Ahem.
Thermoswitch.

 

LOL fair enough on the debate on semantics...

What I don't understand is why it would be shutting out on high, but not on low? Could it be something along the lines of ground loop feedback? I am fighting other electrical problems, which I'm going to tear into tomorrow morning. I've apparently lost some stereo amps, possibly due to ground loop problem. I found the groundloop at the same time that I found my sub amp was not working and my component amp was only working on one side of the car.

Since finding the groundloop in the sound system I disconnected the headunit and all the amps. None of them have been getting any power. Tomorrow morning I'm going to physically remove all my amps and get them out of my electrical system all together.

 

Hmmm.
Where is your thermoswitch located?

 

It's at the OEM location for turbos, on the back of the water pump housing. It's a Starion switch that matches the OEM threads.

 

by "back of water pump housing" i hope you mean, in the neck, as that is the spot for them

 

Yes, there is the thermosensor which sits higher and closer to the thermostat and the OEM thermoswitch which sits a little lower and closer to the engine keg. I'm using the lower and chose the Starion switch because it matches the threads.

My aftermarket gauge thermosensor is a port I had machined on the front of the water pump housing and through one of the air pump bracket threads.

 

It's working as designed, just not exactly as you expected.


When it's on high speed, the fan removes more heat, thus lowering the temp at the sensor to the off temperature.

On low, the temperature does not quite reach that off point, despite what your gauge reads.

My temp sensor is 195 on 180 off. The low speed almost never turns it off even on pretty cool weather. High speed will turn it off occasionally in cool weather. In my case the low shut off temp is too low and the thermostat is nearly closed while the fan is still running.


You could test this by backprobing the sensor to relay connection. When the sensor to relay voltage changes to zero, the fan will shut off.

Try this at both fan speeds and I think you will find it's working fine.

Good Luck!

 

We have similarly specced thermoswitches but my experience is exactly the opposite of yours.
Below @65° (and on the highway in any ambient temp), my fan never comes on.
Recently, in city driving and warmer temps, it's come on a few times but only briefly.
Over the course of a year, I'd estimate the fan is active maybe 10% of all drive time.

 

Check for Resistance at every link in the whole chain. Battery to fuse, fuse to relay, relay to fan, etc. If they all check out around 0.1 ohms, they're good. When my fan was cutting out, it was because I had used the stock Taurus circuit breakers as a short extension. After running or a couple minutes, the fan shut off like clockwork, regardless of coolant temperature. So I took it out and it solved the problem. By your description, it almost sounds like it is shorting out BEFORE the fuse. If so, a cheap way to correct it is to put the cable in some 1/4" split loom.

I'm not sure what LRA and FLA mean specifically, but the majority of us use "Inrush current draw" and "constant current draw" to describe these parameters. By your numbers, the fan motor is perfectly healthy. Also, check your grounds again. Funky behavior like this usually indicates a sketchy ground. For my setup, I have it grounded to a dedicated bus bar which goes down to the Main Engine Ground via some 4awg cable.

 

My aftermarket (Prosport) temp gauge sensor/sender is on the front side of the waterpump housing. My temp switch on on the backside of the water pump housing. The two practically touch and last night my fan was switching off before my temp gauge showed any drop in coolant temp.

Though something did come to me in a dream last night..and Clokker may be on to something with the thermoswitch being bad. I have tested manual switch on low and high. I'll drive around on low and it will stay on for as long as I have it toggled on without bringing temps down. It stayed on with high speed and manually switched on for 5 min. It turned off on it's own while on high speed with the thermoswitch signal before I saw temps drop on my gauge. I have not allowed the thermoswitch to switch the fan low speed on.

I need to head to the store for more spade terminals so that I can re-connect low speed through the maxi fuse and simply let the thermoswitch do it's thing on low speed.

 

LRA = Locked Rotor Amperage
It's terminology in HVAC and I assume means the same as "inrush current draw." It's how many amps a motor pulls when first starting up, as it takes more amps to get a motor moving than it does to keep a motor moving.

