Adding coolant temp gauge?

[RavenTai]

Sorry RT, can't help much with the heavy circuitry. I did try to email tech support for Yazaki Meter last week, to ask what was done to the circuitry to create the flat spot in the gauge. No response yet.


Rookie2

Good deal, I am pretty sure the flat spot caused by the zener diode but it would be great to get confirmation, also if they respond it will surely give some other useful info.

Thank You,

RavenTai,

You are an amazing dude!

Keep plugging away. I have absolute faith in you.

BTW - what do you do for a living - Rocket Scientist?

Thanks you, but electronics on this scale is not beyond any one here, a couple hours of basic electricity would get you there. I am an aircraft mechanic (A&P/AMT) at a major airline (dying). I had some basic electricity and avionics classes in school and have dabbled with computers and automotive electricity since then.


I’ll keep playing with it but to come up with a mod I am going to need a gauge to test it on, if you see a gauge or cluster for sale at a reasonable price let me know.

 

[Gauge]

Aircraft Mechanic - Rocket Scientist - Close enough You are still way out of my league. For all I know a Zener Diode could be another name for "Muffler Bearings".

Like I said, I have faith in you man!

BTW - I don't think a couple of hours of "basic electricity" will get me to your level of understanding, but if you believe in me, I guess I should take that as a compliment.

Thanks for using your knowledge to help solve this.

 

[Heffenoche]

This may be a little too basic for this thread but once one installs a temp gauge, what is normal operating temperature? Than at what temp would one become a little concerned and at what temp would one pull over and shut it down and at what temp am would one be fairly certain they have done severe damage?

My guesses, to see how wrong I might be:

Normal operating temp 175-190
A little concerned 191-220
Pull over & shut down >221 (226 ac cuts off)
Eminent damage >240

 

[Photoman]

I was curious about the chart data I had posted so I set the temp sensor up again and ran the numbers. The equipment I used both times was a Raytek ST80 Pro with the remote temperature probe, and a R Shack Micronta multimeter. I put 15w-40 motor oil in a cup that I used as the medium to heat up with a Coleman stove. I took a pic of the test setup but it turned out pretty messy. If anyone wants it posted let me know. The meter was set on the 200 Ohm scale. I’ve been messing with this all day and keep getting different numbers. If I test with the oil heating up it’s one set of numbers, and cooling down a different set so I think I needed a larger oil volume. Here is a pic of the best numbers I could come up with. If anyone wants an Excel chart/file, PM me with your email address. I would just use the chart for reference. Basically, from 175-200 F the temp changes 2 degrees per ohm. 200-210 3 degrees per ohm. 210-220 4 degrees per ohm. 220-230 5 degrees per ohm. 230-245 6. 245-260 7. From 28 ohms at 175F to 5 ohms at 260F. Good luck.

Bill

 

[Rookie2]

Thanks you, but electronics on this scale is not beyond any one here, a couple hours of basic electricity would get you there.

If you aware of a good beginers manual for learning basic circuitry and testing, I'd love to read up and learn more. I had a 100 level EE circuits class about 15 years ago, but most all that has left me. I know about how to turn my voltage meter on, but I still find myself finding out that something is hot... the hard way .


Rookie2

 

[RavenTai]

For all I know a Zener Diode could be another name for "Muffler Bearings".

Like I said, I have faith in you man!

Thanks for using your knowledge to help solve this.

no problem, I want a accurate cluster gauge also, when I figure out how to do so I will be glad to share the info back to a community that I have learned a lot from.

Just so you know, a diode is an electrical check valve. It allows current to flow one way but snaps shut preventing current in the opposite direction.

A zener diode is a special kind of diode, like a diode it allows foreword current and stops reverse current but if the voltage of the reverse current gets to a certain pre set level it will allow this reverse current to flow.

This range between where the current just starts to reverse and when the zener diode finally lest some of the current pass is where the flat spot in the gauge comes from, during the flat spot the circuit is artificially balanced and the needle does not move

BTW - I don't think a couple of hours of "basic electricity" will get me to your level of understanding, but if you believe in me, I guess I should take that as a compliment.

It is intended as a compliment, all the members here seam to be pretty smart. This circuit is not too bad, with the proper tools you could figure this one out.

This may be a little too basic for this thread but once one installs a temp gauge, what is normal operating temperature?

