Datalogging: Where to find 0-5V linear intake temp sensor?
03-09-2008, 09:04 PM
#1
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
Datalogging: Where to find 0-5V linear intake temp sensor?
I'm using a Zeitronix wideband, and it has the option to datalog one user input. It has to be 0-5V.
I'd like to monitor my intake temps to see how well my IC is doing after a few runs, and to see how the meth kit is cooling them down.
Does anyone datalog with an IAT that is linear and 0-5V? I need help finding an appropriate one. It looks like the 350Z uses one, but I'm not sure it's linear.
Thanks!
-Adam
I'd like to monitor my intake temps to see how well my IC is doing after a few runs, and to see how the meth kit is cooling them down.
Does anyone datalog with an IAT that is linear and 0-5V? I need help finding an appropriate one. It looks like the 350Z uses one, but I'm not sure it's linear.
Thanks!
-Adam
03-10-2008, 12:07 AM
#2
Re: Datalogging: Where to find 0-5V linear intake temp sensor? (boostd92)
I was thinking of doing the same thing but , how would you translate the voltage into temperature?
the user input only reads out in volts. I wrote down the pin on the ecu that needs to be connected, but never tested it, I might do it later.
and let you know.
the user input only reads out in volts. I wrote down the pin on the ecu that needs to be connected, but never tested it, I might do it later.
and let you know.
03-10-2008, 06:24 AM
#4
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
They absolutely can be:
http://www.plxdevices.com/products/sm/ait/
That particular one doesn't read high enough for me, I'm looking for 200F at least, but less than 250F.
Anyone know of one like this that reads up to 200F?
http://www.plxdevices.com/products/sm/ait/
That particular one doesn't read high enough for me, I'm looking for 200F at least, but less than 250F.
Anyone know of one like this that reads up to 200F?
03-10-2008, 06:56 AM
#6
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
I don't care if it's a k-type or any other type, as long as it does what I want. If I can find a k-type to read (with reasonable sensitivity like this PLX one) to 200F, I'm happy.
Anyone have any leads?
Anyone have any leads?
03-10-2008, 08:12 AM
#7
Honda-Tech Member
Join Date: Nov 2003
Location: Kitchener, Ontario, CANADA
Posts: 5,986
Likes: 0
Received 2 Likes
on
2 Posts
Re: (boostd92)
"Measurement Range 0-150 Deg C (32-302 Deg F)"
^from the plx website. That's 0-150 degs Canada. Look at the chart!
you have to buy that unit to do what you want... which is basically a current to voltage converter.
If a temperature sensor (linear) that plugs directly into a 0-5vdc input exist, you can't afford it.
Trending Topics
03-10-2008, 08:31 AM
#8
Junior Member
Join Date: Oct 2006
Location: umea, sweden
Posts: 218
Likes: 0
Received 0 Likes
on
0 Posts
Re: (2point2)
when i datalogged with my "tweblog" i messured the voltage on the oem intake sensor at 6deg C then at 26deg C and 50 and then 80
i just put the sensor in a glas of water that i warmed upp,
i found out that it was linear
i just put the sensor in a glas of water that i warmed upp,
i found out that it was linear
03-10-2008, 09:06 AM
#9
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
Re: (2point2)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by 2point2 »</TD></TR><TR><TD CLASS="quote">If a temperature sensor (linear) that plugs directly into a 0-5vdc input exist, you can't afford it.</TD></TR></TABLE>
I'm not quite sure why everyone is picking my words apart....
The PLX one will do exactly what I'm looking for, thanks for varifying. $125 is a little pricey, but a simple GM one is ~$50... maybe I can find it wholesale for a pinch less than $125.
I'm not quite sure why everyone is picking my words apart....
The PLX one will do exactly what I'm looking for, thanks for varifying. $125 is a little pricey, but a simple GM one is ~$50... maybe I can find it wholesale for a pinch less than $125.
03-10-2008, 03:17 PM
#10
* B A N N E D *
Join Date: Apr 2007
Location: SoCal, CA
Posts: 848
Likes: 0
Received 0 Likes
on
0 Posts
Re: (boostd92)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by boostd92 »</TD></TR><TR><TD CLASS="quote">
I'm not quite sure why everyone is picking my words apart....
The PLX one will do exactly what I'm looking for, thanks for varifying. $125 is a little pricey, but a simple GM one is ~$50... maybe I can find it wholesale for a pinch less than $125.</TD></TR></TABLE>
1. Thermocouples are very slow and shitty for measuring air temperatures. Especially k-types, which are only used because they're the most common (cheapest to produce.)
2. That isn't a linear sensor. It's a non-linear sensor feeding into a box with a microcontroller inside that converts the logarithmic output voltage to temperature.
3. Thermocouples can't measure absolute air temperatures. They can only measure the temperature difference between two points.