FLA = Full Load Amperage
Another HVAC term and I assume similar to constant current draw.

If you look on your home's furnace and a/c unit outside you'll see a tag telling you how many LRA and FLA your blower, compressor, and condensor fan should be pulling. Though some older units list something different for FLA...um...RLA IIRC. Not sure what RLA means, but it's the same numbers as FLA.

A couple things I thought of while getting my breakfast on, is it possible the high speed winding on my motor are failing and shorting out? I don't know enough about DC motors to test that.

As far as grounds, I was going to look at that. IIRC the ground for the fan is an 8ga to match the high speed 8ga

 

Makes sense now. The way I learned it was from Aaron Cake, hence Inrush and Constant Draw. But since they describe the same thing as LRA & FLA, it's all good.

8awg ground is correct, mine is wired the same way and has no glitches, even with the overstuffed ring terminals. Perhaps it would help if you grounded it to the main Engine Ground location to ensure it has a good ground.
Just to be safe, check your battery voltage with the fan running. If it's dipping under 12v, it won't be able to keep the relay coil energized, which makes it 'fipple' between on and off quickly (and makes annoying noise too).

To test the high speed winding, it's as simple as hotwiring the fan straight to the battery or another suitable 12v source. That's how I test electrical devices with an ATX power supply modified with four binding posts.

 

Here is what I have found

If the thermoswitch switches the fan on then the fan will switch off 1-2 min after being switched on. This is true on both low and high speeds. If I manually turn the fan on it will stay running seemingly indefinitely regardless of low or high speed.

Akagi, I'm curious where the .1 ohm resistance came from for your setup. All my resistances for every link of my fan wiring was .5-.7 ohms. I checked the brand new 8ga wire left over from rewiring the maxi fuse and it reads .6 ohms across it. I've checked grounds in the past and they read around .5 ohms.

The only portion of my wiring I could not check was my thermoswitch to relay (86 IIRC). To get to it I have to remove the UIM and alternator and I wasn't in a position to do that today.

One thing I'm curious about regarding my thermoswitch and manual switch. I have the manual switch and thermoswitch spliced together and then going to 86 on the relay. The manual switch is mounted under my steering column and grounded at the driver kick panel. Can this be the cause of my curious fan problems? Should I have two separate wires that meet at pin 86 on the relay?

My plan of action is to get my hands on a different thermoswitch. I'm not terribly happy with the fan coming on at 205 and would prefer a 195 on. Plan is to take the existing Starion thermoswitch off and test it. Source a replacement regardless of the shape of the Starion piece and have it on hand when I remove all the necessary parts to get to my thermoswitch BUT it might be a couple weeks before I can do that. Hopefully I find what Clokker was suggesting, I have a bad thermoswitch....though the one on there has less than 3k miles on it from when I bought it new.

 

For comfirmation, S5 switches and the Starion are normally open, correct? If I get a two prong thermoswitch that simply means one goes to relay the other goes to ground?

 

Sounds right, yes.

 

The standard I use for electrical connectivity is 0.07 ohms, but I let 0.10 ohms pass as well. It sounds like you have corroded connections as evidenced by your resistance readings. Clean every connection with sandpaper and a wire wheel untill you see bare shiny metal, then give each exposed connection some dielectric grease and torque them down so it gets squeezed out and covers the connection to protect it. Your manual override switch is in parallel to the thermoswitch, so the relay coil can get a ground from the Thermoswitch OR the Manual Override Switch to complete the circuit and turn the fan on. Any electrical funkiness on the MOS can't affect the Thermoswitch this way.