That is actually quite important,

1. I would also like to know what those with real gauges are seeing for typical temps. CDan has posted his recent experience above but his 80 is not at all typical in relation to temperatures. this info will be needed to calibrate a linearized gauge. I would still like my fixed gauge to point to the middle when it is at its average temperature. This typical temperature may be different with different weather but would like to get it close. Adjusting the value of the resistor R1 should adjust what temperature the gauge centers at. Assuming the two inductors (L2 and L3) have the same resistance then when the resistor R2 (thermistor) and R1 (set resistor) have the same value the bridge will be balanced and the needle will be in the center

2. I would also like to know what people think would be a good range for this gauge. How high and low do you want to see? A tight range will show the slightest changes in temperature. But miss the extreme highs and lows. A wide range will allow you to watch it warm up on the cold side and also watch your motor melt on the high side.

At first though I am thinking I want a tight range, don’t care to read much below a certain temperature. Pegged “cold” will be good enough for me. At the other end I don’t care to read extremely high temperatures. A pegged gauge at “holly s*!t that’s hot you should have shut down long ago idiot” is good enough. But maybe there are some reasons to see wider ranges?

I am a little less certain how this can be adjusted but I think a resistor in place of the zener diode will give a wider range, the higher the resistance of this new resistor the less sensitive the gauge and the wider the range. For a tighter range we will not be able to go below no resistance, or basically a wire in place of the diode whatever range of temperatures that gives. A no resistor gauge should show slightly narrower than it shows stock.

Resistors could also be added in series to the L2 and L3 inductors to do other things but those would be more difficult to add.

I was curious about the chart data I had posted so I set the temp sensor up again and ran the numbers. The equipment I used both times was a Raytek ST80 Pro with the remote temperature probe, and a R Shack Micronta multimeter. I put 15w-40 motor oil in a cup that I used as the medium to heat up with a Coleman stove. I took a pic of the test setup but it turned out pretty messy. If anyone wants it posted let me know. The meter was set on the 200 Ohm scale. I’ve been messing with this all day and keep getting different numbers. If I test with the oil heating up it’s one set of numbers, and cooling down a different set so I think I needed a larger oil volume. Here is a pic of the best numbers I could come up with. If anyone wants an Excel chart/file, PM me with your email address. I would just use the chart for reference. Basically, from 175-200 F the temp changes 2 degrees per ohm. 200-210 3 degrees per ohm. 210-220 4 degrees per ohm. 220-230 5 degrees per ohm. 230-245 6. 245-260 7. From 28 ohms at 175F to 5 ohms at 260F. Good luck.

Bill

Bill thank you for the updated chart and info, do you think the internal temperature of the thermistor lagged behind the temperature of the oil?

If you don’t mind I would like the see the test setup and any advise on other ways it can be done. I will be doing something similar to test the entire gauge circuit.

PM coming your way.

If you aware of a good beginers manual for learning basic circuitry and testing, I'd love to read up and learn more. I had a 100 level EE circuits class about 15 years ago, but most all that has left me. I know about how to turn my voltage meter on, but I still find myself finding out that something is hot... the hard way .


Rookie2

I had a Glencoe text book,
Aircraft Electricity and Electronics in school, it does a good job of explaining basic theory both AC & DC that would apply to all electronic devices and then goes on to explain aircraft systems both general aviation, witch is very much like 1950’s automotive, alternators, magneto’s, batteries etc and also commercial with is more complex and specialized with communication, radar, navigation, specialized systems, etc. The later parts of the book may not be that useful to you beyond just information but at the used price (4$-20$) it is worth having IMO. I don’t think it would pay for you to buy a new one at $60.

I am sure there are some books more specific to what you are looking for, but I have not read any beyond the above to be able to give any advice on them. Maybe a trip to the library?

There is also a lot on the web, I use Google searches to get refreshers on formulas and components.

Here are some on-line primers. Learn Ohm’s law first. Learn it until you are tired of it. It is the core of electronics. Ohms law is simple to use (multiply and divide) but powerful to apply, using some known values it can solve for unknown values with certainty.

Learn about resistors in parallel and series then you can move on up. Go ahead and work the math examples just reading will not get you a working understanding.

http://www.autoshop101.com/autoshop16.html

http://www.allaboutcircuits.com/

http://www.electronics-tutorials.com/basics/starting-out.htm

http://hyperphysics.phy-astr.gsu.edu/hbase/electric/ohmlaw.html

 

[Photoman]

Bill thank you for the updated chart and info, do you think the internal temperature of the thermistor lagged behind the temperature of the oil?

