Thoughts on Temp Sensor


Forum tech Advisor
Staff member
Jul 31, 2001
Ok, I get a little extra time and things get out of hand around here :)

I'm developing a simple(cheap) protection system for use in conjunction with the alcohol kits.

Initial thought process is pressure switch on pump outlet. Problem is if nozzle were to be clogged, all the pressure in the world would be available and the protection wouldnt be there.

Then I thought, possible current tracking.. in other words see how much the pump is pulling and if the current is within range, leave things as is. Problem is current can vary from vehicle to vehicle. And not really an easy way. Plus the plugged nozzle syndrome will still exist.

Then.. flow meter.. this would be the E-ticket. Get a chemically resistant flow device that outputs pulses as the liquid flows through it. Problem I have found is the meters will be upwards of 150+ plus fittings/electronics/interfacing.

Finally what I've been using has been my ATS sensor relocated into my plenum. Thats the sensor located by the air filter on the GN. Ive been tracking intake air temps with that device now for the last year. So here's my thoughts. Please chime in.. make this happen.

Run a conditional switch, so that when the alky pump is energized, the circuit activates. Then have a wire sense the reading of the sensor and if the reading goes past a certain point.. trigger a relay that fires a warning device.

Elaborated a little further. I'm going to use an autometer pencil shift light tied into the relay. I'll set the threshold like on my own car at 110 degrees. If the temps inside the intake climb past 110 degree's.. it fires the light.. and can also control a boost solenoid or other protection of your choosing. Audible alert, lamp, etc...

Issues.. must drill a pretty big hole into the plenum to mount the sensor. Its 3/8 pipe NPT thread. Other option is to mount sensor after the nozzle. There are thermistors that are really small that may be slipped in under the pressure hoses. The neat thing about the GM sensor.. the thing works flawless.

If the alky system ever stopped spraying for whatever reason, the temps would climb fast and trigger the device.

Now the issue is to make the system bare bones(cheap) or add electonics like a bar graph to show temps. Bare bones, would set it, stick it in the glove box.. forget about it. Reasoning.. think we have enough to look at while going down the track. And a DS can record the sensors operations through the MAT input.

Thoughts? Opinions? Ideas?

Protection is always a good idea. :)

Probably need some adjustability with the device temp. wise.

Temp. tracking is always a plus.

Other devices like the flow meters are too expensive unfortunately, as mentioned.

I was going to use the last yellow of the Caspers gauge to trigger a relay opening power to the boost solenoid(s), reverting back to spring pressure should that light ever trigger.

Easy to do with simple electronics also.

That scenario assumes no false KR on the car which right now I don't have.
Wouldn't an audible knock sensor be adequate? If the alky suddenly goes off, I would think your gonna have knock ultra quick which would set off a knock alarm instantly so you can try get out of it in time. Do you really think the addition of the temp sensor system would lower this reaction time? I would think that if you had your car tuned to the high boost edge with alky and you had a pump failure .... not much is gonna save you:eek: Just the risk you take I've always figured.

How about this ideal.... a backup pump that comes on through a separate nozzle if any loss of pump pressure in the initial system is triggered along with some kind of indicator.;)
You said
"How about this ideal.... a backup pump that comes on through a separate nozzle if any loss of pump pressure in the initial system is triggered along with some kind of indicator."

Indicator is the question. And a backup would be more costly than the flow meter mentioned. Let alone complexity.

Having been around these systems long enough. Two things are going to happen. Either they'll work or they wont. Not one issue has been from it stopped mid track. Think if the issue arose, IAT's will go up.

I run an audible, I have a scanmaster, and record with direct scan. The audible is hard to hear with a helmet on, or have the cut-out open. Scanmaster requires you to be looking at it. In the heat of the moment..that can sometimes be tough.

And if the system were not to spray, at 10 PSI boost there would be over 120 degree's inside the motor. So this would catch it way earlier :)

For every lb of boost the air gets heated 10 degree's. 10 PSI =100 degree's + ambient coming out of the turbo. A big IC will cuel that temp some.. but looking for a pattern. This will go off way before a knock sensor will. Thats the idea.

I have had ZERO problems with the kits. Just looking for an umbrella policy ;) making sure all is covered.

My car at 25 PSI will run ambient. So if its 70 degree's outside. 65-75 degrees through the traps. I can set the sensor for 90 degree's and still have room. Zero chance :)

And if the system were not to spray, at 10 PSI boost there would be over 120 degree's inside the motor. So this would catch it way earlier

Yeah, but what is our definition of way earlier.... miliseconds. How long does it take your car to build boost in first gear? How is this gonna be fast enough to prevent that ultra lean mixture from getting to the combustion chamber where it will go BOOM. Reading your past posts, you recomend to your customers to lean the fuel out and run as much alky as possible. You claim that you have proven that this provides the best performance gain with alchohol and I feel you have proven your point. The problem now though, which I commend you for trying to figure out, is that if there is any kind of failure in the system, you'll be ultra lean at a high level of boost. Your screwed.

I have had ZERO problems with the kits. Just looking for an umbrella policy making sure all is covered.

Reading your posts over the last couple of years, this has been a subject you have always tried to overcome, and its awesome that you've tried but ......

In the case of a pump failure or clogged nozzle that cuts the alcohol immediatly at WOT, your temp sensor will not make any more difference when racing than a knock sensor. As you clearly understand, severe detonation has to be prevented and your sensor tied to whatever means of stopping ignition, will probably be too late. Once that ultra lean fuel mix is in that cylinder under that much boost .... BOOM. Your temp sensor will not prevent this..... not much will. Even my NASA'esqe ideal of adding a backup system would not work because by the time the pump pressured up to combat the problem to spray... it would be too late.