Check the resistance between your negative cable at the starter, main engine ground and negative terminal. If they're above 0.10 ohms, clean them untill you see bare shiny metal as Aaron Cake mentions here:
Proper RX-7 Grounding Procedures

Voltage Drop= Current x Resistance (in ohms)
Assuming it takes 1 amp to trigger the relay coil, 0.5 ohms means that the circuit only gets 11.5 volts. But since we know the current draw of the fan motor (20A on low, 30A on high for this ballpark example), we can apply Watt's Law to tell how much current is being drawn at the reduced voltage

Watts = Amps x Volts
20 x 12= 240 Watts being drawn at 12v, so divide by 11.5 to account for the voltage drop we calculated above.
My math shows 20.87 amps for low, and 31.3 amps on high when running at 11.5v

The above math assumes that there is only 0.5 ohms resistance in the ENTIRE circuit. If each link in the circuit has 0.5-0.7 ohms, add it all up, then apply Ohm's and Watt's Laws as described above.

Standard Brand TS-84, intended for a 88-91 Honda CRX 1.5L or 1.6L SOHC is the thermoswitch you seek. It cross-references to a bunch of other Hondas, such as a 77 Civic, so they're not TOO difficult to find. Same M16x1.5 threads stock, 195 degree ON temperature (off at 185) and uses common bullet connectors. It has two terminals, so link the second one to a ground spot and you should be golden. Alternately, since it has two terminals, you could use it on the positive side of the relay coil too

 

I know the math on Ohms Law, I've had a handful of electrical classes for HVAC. I'd have to check with one of my instructors, but I don't believe you add all the resistance for all the links of wire connecting the fan. All the wire is considered to have a very small resistance. So for instance an 8ga stranded copper wire will be rated for "x" amount of amps up to "x" amount of distance. If you go beyond the rated distance for your amperage draw you have to increase the wire size. If the loads (generally considered to be motors, electric heat exchangers, etc) are in series you take the sum off all the resistances. If the loads are in parallel you take the reciprocal of the resistances. So for parallel resistances are 1/R(total)= 1/R1+1/R2+1/R3+...(etc, etc)

About resistance, I've got a true RMS Fieldpiece for work (HS36) which retails for right around $300, though you can get them for less than that. I think I paid $225-250 at an HVAC supply shop. Regardless of the model and price of it, it's a pretty nice piece (though not a Fluke) which many people in my profession trust the same model meter to troubleshoot really expensive equipment. My $50 Craftsman DMM shows very similar resistances on the leads. I can put each DMM on brand new 4ga wire (bout 5' of it) and see virtually the same resistance on both meters. Maybe if I had a different meter I would see different numbers. Maybe my meter for work is designed for just that..HVAC and they aren't concerned about 1/10th of an ohm type resistance.

I'm curious what meter you are using for your numbers?

I ordered the thermoswitch you referenced, it will be here Tues...and not planned but coincedentally I took the day off work Weds to get a few things done in the AM. Maybe I'll have time to swap out and test a few more things.

 

i saw 28° at the center vent today. what was yours?

 

Dusty.

 

lol

 

I've just been using a common meter from Harbor Freight, nothing fancy but it does the job well. 0.7 ohms is the figure I got from doing a resistance test on a known good cable. Resistance in the body from one ground point to another just happens to be the same on my car, meaning that the connection is good.

Apparently I had read the meter wrong before. It said 0.7 ohms, not 0.07. Either way, resistance on a GOOD connection is always a tiny number as you said. Bad connections show more than 1 ohm.

Not knocking your training in HVAC, but try putting together 3 resistors in series and checking it with a meter. Now compare it with the reading of a single resistor...see a pattern here? Resistance is cumulative in any circuit

My gut feeling says you have a sketchy ground for the fan. Might try running an accessory bus bar from the main engine ground point after giving it a good scrubbing with the abrasive of your choice. I have all my underhood extras grounded this way and it works damn well. Did the same with anything needing power too. Plus, the two bus bars double ad the underhood jumpstart points as my battery is in the passenger side storage bin

Bit of advice: Try putting the main ground point's bolt in from the wheelwell side after you clean it up. then put the cables on, followed by a nut. Much easier and expandable where appropriate. A 20mm long bolt should be plenty here.

 

Good, good...we are seeing the same numbers on our meters then....I was trying to figure out how you were able to transcend electrical laws and see lower resistances on wire than I can. I thought maybe you had a precision piece of equipment that read lower than mine was capable of.

