Need your input on replacement IAT packaging
Thread Starter
* B A N N E D *
Joined: Apr 2007
Posts: 848
Likes: 0
From: SoCal, CA
Re: (boostd92)
Lots of progress. I'm currently working on an LCD module that will display the iat temp. It has some cool functions, like a hold mode that captures the highest temp during a run. The module is also very small, about the size of a box of matches. Hopefully I can build these for cheap (shooting for <$100). The prototype parts are very expensive since I'm not buying in bulk. The LCD screen alone cost me $40.
As for the sensor element itself - progress is very slow. I still haven't found a manufacturer to buy from in bulk who won't rip me off. Most quotes I'm getting are around $20/unit for 100+ units. I'm sorry, but I'm not going to eat a $2,000 up-front cost when I don't even know if these will sell.
As for the sensor element itself - progress is very slow. I still haven't found a manufacturer to buy from in bulk who won't rip me off. Most quotes I'm getting are around $20/unit for 100+ units. I'm sorry, but I'm not going to eat a $2,000 up-front cost when I don't even know if these will sell.
Thread Starter
* B A N N E D *
Joined: Apr 2007
Posts: 848
Likes: 0
From: SoCal, CA
Re: (Legion)
Change of plans guys. Since I want to implement a more accurate sensor than the Honda ECU can make use of, I'm going to scrap my progress and take a different route.
The problem arises from different makes/models/generations of cars all using different R/T curves with their IAT sensors. Spending thousands of dollars having a bunch of custom sensors made is not practical. The workaround is that I can use the same sensor for every year/make/model, but I'll have to pass the output signal through a microcontroller to convert it into something the ECU will like.
The only real downside to this method is that the sensor will have a converter box inline between the sensor and the pug. It might be kind of ugly. Like a USB/TTL converter box.
The problem arises from different makes/models/generations of cars all using different R/T curves with their IAT sensors. Spending thousands of dollars having a bunch of custom sensors made is not practical. The workaround is that I can use the same sensor for every year/make/model, but I'll have to pass the output signal through a microcontroller to convert it into something the ECU will like.
The only real downside to this method is that the sensor will have a converter box inline between the sensor and the pug. It might be kind of ugly. Like a USB/TTL converter box.
i HAS questions ?
Joined: Feb 2003
Posts: 7,850
Likes: 0
From: OH
Re: (Legion)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">Change of plans guys. Since I want to implement a more accurate sensor than the Honda ECU can make use of, I'm going to scrap my progress and take a different route.
The problem arises from different makes/models/generations of cars all using different R/T curves with their IAT sensors. Spending thousands of dollars having a bunch of custom sensors made is not practical. The workaround is that I can use the same sensor for every year/make/model, but I'll have to pass the output signal through a microcontroller to convert it into something the ECU will like.
The only real downside to this method is that the sensor will have a converter box inline between the sensor and the pug. It might be kind of ugly. Like a USB/TTL converter box.</TD></TR></TABLE>
Makes more sense to do it that way. Why limit this to just Hondas?
The problem arises from different makes/models/generations of cars all using different R/T curves with their IAT sensors. Spending thousands of dollars having a bunch of custom sensors made is not practical. The workaround is that I can use the same sensor for every year/make/model, but I'll have to pass the output signal through a microcontroller to convert it into something the ECU will like.
The only real downside to this method is that the sensor will have a converter box inline between the sensor and the pug. It might be kind of ugly. Like a USB/TTL converter box.</TD></TR></TABLE>
Makes more sense to do it that way. Why limit this to just Hondas?
Thread Starter
* B A N N E D *
Joined: Apr 2007
Posts: 848
Likes: 0
From: SoCal, CA
Re: (.dave)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by .dave »</TD></TR><TR><TD CLASS="quote">
Makes more sense to do it that way. Why limit this to just Hondas? </TD></TR></TABLE>
Especially since my main marketing focus is BMWs. See, they have the money, and they'll pay anything for this stuff. I actually hate BMWs though. Hondas are my passion, but Honda people are broke and vehemently oppose paying for R&D work.