If you don’t mind I would like the see the test setup and any advise on other ways it can be done. I will be doing something similar to test the entire gauge circuit.

RavenTai,

I think the speed at which the small amount of oil heated up affected the results by creating temperature layers. Along with that, the probe depth with this amount of oil appears critical. The sensor has to be near the top of the oil in order to keep the ground preserved (not shorted in the oil). I tried to match this depth with the probe tip to keep the temp readings as accurate as possible. To do this test correctly in my opinion, a larger amount of oil should be heated more slowly. Possibly doing the test backwards by heating the oil, then taking the readings as it cools, with some agitation, may work better (but much slower). The first time I did the test I used a small propane torch and played the flame on the cup, heating the oil more slowly. This time, even on it’s lowest setting, the propane heater caused the temps to rise rapidly. It would have been better with two people, one to call off the readings and the other to record them. As it was I just held the trigger down on the IR gun, and took the probe readings every 1 ohm change.
The setup in the pic has a blue cup with the oil in it. On top of that I just used a small piece of thin plywood drilled to fit the sensor and beside that, drilled for the probe. I made sure the oil level was up on the sensor and the probe was at that same height. I did have to put a large washer under the cup to get it to fit the Coleman; but I don’t think that would affect anything. A little macro tripod to hold the wires away and that was it. Pretty booty fab.
The first pic shows the whole setup. The second closer pic shows a meter reading of 172.6 which was at oil and ambient 73 F. Let me know if you need anything else.

Bill

 

[cary]

Important question, what kind of oil did you use for the test?

Seriously, I am so glad that someone is finally working out a fix. My electrical skills are limited to crossover design, and I haven't done that in years.

Concerning the range of the guage, I would say a wide range. This would keep the guage from bouncing like a pogo stick every time you speed up or slow down, but would show when the motor starts to heat up.

 

[Rookie2]

I had a Glencoe text book,
Aircraft Electricity and Electronics in school, it does a good job of explaining basic theory both AC & DC that would apply to all electronic devices and then goes on to explain aircraft systems both general aviation, witch is very much like 1950’s automotive, alternators, magneto’s, batteries etc and also commercial with is more complex and specialized with communication, radar, navigation, specialized systems, etc.

Awesome. Got one on the way. This is perfect, learn a little about electronics, and have another resource for enhancing the GA knowledge.


Rookie2

 

[semlin]

great work guys, this is fun to read...

concerning the average temp, I suspect that will fluctuate somewhat between trucks. So many possible variables in the vehicle to affect temp and so many miles on them. I would suggest instead targeting the standard thermostat opening temp at say 1/3 or 1/4 of the dial.

I vote for a wide range that would include all possible operating temps. I think that once we have the flat spot eliminated, a wide range will have enough sensitivity for what we need. A significant gauge bounce every time you went up a hill or stopped at a traffic light would be distracting.

Once we have a working gauge, It would also be great to have an overlay sticker to apply directly to the gauge face (not the bezel) with a few data points. That seems like a job for a graphic design person to come up with something that looked reasonably oem.

Another thought I have had on the LED alarm issue is that the turbo diesel 80's have an LED light in the tachometer instrument face to indicate when the turbo is on. The tach is otherwise identical so I wonder if the 80 gas series tachs might already have a ready made mounting point and punchout hole for an LED that could be drilled to allow for this as a temp alarm.

 

[Gauge]

I vote for a wide range that would include all possible operating temps. I think that once we have the flat spot eliminated, a wide range will have enough sensitivity for what we need. A significant gauge bounce every time you went up a hill or stopped at a traffic light would be distracting.

Actually I don't mind watching the gauge rise when I am climbing a hill and go down when I am descending - at least I know that my gauge is working. And if my engine gets hot, even if it is for a short time, I don't mind knowing that.

Once we have a working gauge, It would also be great to have an overlay sticker to apply directly to the gauge face (not the bezel) with a few data points. That seems like a job for a graphic design person to come up with something that looked reasonably oem.

We should consider doing this for the Oil Pressure gauge as well - maybe throw some PSI amounts next to those tick marks.

Just my $0.02

 

[turbocruiser]

Wow, great reading, great posts, great thread, however, and i super sincerely hope not to offend any at all, there are way way way too many great guys with great minds here spending way way way too much time on this problem. I say this for several specific reasons:

1. No matter how you electrically engineer the stock gauge to get more accuracy you are still working with something that is not marked and not metered and something that will travel only 1.5 inches or so from completely cold to dangerously hot. When you consider the travel to the top of the gauge from the middle of the gauge where normal operating temperature is you have something that will travel only .75 inches or so. Suppose the stock needle on the stock gauge moves some 1/8th of an inch up from usual operating postion??? Well, so what? What would that temp be? What would the temp be at another 1/8th of an inch. Basically by doing this you are reverse engineering the "fix" that toyota made to prevent people panicing and pulling into the dealer to declare an overheat. YOU DONT WANT TO DO THIS!