I think you are getting confused on resistance for the wire and actual resistors and other types of loads on a series circuit. You are correct that resistors in series are cumulative and summed up to find R(t) on the circuit. However, electrical code rates total wire length for resistance....that resistance is not cumulative as long as the wiring stays below a very very long length. Below is one of the code charts for wiring which shows Ohms ratings on different gauge of solid copper wire. Those ratings are for 1000 ft and a certain temperature.



I think you are getting confused on having different sections/links of wiring on the fan circuit. You do not add resistance for each link of the wire used like you do for actual loads (resistors, motors, heat exchangers, etc). So far as electrical current knows the wire is all one piece and merely providing a conductor and path to reach the loads and ground and thus complete the circuit.

Just to clarify though, resistance is absolutely not cumulative in any circuit. It is cumulative in an series circuit. However in a parallel circuit resistance is reciprocal, not cumulative. In a parallel circuit resistance will be slightly lower than the lowest resistor/load..hence the reciprocal.

To test this wire in some light bulbs in series. You will see the first light in the circuit is the brightest and the next bulb will be considerably more dim. That is because the first bulb is the first load and in a sense "soaked up" all the voltage it's load would allow. The remaining voltage continues on the next bulb, the remaining voltage from the second bulb will continue to the third. Each bulb in the circuit will dim cumulatively.

Wire those same bulbs in parallel and you will see an even brightness on the lights. This is because of the reciprocal nature of resistance in parallel where voltage remains the same and each bulb in the circuit sees the same voltage.

As far as my ground for the fan, I believe it is a good one. I'm using the factory spot at the driverside shock tower. I checked resistance from the fan to the ground and saw .5-.7ohms. Same as the wire.

Now, I may still have ground loops somewhere and that might be causing that thermoswitch to bug out. That's what I got to thinking over the last couple days. That if the switch is grounded to the engine but if I have a ground loop problem on the engine or chassis of the car then that's a red flag. But as far as that particular fan ground is concerned, it's a good one. I've sanded down the paint to bare metal, lithium greased it, etc. I've been going through all my grounds and checking them. I have a feeling I"m eventually going to find more ground loops or something shorting because I have other electrical problems.

1. Dome light hasn't worked in forever (no the bulb isn't burned out)
2. Blew a 40amp or so fuse on defrost fuse (which should only be like a 5 or 10 amp)
3. My Prosport water temp gauge occasionaly spontaneously switches from orange illumination when the lights are on, to it's default white illumination whenever it gets power
4. Also occasionally I will start my engine up and that same water temp gauge will zero itself out at 200ish. As it sees power it will slowly drop down and if i wait long enough it will drop to zero.

 

I'm picking up my thermoswitch tomorrow. Parts store lied to me, it was not in Tues and I haven't had time to go get it. When I wire up the new two prong thermoswitch, does it matter which goes to ground? Also I'm going to test it before I put it on. For testing on the stove with a pot of water do I simply put a lead of my DMM on each prong/pole, or does one need to be grounded...say to the stove? When testing the one I'm removing, which is a single prong/pole, do I simply put one of the leads on the thermoswitch threads or do I ground the lead to the stove?

 

There is no polarity, so you can run +12v to either pin of the sensor.

When testing in water you're just measuring continuity between the two sensor pins.
No grounds required for the test.

 

Doesn't matter which wire goes to what, it just connects the two terminals together at the preset temperature of 195 or whatever. This kind of thermoswitch can be on the positive or negative side of a circuit, so they're more flexible than a one-terminal thermoswitch. Mitsubishi did this on the 3000GT and Diamante.

Try hooking up a battery to one terminal like this:
Battery Negative to Negative probe
Battery positive to Thermoswitch terminal
Positive probe to the second Thermoswitch terminal

When it hits 195 degrees (or whatever the switch is rated), you'll see battery voltage register on the meter. In the car, connect the second thermoswitch terminal to a good ground.