aug 02, 2016 9:02 pm ]
Post subject:	Re: MAF accuracy?
WWDiesel wrote: What I posted on another thread: I think you are correct and it (MAP) uses IAT to determine the volume of air entering the motor. Found this in the FSM: OPERATION The intake air temperature sensor is a negative temperature coefficient (NTC) thermistor (resistance varies inversely with temperature). This means at cold air temperature its resistance is high, so the voltage signal will be high. As intake air temperature increases, sensor resistance decreases and the signal voltage will be low. This allows the sensor to provide an analog voltage signal (0.2-4.8 volts) to the ECM. When the intake manifold pressure is low sensor voltage output is 0.25-1.8 volts at the ECM. When the intake manifold pressure is high due to turbo boost, sensor voltage output is 2.0-4.7 volts. The sensor receives a 5-volts reference from the ECM. Sensor ground is also provided by the ECM. The ECM uses boost pressure combined with intake air temperature to determine the volume of air entering the engine. Correct, you need the intake air temp to calculate flow, there is an IAT sensor on the airbox, the outside air temp(used for EVIC display and some other stuff I think) and then a "manifold air temp" that I think also from the MAP.

Author:	mass-hole [ Tue Aug 02, 2016 9:04 pm ]
Post subject:	Re: MAF accuracy?
Sir Sam wrote: WWDiesel wrote: What I posted on another thread: I think you are correct and it (MAP) uses IAT to determine the volume of air entering the motor. Found this in the FSM: OPERATION The intake air temperature sensor is a negative temperature coefficient (NTC) thermistor (resistance varies inversely with temperature). This means at cold air temperature its resistance is high, so the voltage signal will be high. As intake air temperature increases, sensor resistance decreases and the signal voltage will be low. This allows the sensor to provide an analog voltage signal (0.2-4.8 volts) to the ECM. When the intake manifold pressure is low sensor voltage output is 0.25-1.8 volts at the ECM. When the intake manifold pressure is high due to turbo boost, sensor voltage output is 2.0-4.7 volts. The sensor receives a 5-volts reference from the ECM. Sensor ground is also provided by the ECM. The ECM uses boost pressure combined with intake air temperature to determine the volume of air entering the engine. Correct, you need the intake air temp to calculate flow, there is an IAT sensor on the airbox, the outside air temp(used for EVIC display and some other stuff I think) and then a "manifold air temp" that I think also from the MAP. I can see the manifold temps with torque pro but not the air box temps. I really need to just buy a drb. But again, I don't see how you can calculate flow with just boost and IAT and not know the volumetric efficiency of the motor. If my engine had performance cams and ported heads(and my manifold didn't have a half inch of soot lining the walls) I would flow more at the same boost. Sent from my Nexus 6 using Tapatalk

Author:	Sir Sam [ Tue Aug 02, 2016 9:07 pm ]
Post subject:	Re: MAF accuracy?
mass-hole wrote: [ Would IAT and boost tell you flow though? YES. These two are absolutely critical for flow calculation. It's funny that you posted about your compressor being in the surge region yesterday, I spent the day yesterday trying to explain that a certain piece of rotating equipment was not crossing the surge line. One of the issues that comes up is the suction temperature, in our case suction temp is staying steady until hot gas is introduced from the drier, like if you had a very large dose of hot air get sucked into the engine. So yes, suction temperature(IAT) is very important.

Author:	Sir Sam [ Tue Aug 02, 2016 9:14 pm ]
Post subject:	Re: MAF accuracy?
mass-hole wrote: Sir Sam wrote: WWDiesel wrote: What I posted on another thread: I think you are correct and it (MAP) uses IAT to determine the volume of air entering the motor. Found this in the FSM: OPERATION The intake air temperature sensor is a negative temperature coefficient (NTC) thermistor (resistance varies inversely with temperature). This means at cold air temperature its resistance is high, so the voltage signal will be high. As intake air temperature increases, sensor resistance decreases and the signal voltage will be low. This allows the sensor to provide an analog voltage signal (0.2-4.8 volts) to the ECM. When the intake manifold pressure is low sensor voltage output is 0.25-1.8 volts at the ECM. When the intake manifold pressure is high due to turbo boost, sensor voltage output is 2.0-4.7 volts. The sensor receives a 5-volts reference from the ECM. Sensor ground is also provided by the ECM. The ECM uses boost pressure combined with intake air temperature to determine the volume of air entering the engine. Correct, you need the intake air temp to calculate flow, there is an IAT sensor on the airbox, the outside air temp(used for EVIC display and some other stuff I think) and then a "manifold air temp" that I think also from the MAP. I can see the manifold temps with torque pro but not the air box temps. I really need to just buy a drb. Sent from my Nexus 6 using Tapatalk Ya, it would be good if we had some time to compare the DRBIII readings with the torque pro readings. You should have an intake air temp from the airbox(the mercedes labeled sensor that gave geordi so much trouble years back), aka suction temp. We have a manifold temp, which is downstream of the intercooler. There is some lag in the downstream reading because of the pressure drop across the intercooler, and the "spongyness" of it. A thought occurs to me, since you are not reading boost pressure at compressor discharge and temperature at the same point(before or after the compressor), you will have some uncertainly because of the intercooler and piping "spgonyness"

Author:	WWDiesel [ Wed Aug 03, 2016 11:31 am ]
Post subject:	Re: MAF accuracy?
Sir Sam wrote: mass-hole wrote: Would IAT and boost tell you flow though? YES. These two are absolutely critical for flow calculation. So yes, suction temperature(IAT) is very important. Oh yes, you are totally correct! The temperature of the incoming air can have a direct effect on the combustion temperatures. The cooler and more dense the air coming into the engine the more oxygen it will contain which relates to better combustion and more complete burning of the fuel and in turn more power for a given amount of fuel.... My engine always seem to have a little more power on cooler winter days than it does on the hot high 90 days of summer! The CAC cools the incoming combustion air after being exposed to the compressor wheel to try and offset the temperature rise from the air being compressed. I suspect there is a linear relationship between the incoming suction air temperature and the combustion air entering the motor at a given boost level. Of course the relationship would get nonlinear as boost is increased or decreased across the operating range of the engine due to the heat generated from the air being compressed... Wish we had a way to capture and chart all the data points and learn the relationships, bet we could learn a lot!!!!