Hopefully this new approach will satisfy both worlds!
Makes more sense to do it that way. Why limit this to just Hondas?
</TD></TR></TABLE>Especially since my main marketing focus is BMWs. See, they have the money, and they'll pay anything for this stuff. I actually hate BMWs though. Hondas are my passion, but Honda people are broke and vehemently oppose paying for R&D work.
Hopefully this new approach will satisfy both worlds!
Honda-Tech Member
Joined: Sep 2006
Posts: 5,856
Likes: 0
From: poopfacepartytime, ca, usa
Re: (Legion)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">
Especially since my main marketing focus is BMWs. See, they have the money, and they'll pay anything for this stuff. I actually hate BMWs though. Hondas are my passion, but Honda people are broke and vehemently oppose paying for R&D work.
Hopefully this new approach will satisfy both worlds!</TD></TR></TABLE>
honda customers are cheap effs, but DSM customers are the worse!! i really hope this works out for you and the people who will benefit from this
Especially since my main marketing focus is BMWs. See, they have the money, and they'll pay anything for this stuff. I actually hate BMWs though. Hondas are my passion, but Honda people are broke and vehemently oppose paying for R&D work.
Hopefully this new approach will satisfy both worlds!</TD></TR></TABLE>
honda customers are cheap effs, but DSM customers are the worse!! i really hope this works out for you and the people who will benefit from this
i HAS questions ?
Joined: Feb 2003
Posts: 7,850
Likes: 0
From: OH
Re: (Legion)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">
Especially since my main marketing focus is BMWs. See, they have the money, and they'll pay anything for this stuff. I actually hate BMWs though. Hondas are my passion, but Honda people are broke and vehemently oppose paying for R&D work.
Hopefully this new approach will satisfy both worlds!</TD></TR></TABLE>
So the BMW guys will pay for the R&D work?
I'm guessing scaling the output will be all software side?
Especially since my main marketing focus is BMWs. See, they have the money, and they'll pay anything for this stuff. I actually hate BMWs though. Hondas are my passion, but Honda people are broke and vehemently oppose paying for R&D work.
Hopefully this new approach will satisfy both worlds!</TD></TR></TABLE>
So the BMW guys will pay for the R&D work?
I'm guessing scaling the output will be all software side?
Honda-Tech Member
Joined: Jul 2006
Posts: 4,009
Likes: 0
From: missouri
Re: (.dave)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by .dave »</TD></TR><TR><TD CLASS="quote">
So the BMW guys will pay for the R&D work?
</TD></TR></TABLE>
yea, that's pretty nice of them!
So the BMW guys will pay for the R&D work?
</TD></TR></TABLE>
yea, that's pretty nice of them!
Thread Starter
* B A N N E D *
Joined: Apr 2007
Posts: 848
Likes: 0
From: SoCal, CA
Re: (.dave)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by .dave »</TD></TR><TR><TD CLASS="quote">So the BMW guys will pay for the R&D work? </TD></TR></TABLE>
Yep. They pay for anything. They don't care how it works or how long it takes to make. They just want to throw away their money. Literally. That stuff REALLY doesn't fly in the Honda crowd
Anyway, I have some good news! I've been talking all day with a new sensor company and we just might be going into business. My amazing new LCD module arrives in the mail tomorrow, so I'll start tearing into the programming side of that.
</TD></TR></TABLE>Yep. They pay for anything. They don't care how it works or how long it takes to make. They just want to throw away their money. Literally. That stuff REALLY doesn't fly in the Honda crowd
Anyway, I have some good news! I've been talking all day with a new sensor company and we just might be going into business. My amazing new LCD module arrives in the mail tomorrow, so I'll start tearing into the programming side of that.
Thread Starter
* B A N N E D *
Joined: Apr 2007
Posts: 848
Likes: 0
From: SoCal, CA
Re: (Legion)
Here's some stuff for you guys to chew on:
1. Honda IAT sensors flat out don't work on boosted motors. They just cannot respond fast enough. That's why some tuners have learned to mount them in the charge pipes, away from the intake manifold. This yields good results because the sensors are so slow that by the time the ECU reads the temperature in the charge pipes, that volume of air has already started to enter the combustion chamber.