4. IAT sensors use thermistors. I just so happen to be designing one as we speak. You can read up on it here if you want:
https://honda-tech.com/zerothread?id=2240305
Bottom line, thermocouple probes are a joke and can't respond fast enough to register when you're spinning a turbo past it's efficiency range and/or maxing out an intercooler.
I'm not quite sure why everyone is picking my words apart....
The PLX one will do exactly what I'm looking for, thanks for varifying. $125 is a little pricey, but a simple GM one is ~$50... maybe I can find it wholesale for a pinch less than $125.</TD></TR></TABLE>
1. Thermocouples are very slow and shitty for measuring air temperatures. Especially k-types, which are only used because they're the most common (cheapest to produce.)
2. That isn't a linear sensor. It's a non-linear sensor feeding into a box with a microcontroller inside that converts the logarithmic output voltage to temperature.
3. Thermocouples can't measure absolute air temperatures. They can only measure the temperature difference between two points.
4. IAT sensors use thermistors. I just so happen to be designing one as we speak. You can read up on it here if you want:
https://honda-tech.com/zerothread?id=2240305
Bottom line, thermocouple probes are a joke and can't respond fast enough to register when you're spinning a turbo past it's efficiency range and/or maxing out an intercooler.
03-10-2008, 03:56 PM
#11
Honda-Tech Member
Join Date: Jul 2006
Location: missouri
Posts: 4,009
Likes: 0
Received 0 Likes
on
0 Posts
Re: (boostd92)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by boostd92 »</TD></TR><TR><TD CLASS="quote">
I'm not quite sure why everyone is picking my words apart....
The PLX one will do exactly what I'm looking for, thanks for varifying. $125 is a little pricey, but a simple GM one is ~$50... maybe I can find it wholesale for a pinch less than $125.</TD></TR></TABLE>
lol, a gm iat sensor is like $12
I'm not quite sure why everyone is picking my words apart....
The PLX one will do exactly what I'm looking for, thanks for varifying. $125 is a little pricey, but a simple GM one is ~$50... maybe I can find it wholesale for a pinch less than $125.</TD></TR></TABLE>
lol, a gm iat sensor is like $12
03-10-2008, 04:05 PM
#12
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
So would a GM IAT be a reasonable solution?
Could I simply put it in a pot of water and record temperature vs. voltage output as I raise the temp to known points?
Could I simply put it in a pot of water and record temperature vs. voltage output as I raise the temp to known points?
03-10-2008, 04:40 PM
#14
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
I'm really dumbfounded that no one is reading the first post or understanding what I'm after..
How can I use that chart? I can only log VOLTAGE FROM A 0-5V SOURCE.
If I'm stupid and can't get that from the chart you posted, you have my full apology in advance!
How can I use that chart? I can only log VOLTAGE FROM A 0-5V SOURCE.
If I'm stupid and can't get that from the chart you posted, you have my full apology in advance!
03-10-2008, 05:09 PM
#15
Honda-Tech Member
Join Date: Jun 2004
Location: Dont Steal My Car, Ct, USA
Posts: 2,852
Likes: 0
Received 0 Likes
on
0 Posts
Re: (boostd92)
Cant u just use the gm sensor and hook it up to a thing that just displays the voltage numbers?
Then u would know that at 1volt is 10 deg and 2 is 20deg or something like that?
Then u would know that at 1volt is 10 deg and 2 is 20deg or something like that?
03-10-2008, 05:17 PM
#17
Honda-Tech Member
Join Date: Nov 2003
Location: Kitchener, Ontario, CANADA
Posts: 5,986
Likes: 0
Received 2 Likes
on
2 Posts
Re: (mike93eh2)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">Bottom line, thermocouple probes are a joke and can't respond fast enough to register when you're spinning a turbo past it's efficiency range and/or maxing out an intercooler.</TD></TR></TABLE>
I... sort of agree. A lot of people dont realize that an egt (for example) is not instantaneous. your definition of a 'joke' however might be adequate for some people.
according to my google search...
A 1/4" exposed tip has a response time of 0.090s.
Your typical 1/8" (grounded) TC has a response time of 0.34
and to put that into perspective. "On average, a blink takes approximately 300 to 400 milliseconds." Denso claims a 100ms response time for their o2 sensors.
Bottom line, thermocouples are good enough for what I need!
I... sort of agree. A lot of people dont realize that an egt (for example) is not instantaneous. your definition of a 'joke' however might be adequate for some people.
according to my google search...
A 1/4" exposed tip has a response time of 0.090s.
Your typical 1/8" (grounded) TC has a response time of 0.34
and to put that into perspective. "On average, a blink takes approximately 300 to 400 milliseconds." Denso claims a 100ms response time for their o2 sensors.
