[sti robot]

Originally Posted by fullerton

cool.

question: why are you soldering a pressure sensor? shouldn't it have a hose that connects to it so you can tap into the BPV's line?

True...

It goes bpv---(hose)---pressure sensor inlet (male)---3 wires to microcontroller(1 is the reading and the other two are power and ground)

I've got to wire those three to the sensor. The sensor is really meant to be popped onto a board with a chip socket. I might use one of those...but then it'll cost just as much to switch to a pressure sender...one that we know can take the abuse a car ride dishes out. The think I have right now is downright delicate.

the vs on these two types of sensor:

mini pressure (what I have now) - tiny - $14 - works on 5 volt scale -
regular pressure sender - bigger - $18 - works on resistance values - does neg (vac) pressure better -



Back on with the temp sensors:
I found a chart of the temp sensor's R-T (resistance - temp) curve/values. I tried fitting a line to it but that isn't happening easily. I'm going to have to fit various lines to various portions of it. When I get home (dang lunch break is almost over) I'll try to fit a better line or figure out how many segments I'll need to figure curves for (to get the right amount of tolerance). I'll also have to wire it up with a 1k resistor vice the 10k so that the Voltage values coming out fit within the 0-5V window of the analog in pin.

EDIT: Apparently there is a forumla for this type of stuff. Unfortunately I may have to stick to the original segmented power curves idea due to limitations in the math libraries provided for the microcontroller.

http://www.moshier.net/rtd-readme.html
http://en.wikipedia.org/wiki/Thermistor

Or I could do lookup tables. With 2degrees of resolution that would waste a whole lot of space on the chip.

[sti robot]

Doing ln() or log() with the chip I have would require lookup tables to report anything near real time (you could do a aprox forumal but it would create an unnecessary lag in temp reporting time). So lookup tables for the actual values makes more sense.

Look up tables would be easier (ie take up less space on the chip) if I had an idea of what temps we would be measuring.

-160 tops for tire temps
-ic temps turbo side are probably 350???
-ic cold side are ??
-intake temps top out at 120 (autometer gauges for this go to 300!)??
-Anything in the EGT range should probably be done with an EGT probe, so that's out.
Low threshold would probably be freezing...just in case someone want to use it as an ambient temp thing.

So what would this be useful for...
-underhood ambient temps (lay to rest the intake scoop size and diffuser debates)
-intercooler fin temps (compare one side to the other)
-radiator fin temps (not really useful)
-Cold air at the fender
-Air temps with one of those nifty shorty intake shields and/or boxes
-tire temps (hop out and measure in between autox runs or between laps)

I found this article in autospeed which was pretty interesting. http://autospeed.drive.com.au/cms/article.html?&A=0578 I like the way he mounts his sensors.

Also, I wouldn't dare mount this in an air intake (you risk sucking things up into your engine...that would be bad). Doing so would be dangerous. If this is what I wanted to do I would use one of these http://www.egauges.com/vdo_send.asp?Sender=Intake_Temp. This probe looks more like an egt probe in that is screws in place (you'd have to tap or weld a bung into your intake)

[Mack]

well, that seems to solve the question of where to put the temp sensors on the IC. would be nice to mount on on the filter of my CAI, so I can see three readings: Intake, Pre-IC and Post IC.

[sti robot]

One just wouldn't be able to (easily) get a reading from inside the inlet hoses.

[Mack]

easier, albeit less accurate, to place the sensors on the outside vs inside the hoses.

[Meridock]

Robot and I were at a local subby meet and eat and he showed me the demo mode. Very cool. he has a decent interface. (needs bolder "red" IMHO for bright daylight use) but once the senders are working the unit is solid.


Robot has some aesthetic things to work on (button location,) but all and all - I will source one but will need to have a custom mount or display since I love my stock clock


this unit is very flexible. You want pressure sensors instead of temp - he is working on it. It even looks like he could do "if then" ie if pressure/temp hits a threshold - perform XX and if XX could be activate a relay to do YY - you have the makings for an affordable temp / boost management system. (thinking of autospray of IC / CO2 on intercooler, etc)


This project leaves a bunch of things open to the imagine.




Good work so far.

[sti robot]

^ Thanks for the good words.

And to keep this above board:
Remember I'm not a vendor (currently) and you cannot approach me (currently) and ask to purchase anything.

[sti robot]

Originally Posted by Meridock

this unit is very flexible. You want pressure sensors instead of temp - he is working on it. It even looks like he could do "if then" ie if pressure/temp hits a threshold - perform XX and if XX could be activate a relay to do YY - you have the makings for an affordable temp / boost management system. (thinking of autospray of IC / CO2 on intercooler, etc)

Essentially anything that has an on/off type circuit is accessible via relays. So your IC spray, C02, meth (these have good controllers already though), etc. could be controlled.

[Mack]

Originally Posted by sti robot

^ Thanks for the good words.

And to keep this above board:
Remember I'm not a vendor (currently) and you cannot approach me (currently) and ask to purchase anything.

I am sure the powers that be will frown upon all-out selling without a vendor account.

However, I would bet they're ok with you using a few beta testers to work through your product and get it ready for vendorship.

[sti robot]

The beta test thing is my plan...

I figure as long as I'm not using IWSTI to turn a profit they won't have a problem. Beta testers would be merely purchasing the materials.