2. You cannot use aftermarket sensors on your engine (except for the one I'm developing). People seem to think that they can swap in a GM or Bosch sensor and compensate for it with tuning software. Unfortunately, thermistors all have wildly different R/T curves, nominal resistance values, Beta constants, Alpha constants, resistance ratios, etc. etc.. There are dozens of mathematical values that differ from thermistor to thermistor. These relationships are all exponentially and logarithmically related. To accurately swap in a GM or Bosch sensor, you'd need a microcontroller to intercept the IAT signal, then do lots of complicated math before outputting the signal to the ECU. I have no idea how you guys think you are using your tuning software to compensate for these sensors, but I can tell you right now - it isn't working how you think it is. Just because someone added a few buttons or sliders into their software doesn't mean they can magically circumnavigate the physical/thermal properties of the IAT.
3. Why am I building an LCD screen? Well, the OEM Honda IAT sensor has a large tolerance built in, which means the ECU is prepared for this tolerance as well. So the replacement sensors I'm working on are very accurate, but it doesn't matter on the ECU side because the ECU can't make use of the added precision. My LCD module, however, is programmed by me, and makes perfect use of the IAT precision. Plus, I don't want to carry a laptop around and have to datalog if I feel like monitoring my IATs on a hot day. I can just set the module in "hold" mode, make a run, and see if my IATs are getting out of control.
4. How important is an accurate IAT sensor? Well, that depends on what you consider important. Is tuning important? I mean, you can just throw an FMU on your car and call it a day. But that's sloppy. As an engineer, my goal in everything I do is to achieve precision. A fast-response IAT sensor will allow your engine to actually adjust fuel delivery based on what your ECU is telling it to. Is that necessary? Again, what do you consider necessary? It's as necessary as setting a proper injector dead time, but most people don't do that (or even know what that is for that matter). The main goal of tuning is to get the engine to inject exactly as much fuel as you want it to under various conditions. This sensor will allow you to get that much closer to achieving perfect fuel maps. After all, your car doesn't spend it's entire life operating at the exact temperature it was tuned at.
1. Honda IAT sensors flat out don't work on boosted motors. They just cannot respond fast enough. That's why some tuners have learned to mount them in the charge pipes, away from the intake manifold. This yields good results because the sensors are so slow that by the time the ECU reads the temperature in the charge pipes, that volume of air has already started to enter the combustion chamber.
2. You cannot use aftermarket sensors on your engine (except for the one I'm developing). People seem to think that they can swap in a GM or Bosch sensor and compensate for it with tuning software. Unfortunately, thermistors all have wildly different R/T curves, nominal resistance values, Beta constants, Alpha constants, resistance ratios, etc. etc.. There are dozens of mathematical values that differ from thermistor to thermistor. These relationships are all exponentially and logarithmically related. To accurately swap in a GM or Bosch sensor, you'd need a microcontroller to intercept the IAT signal, then do lots of complicated math before outputting the signal to the ECU. I have no idea how you guys think you are using your tuning software to compensate for these sensors, but I can tell you right now - it isn't working how you think it is. Just because someone added a few buttons or sliders into their software doesn't mean they can magically circumnavigate the physical/thermal properties of the IAT.
3. Why am I building an LCD screen? Well, the OEM Honda IAT sensor has a large tolerance built in, which means the ECU is prepared for this tolerance as well. So the replacement sensors I'm working on are very accurate, but it doesn't matter on the ECU side because the ECU can't make use of the added precision. My LCD module, however, is programmed by me, and makes perfect use of the IAT precision. Plus, I don't want to carry a laptop around and have to datalog if I feel like monitoring my IATs on a hot day. I can just set the module in "hold" mode, make a run, and see if my IATs are getting out of control.