2. No matter how you plug and play an OBD setup to show the true temps, you have some huge (laptop) or small (pda) display full of all sorts of settings that you have to refer to to tell true temps. YOU DONT WANT TO DO THIS WHILE DRIVING! Sure, an OBD setup would be sweet, and I still hope to have one, BUT, short of recording an entire driving event and reviewing it later, you cannot safely repeatedly refer to some such display while driving. The OBD option would be wonderful with willing copilots to tell the temps, and, it would be wonderful to review an entire driving event at a later time, and, it would be wonderful to determine the capability of cooling systems in general in garage while parked, but again, not while driving.

3. Any attempt at fixing some sensor to the block or the head or the rad is going to give false data; unless the sensor is in the coolant, you do not know the temp of the coolant, period.

With these things in mind, my own opinion is that the only option for telling the true temps is with wide range accurate aftermarket meters such as the Greddy Gauge I use, the ISSPRO Dan and others use, the Blitz that Reffug uses, or with the HKS plug and play temp thing I have posted here before. You really need something simple, and something setup right in regular view. If you have to pull the eyes from the road you recklessly risk real problems.

Again, I hope not to offend any at all, its just one of those things where the incredible collective intelligence here is solving a problem (s***ty stock gauge) that should NOT be solved. Get a good aftermarket gauge, install it in an easily seen spot, and call it good. Trust me, this modification done this way will easily end up as the absolute most "peace of mind" modification you make. HTH

 

[turbocruiser]

A pic of sensor ...

 

[turbocruiser]

A pic of gauge ...

 

[clownmidget]

that's hard to miss and nicely covers that annoying burned out "D" at the same time An unbeatabel mod!

I'm curious why you didn't just "T" into the coolant line you already have going over to the turbo? It still looks slick and perfectly fine where it is.

 

[cruiserdan]

Did you see how much gas he has left? That's gotta be at least 50 bucks worth..

 

[turbocruiser]

that's hard to miss and nicely covers that annoying burned out "D" at the same time An unbeatabel mod!

I'm curious why you didn't just "T" into the coolant line you already have going over to the turbo? It still looks slick and perfectly fine where it is.

Thanks, yes it is indeed hard to miss and only covers over the "D" and part of the "CRUISE" light when cruise control is on. It completely avoids covering any other idiot light which is wonderful cause I need all of em!!! The picture perspective is weird though; since I had to shoot the camera through the top of the open steering wheel section, it makes the gauge look a lot bigger than it is, but we basically get the idea.

As far as "T" ing into the coolant line to the turbo, to be truthful, I just did not think of it, but since you did, and now since I am, the probe tip is pretty long and I am thinking that the thing should sit immersed in as much coolant as possible - finding an adaptor to do that might be more difficult and because the basic Greddy adaptor was so simple, I just went with that. Also, I really wanted to know the temp that the coolant was JUST before returning to the rad so that spot made the most sense to me.

Anyways, thanks again.

 

[turbocruiser]

Did you see how much gas he has left? That's gotta be at least 50 bucks worth..

Yea, I always run with alota gas!

 

[Rich]

I don't think extreme precision is required from the coolant temp gauge. Here is all I really need to know 1) is the engine warmed up and running at normal operating temp, 2) is it running hotter than normal, but at a fixed temp and not increasing, 3) is it running hotter than normal and continuing to increase, approaching too hot, and 4) is it running too hot.

The factory gauge omits displaying category 2 and 3 info listed above. If the display were more linear, even without degree markings, I would have all the info I need. I would do a one time calibration to determine where "too hot" begins on the gauge.

 

[Rich]

Regarding the temp chart I posted a long time ago, which I believe Raven refers to: I did the test in a water bath, which is why the cutoff temp is below 212. I used an uncalibrated lab thermometer mark in single degrees C, and good quality Fluke multimeter last calibrated in the previous century. I recorded and plotted the change in resistence for each degree centigrade change in bath temperature, stirring the water bath constantly with the thermometer.

It is possible that the water vapor on the sensor skewed the resistence measurements.