Bottom line, thermocouples are good enough for what I need!
03-10-2008, 06:45 PM
#18
* B A N N E D *
Join Date: Apr 2007
Location: SoCal, CA
Posts: 848
Likes: 0
Received 0 Likes
on
0 Posts
Re: (2point2)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by boostd92 »</TD></TR><TR><TD CLASS="quote">I'm really dumbfounded that no one is reading the first post or understanding what I'm after..
How can I use that chart? I can only log VOLTAGE FROM A 0-5V SOURCE.
</TD></TR></TABLE>
You CAN'T. I already told you that. Unless you build a resistor network, or program a microcontroller to decode the sensor and output a voltage.
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by 93supercoupe »</TD></TR><TR><TD CLASS="quote">Cant u just use the gm sensor and hook it up to a thing that just displays the voltage numbers?
Then u would know that at 1volt is 10 deg and 2 is 20deg or something like that?</TD></TR></TABLE>
Knowing the values doesn't help - that's what's already posted in the chart above.
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by 2point2 »</TD></TR><TR><TD CLASS="quote">
according to my google search...
A 1/4" exposed tip has a response time of 0.090s.
Your typical 1/8" (grounded) TC has a response time of 0.34
Bottom line, thermocouples are good enough for what I need! </TD></TR></TABLE>
BS! Most bare-wire thermocouples take several seconds to stabilize. The autometer temperature probe had a "fast response thermocouple" (typical k-type) which had a time constant of about 1 second. That means full temperature stabilization would take about 5 seconds.
The factory Honda IAT sensor has a time constant of 15 seconds in MOVING air. People who datalog their IATs don't seem to realize that they're basically just measuring how much their sensor/manifold has heat soaked. Nobody knows that they aren't really logging their air temps.
And thermocouples just suck all out. They don't even measure air temperatures - they measure the difference between the "hot" side and the "cold" side. So you have to mount the end of the probe somewhere where it will always reflect ambient temps. And the leads are part of the sensor calibration, so you can never extend or shorten the probe. They're also just flat out inaccurate, due to EMI.
How can I use that chart? I can only log VOLTAGE FROM A 0-5V SOURCE.
</TD></TR></TABLE>
You CAN'T. I already told you that. Unless you build a resistor network, or program a microcontroller to decode the sensor and output a voltage.
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by 93supercoupe »</TD></TR><TR><TD CLASS="quote">Cant u just use the gm sensor and hook it up to a thing that just displays the voltage numbers?
Then u would know that at 1volt is 10 deg and 2 is 20deg or something like that?</TD></TR></TABLE>
Knowing the values doesn't help - that's what's already posted in the chart above.
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by 2point2 »</TD></TR><TR><TD CLASS="quote">
according to my google search...
A 1/4" exposed tip has a response time of 0.090s.
Your typical 1/8" (grounded) TC has a response time of 0.34
Bottom line, thermocouples are good enough for what I need!
</TD></TR></TABLE>BS! Most bare-wire thermocouples take several seconds to stabilize. The autometer temperature probe had a "fast response thermocouple" (typical k-type) which had a time constant of about 1 second. That means full temperature stabilization would take about 5 seconds.
The factory Honda IAT sensor has a time constant of 15 seconds in MOVING air. People who datalog their IATs don't seem to realize that they're basically just measuring how much their sensor/manifold has heat soaked. Nobody knows that they aren't really logging their air temps.
And thermocouples just suck all out. They don't even measure air temperatures - they measure the difference between the "hot" side and the "cold" side. So you have to mount the end of the probe somewhere where it will always reflect ambient temps. And the leads are part of the sensor calibration, so you can never extend or shorten the probe. They're also just flat out inaccurate, due to EMI.
03-10-2008, 08:16 PM
#19
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
Re: (Legion)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">You CAN'T. I already told you that. Unless you build a resistor network, or program a microcontroller to decode the sensor and output a voltage.
</TD></TR></TABLE>
Like this for the GM IAT?
</TD></TR></TABLE>
Like this for the GM IAT?
03-10-2008, 08:49 PM
#20
* B A N N E D *
Join Date: Apr 2007
Location: SoCal, CA
Posts: 848
Likes: 0
Received 0 Likes
on
0 Posts
Re: (boostd92)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by boostd92 »</TD></TR><TR><TD CLASS="quote">
Like this for the GM IAT?
[img]https://www.innovatemotorsports.com/images/schematic.jpg[img]</TD></TR></TABLE>
So the LM-1 software has been programmed to decode the GM sensor resistance/temperature curve? That's a pretty big help. Looks like your job just got a lot easier.
That circuit is overly complex. It assumes you'll be using a dirty 12v power supply, which it then regulates to 5v via the zener diode. If you tap into a clean 5v power source through the ECU you can power the GM sensor directly, then all you'll need is the load resistor!