4. How important is an accurate IAT sensor? Well, that depends on what you consider important. Is tuning important? I mean, you can just throw an FMU on your car and call it a day. But that's sloppy. As an engineer, my goal in everything I do is to achieve precision. A fast-response IAT sensor will allow your engine to actually adjust fuel delivery based on what your ECU is telling it to. Is that necessary? Again, what do you consider necessary? It's as necessary as setting a proper injector dead time, but most people don't do that (or even know what that is for that matter). The main goal of tuning is to get the engine to inject exactly as much fuel as you want it to under various conditions. This sensor will allow you to get that much closer to achieving perfect fuel maps. After all, your car doesn't spend it's entire life operating at the exact temperature it was tuned at.
Honda-Tech Member
Joined: Jul 2006
Posts: 4,009
Likes: 0
From: missouri
Re: (Legion)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">
2. You cannot use aftermarket sensors on your engine (except for the one I'm developing). People seem to think that they can swap in a GM or Bosch sensor and compensate for it with tuning software. Unfortunately, thermistors all have wildly different R/T curves, nominal resistance values, Beta constants, Alpha constants, resistance ratios, etc. etc.. There are dozens of mathematical values that differ from thermistor to thermistor. These relationships are all exponentially and logarithmically related. To accurately swap in a GM or Bosch sensor, you'd need a microcontroller to intercept the IAT signal, then do lots of complicated math before outputting the signal to the ECU. I have no idea how you guys think you are using your tuning software to compensate for these sensors, but I can tell you right now - it isn't working how you think it is. Just because someone added a few buttons or sliders into their software doesn't mean they can magically circumnavigate the physical/thermal properties of the IAT.</TD></TR></TABLE>
are you talking about aem's standalone ? because that is the only one you can use a gm iat sensor with and only people who use aem have used a gm iat sensor. if youre thinking people with a honda ecu are using a gm iat sensor then you are wrong and just wrote all that for nothing.
go look at aem's software..
2. You cannot use aftermarket sensors on your engine (except for the one I'm developing). People seem to think that they can swap in a GM or Bosch sensor and compensate for it with tuning software. Unfortunately, thermistors all have wildly different R/T curves, nominal resistance values, Beta constants, Alpha constants, resistance ratios, etc. etc.. There are dozens of mathematical values that differ from thermistor to thermistor. These relationships are all exponentially and logarithmically related. To accurately swap in a GM or Bosch sensor, you'd need a microcontroller to intercept the IAT signal, then do lots of complicated math before outputting the signal to the ECU. I have no idea how you guys think you are using your tuning software to compensate for these sensors, but I can tell you right now - it isn't working how you think it is. Just because someone added a few buttons or sliders into their software doesn't mean they can magically circumnavigate the physical/thermal properties of the IAT.</TD></TR></TABLE>
are you talking about aem's standalone ? because that is the only one you can use a gm iat sensor with and only people who use aem have used a gm iat sensor. if youre thinking people with a honda ecu are using a gm iat sensor then you are wrong and just wrote all that for nothing.
go look at aem's software..
Thread Starter
* B A N N E D *
Joined: Apr 2007
Posts: 848
Likes: 0
From: SoCal, CA
Re: (mike93eh2)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by mike93eh2 »</TD></TR><TR><TD CLASS="quote">
are you talking about aem's standalone ? because that is the only one you can use a gm iat sensor with and only people who use aem have used a gm iat sensor. if youre thinking people with a honda ecu are using a gm iat sensor then you are wrong and just wrote all that for nothing. go look at aem's software.. </TD></TR></TABLE>
I'm not talking about standalones. I'm talking about cracked Honda ECUs. People keep telling me that their software can compensate for GM sensors. I'm dispelling that myth.
are you talking about aem's standalone ? because that is the only one you can use a gm iat sensor with and only people who use aem have used a gm iat sensor. if youre thinking people with a honda ecu are using a gm iat sensor then you are wrong and just wrote all that for nothing.
go look at aem's software.. </TD></TR></TABLE>I'm not talking about standalones. I'm talking about cracked Honda ECUs. People keep telling me that their software can compensate for GM sensors. I'm dispelling that myth.
Thread
Thread Starter
Forum
Replies
Last Post
Brugalontherocks
Honda Civic / Del Sol (1992 - 2000)
7
Apr 28, 2008 08:10 PM