Author:	Sir Sam [ Wed Aug 03, 2016 2:59 pm ]
Post subject:	Re: MAF accuracy?
WWDiesel wrote: Wish we had a way to capture and chart all the data points and learn the relationships, bet we could learn a lot!!!! I bet you could rig up some data IO with a raspberry pi. Right now we have compressor suction temp, and manifold intake temp, but we do not have compressor discharge temp, or compressor discharge pressure. Heck if you wanted to frankenstein something together get a couple of extra sensors and an ecu and shove them on the engine and read it off a second obdii port....... Which actually gives me an idea, I have several SI ECUs that my company makes that have a ton of IO on them. Get a couple of cheap GM map sensors to monitor compressor inlet pressure/temp, discharge pressure/temp, intercooler discharge pressure/temp, and then I can pull the data onto a laptop serially, or off the ECU with an OBDII port. Would make an interesting bit of data.

Page 1 of 2	All times are UTC - 5 hours [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/

Author:	mass-hole [ Wed Aug 03, 2016 5:06 pm ]
Post subject:	Re: MAF accuracy?
Sir Sam wrote: WWDiesel wrote: Wish we had a way to capture and chart all the data points and learn the relationships, bet we could learn a lot!!!! I bet you could rig up some data IO with a raspberry pi. Right now we have compressor suction temp, and manifold intake temp, but we do not have compressor discharge temp, or compressor discharge pressure. Heck if you wanted to frankenstein something together get a couple of extra sensors and an ecu and shove them on the engine and read it off a second obdii port....... Which actually gives me an idea, I have several SI ECUs that my company makes that have a ton of IO on them. Get a couple of cheap GM map sensors to monitor compressor inlet pressure/temp, discharge pressure/temp, intercooler discharge pressure/temp, and then I can pull the data onto a laptop serially, or off the ECU with an OBDII port. Would make an interesting bit of data. Raspberry pi is a good idea. It outputs 5v doesn't it, so it could run out tmap sensors without and mods. Sent from my Nexus 6 using Tapatalk

Author:	Sir Sam [ Wed Aug 03, 2016 5:09 pm ]
Post subject:	Re: MAF accuracy?
Yes, it will output 5v, but a 5v regulator would also be easy to have standalone, then just route the signal to the pi. Some thermistors could also be used, though I swear the map sensor has temp as well.

Author:	diesel_guy86 [ Wed Aug 03, 2016 5:31 pm ]
Post subject:	Re: MAF accuracy?
Sir Sam wrote: WWDiesel wrote: Wish we had a way to capture and chart all the data points and learn the relationships, bet we could learn a lot!!!! I bet you could rig up some data IO with a raspberry pi. Right now we have compressor suction temp, and manifold intake temp, but we do not have compressor discharge temp, or compressor discharge pressure. Heck if you wanted to frankenstein something together get a couple of extra sensors and an ecu and shove them on the engine and read it off a second obdii port....... Which actually gives me an idea, I have several SI ECUs that my company makes that have a ton of IO on them. Get a couple of cheap GM map sensors to monitor compressor inlet pressure/temp, discharge pressure/temp, intercooler discharge pressure/temp, and then I can pull the data onto a laptop serially, or off the ECU with an OBDII port. Would make an interesting bit of data. This would be awesome to test my new turbo with, exhaust temp/pressure would be nice to data log, as well as vane signal too.

Author:	flash7210 [ Wed Aug 03, 2016 5:33 pm ]
Post subject:	Re: MAF accuracy?
I thought that little sensor on the airbox lid was the inlet pressure sensor. Is it really temp sensor? And yes, the MAP sensor is also your IAT sensor.

Author:	diesel_guy86 [ Wed Aug 03, 2016 6:18 pm ]
Post subject:	Re: MAF accuracy?
I thought i remember kieth telling me it was a pressure sensor to sense when the air filter was plugged. It never got programmed so to speak so it just sits there pretty much doing nothing. That was 2 years ago tho so i wouldnt rely on that being 100% accurate...

Author:	flash7210 [ Wed Aug 03, 2016 7:30 pm ]
Post subject:	Re: MAF accuracy?
diesel_guy86 wrote: I thought i remember kieth telling me it was a pressure sensor to sense when the air filter was plugged. It never got programmed so to speak so it just sits there pretty much doing nothing. That was 2 years ago tho so i wouldnt rely on that being 100% accurate... I'm pretty sure that sensor is disabled in all GDE tunes. But if it does nothing, why does a stock tune go into a kind of limp mode when its unplugged? Which brings up another point... On all the GDE tunes, that airbox sensor is disabled and so is the MAF. So the only sensor used for intake air is the MAP sensor, which is also the IAT sensor. The GDE tunes seem to work just fine like this and provide good power and fuel economy.