Like this for the GM IAT?
[img]https://www.innovatemotorsports.com/images/schematic.jpg[img]</TD></TR></TABLE>
So the LM-1 software has been programmed to decode the GM sensor resistance/temperature curve? That's a pretty big help. Looks like your job just got a lot easier.
That circuit is overly complex. It assumes you'll be using a dirty 12v power supply, which it then regulates to 5v via the zener diode. If you tap into a clean 5v power source through the ECU you can power the GM sensor directly, then all you'll need is the load resistor!
03-11-2008, 05:39 AM
#22
Honda-Tech Member
Re: (Legion)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">
The factory Honda IAT sensor has a time constant of 15 seconds in MOVING air. People who datalog their IATs don't seem to realize that they're basically just measuring how much their sensor/manifold has heat soaked. Nobody knows that they aren't really logging their air temps.
</TD></TR></TABLE>
I've been saying this for a long time. From the test data that i have on the track, there's about a 12 second delay with the honda air temp sensor. On a 1/4 mile pass the air temp will peak 12 seconds after the throttle is lifted at the end of the run. This is with an obd1 temp sensor and a motec ecu. Same ecu with a GM temp sensor and the air temp peaks 1 sec after the throttle is lifted.
This is why i laugh at "data" from people saying their ebay intercoolers are good because they only see a few degrees temp gain on a dyno pull. lol
The factory Honda IAT sensor has a time constant of 15 seconds in MOVING air. People who datalog their IATs don't seem to realize that they're basically just measuring how much their sensor/manifold has heat soaked. Nobody knows that they aren't really logging their air temps.
</TD></TR></TABLE>
I've been saying this for a long time. From the test data that i have on the track, there's about a 12 second delay with the honda air temp sensor. On a 1/4 mile pass the air temp will peak 12 seconds after the throttle is lifted at the end of the run. This is with an obd1 temp sensor and a motec ecu. Same ecu with a GM temp sensor and the air temp peaks 1 sec after the throttle is lifted.
This is why i laugh at "data" from people saying their ebay intercoolers are good because they only see a few degrees temp gain on a dyno pull. lol
03-11-2008, 06:48 AM
#23
can the Gm IAT replace the honda sensor using the s300? it might listed already but I did not find it?
don't mean to thread jack but we are looking for the same answers in logging iat with our zt2 wideband, I will call zeitronix tonight to see what they have in planning later. or if they have any way of making the unit reads temperature.
don't mean to thread jack but we are looking for the same answers in logging iat with our zt2 wideband, I will call zeitronix tonight to see what they have in planning later. or if they have any way of making the unit reads temperature.
03-11-2008, 07:16 AM
#24
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
Re: (Legion)
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>Quote, originally posted by Legion »</TD></TR><TR><TD CLASS="quote">So the LM-1 software has been programmed to decode the GM sensor resistance/temperature curve? That's a pretty big help. Looks like your job just got a lot easier.</TD></TR></TABLE>
I don't believe it decodes anything. From the threads I've read over there, people are just using voltage vs temperature charts to convert the output from that schematic into useful info.
..and I have a Zeitronix... which I think is the same in terms of input as the LM-1 software... no decoding.
So... tell me again why I couldn't put the GM sensor in a pot of water and record voltage vs. temperature and use that chart to reference the datalogged voltage from the sensor in the car? I'm still a little confused..
I don't believe it decodes anything. From the threads I've read over there, people are just using voltage vs temperature charts to convert the output from that schematic into useful info.
..and I have a Zeitronix... which I think is the same in terms of input as the LM-1 software... no decoding.
So... tell me again why I couldn't put the GM sensor in a pot of water and record voltage vs. temperature and use that chart to reference the datalogged voltage from the sensor in the car? I'm still a little confused..
03-11-2008, 07:22 AM
#25
Junior Member
Thread Starter
Join Date: Apr 2006
Location: Chicago, IL, USA
Posts: 808
Likes: 0
Received 0 Likes
on
0 Posts
I suppose I should be more clear. I don't care if my datalogging software can't convert/decode the voltage and turn it into temperature. As long as it can RECORD the voltage, and that voltage is going to mean the same thing everytime, I don't mind referencing an Excel chart to see what the voltage means.
For instance, if I put the GM IAT in a pot of water with a thermometer (even if I have to take a crock pot out to the car to make sure the supply voltage/grounds are consistant), if I record that 1.5V = 90F and do similar for say 5F increments, will that info be accurate with the temp probe installed in the car?
For instance, if I put the GM IAT in a pot of water with a thermometer (even if I have to take a crock pot out to the car to make sure the supply voltage/grounds are consistant), if I record that 1.5V = 90F and do similar for say 5F increments, will that info be accurate with the temp probe installed in the car?