Tuning in the way you recommend to your customers increases the inherant risks a bit. Basically if the pump or anything goes, your way lean and screwed. To your credit you have tried to find the best pump and have provided the best parts you could find to prevent this but at the end of the day, the risks, small as they may be, are still there. Thats part of going fast.

The part that makes this the toughest is that your trying to sovle this problem on a limited budget. Any system you design will have to be tested under real world conditions and that could get pricey....;)

As far as not being able to hear your knock detector

:confused: That should be a very small problem for your electrical prowess.;)

None of this post is meant to ruffle your feathers as I think you've done an awesome job with your kits. I'm just throwing thoughts out there like you requested and I do hope that you are right and that this would make a difference, although I don't wanna test it on my car.;)
I don't have a problem hearing the Caspers sensors they are over by the ECM hanging down a bit into the footwell.

Problem is I like to ignore them at the track. ;) :eek:

I think the LED trigger from the knock sensor on whichever LED you prefer, last green, first yellow, last yellow, etc. will catch the escalating KR the quickest and help save the motor.

Assuming your spring pressure isn't well over 20 psi. :p
One thing that is absolutely certain, while we sometimes ignore knock sensors thinking its fasle knock... a temp sensor wont lie.

I understand what you mean about time. After all, we are dealing with time. How long.. man if multiple conditions are met.. its the checks and balances.

As Salvagev6 wrote
"Problem is I like to ignore them at the track. "

True'r words have never been spoken. Ask guys like Eric Marshall and Steve Yaklin.. Theyre chips have knock ignore as features.. So while your getting out the gate.. the knock sensor reads ZERO.. so timing is not pulled out due to false knock.

The motors will handle a bunch of brief knock. They wont handle sustained knock. So even if the system didnt spray... by the time your knock sensor and temp sensor went off.. you'd be more than covered. So even if you leaned the motor out, threw timing.. you can still recover from that. Just dont keep the pedal down..else the problem will come up.

Our knock sensors start getting treated like car alarms.. we choose when to pay attention to them. There is always that thing in the back of the mind.."Is it false" hence why I wanted to add something.

Turbot2496, no feathers getting ruffled. I want this kind of feedback. That way I know if there is interest in such a device. If not, I wont worry bout cosmetics and make only one and install it on my car :D
how about a n/o lowpressure switch and n/o highpressure switch (might be able to use a/c senders) , mount them before the filter with the switches controlling a large wg soleniod (like the emmission canister solenoid ) , the low switch would need another hobbs switch inline preventing it from completing the ground until preset boost is reached say at 12lbs
,which should be enough to have alky pressure above the preset low pressure switch , if pressure drops the low press switch
closes completes ground and the wg solenoid which is T'd in to the wg opens and boost drops , if pressure builds above the prest highpressure switch circuit closes completes ground and boost drops

should work and will work fast but dont know if it will prevent damage since it can occur quikly
Rather than trying to make it an alarm type system requiring quick action by the driver, why not have the temp sensor trigger a relay to cut off the spark to the engine, just like the ECM would with the stock chip if the max speed was reached. Then you don't have to rely on the driver seeing/hearing an indicator....the motor just shuts off.....will probably scare the crap out of you if it ever happens, but at least it would save the motor...... thoughts ???

May not have to be temp related, but maybe pressure related in the alky line...if pump doesn't work and should be on...then no pressure....if nozzle is clogged then I'd think it would have too high pressure....pressure sensor would be as close to nozzle as possible.
Actually, i'll take it a step further.

I spoke with Steve Yaklin today about possibly incorporating this into the chip. Gotta get ahold of Eric.. sounds easy. Set your temp table just like you would your fueling/timing..make it programmable. Just like we do for the version 5.3 chips and adjust fueling/timing. Then if the MAT temp goes past that certain point, fire the check engine lite and shut the motor down. In the case of the ME-r using one of the 16 TW positions may be a possibility.

Problem is anything having to do with the chip... only allows fuel to be cut. In other words.. hit the top end limit.. TPS.. MAF.. RPM.. cuts fuel.. not to say its a bad thing.. but some thoughts..

Now incorporating the WG solenoid into the equation may be the neat thing to do bringing boost down to the base setting while doing other things like timing/adding fueling etc to safely get away. Having the check engine light come on would set a MAL code..whereby you can see the reason it was set off. In this case ATS(air temp sensor) error :)

Conditions.. If MAF is greater than 220 grams..and temps greater than the threshold.. bam fire the lamp??

Hows that for an alky chip :D

Thats easier than making electronics.

Still.. I got my Autometer 3239 mini-Pro-Lite in the mail today.. :D

results may vary, use at your discretion, not responsible, bla bla bla ;)
IMO, you want something with an instant response, ie drop the timing.

Now, what to sense:
The best answer would be an ionization abnormal combustion devise, but that's still some time away.

A high egt,
High MAT,
but both are rather slow in response to detonation.

Maybe something like the MSD knock sensor monitor. Just use an op-amp, and then a simple voltage threshold for knock. Without the frequency filtering like the oem.

What ever the route, while you can work at detecting Knock, any acoustic devise, is prone to missing preignition, and that can be caused by detonation, or NOT.

Or maybe, a cylinder miss detector using a ring gear mounted sensor.

Speed and response, are critical, in saving an engine. While an engine can take some detonation, if that detonation leads to preignition, you don't have long at all. Might read the Abnormal Combustion thread for more details.