Coolant Temperature Sender

Scorpio / Scorpio Cosworth Discussions - Questions, problem resolution, general talk, technical tips and modifications.
User avatar
Bob Weir
Membership Coordinator
Posts: 953
Joined: Tue Oct 14, 2003 6:49 pm
Location: Allentown PA

Re: Coolant Temperature Sender

Post by Bob Weir » Mon Sep 09, 2019 8:49 am

Have you used an infrared thermometer to read the temps at various locations?

Reference https://blog.thermoworks.com/tips/infrared-thermometry/

brokencase
Level 4
Level 4
Posts: 383
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: Coolant Temperature Sender

Post by brokencase » Tue Sep 10, 2019 11:42 pm

When you say the reading is on the "high side of normal" what exactly do you mean?

john keefe
Level 3
Level 3
Posts: 250
Joined: Sat Sep 02, 2006 1:13 pm

Re: Coolant Temperature Sender

Post by john keefe » Wed Sep 11, 2019 1:46 am

Bob, I'd love to have one, but on a limited budget.

Brokencase, perhaps I should have written, "... on the high side, above NORMAL." The gauge had always registered between R & M in the letters "NORMAL" before the sensor disintegrated. Now its above the N, just touching or under the lower of the two hash marks indicating very hot (overheating).

brokencase
Level 4
Level 4
Posts: 383
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: Coolant Temperature Sender

Post by brokencase » Fri Sep 13, 2019 9:01 pm

John - You need to confirm coolant temp by another means. You could be running hot.

john keefe
Level 3
Level 3
Posts: 250
Joined: Sat Sep 02, 2006 1:13 pm

Re: Coolant Temperature Sender

Post by john keefe » Sat Sep 14, 2019 12:22 pm

Ha! I thought those lazer thermometers were all in the $250+ Snap-On tool range. I know an HVAC contractor who spent more than that on his to evaluate windows, doors, etc.. Didn't even think about the "home use/cooking" $20 models, so I'll pick one up later today. Thanks for the advice.

john keefe
Level 3
Level 3
Posts: 250
Joined: Sat Sep 02, 2006 1:13 pm

Re: Coolant Temperature Sender

Post by john keefe » Sun Sep 15, 2019 9:35 pm

OK, thermometer reading is at 178deg., with coolant gauge needle at the edge of the "normal" box, just before it crosses over into the red. Since this is a dedicated circuit from the sender, and the gauge was working fine before, I conclude the Euro sensor is not calibrated to gauge.

So... again, would a resistor in-line before the gauge work to step down the signal voltage? Assuming its a 5V signal, what do you think the resistor should be to drop it 30% or so?

brokencase
Level 4
Level 4
Posts: 383
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: Coolant Temperature Sender

Post by brokencase » Sun Sep 22, 2019 12:22 am

It is not 5v to the temp gauge. It is 12 volts.

Basically you have 12 volts going into the gauge through the sensor to ground. As the temperature goes up the resistance in the sensor drops and the more current runs through the gauge and the needle deflects more.

It is more current than you would suspect. The manual states that the sensor should read 19 ohms hot and 250 cold.
I=V/R says that is about 0.65 amps in the hot state.

To do what you want your going to need a high power resistor.

But I'm not so sure your sensor is incorrectly calibrated. Defective maybe, or perhaps there is something wrong with the gauge.
I find it hard to believe that they would purposely make a sensor that would go less than 19 ohms.

I say this because "19 ohms hot and 250 cold" = GE55 thermistor material.
It's not like they had a lot of options to choose from, especially at this low resistance level.

john keefe
Level 3
Level 3
Posts: 250
Joined: Sat Sep 02, 2006 1:13 pm

Re: Coolant Temperature Sender

Post by john keefe » Sun Sep 22, 2019 12:57 pm

Yeah, I'd not expect Ford would spec a different sensor when the (then) system worked well. Pennies per car adds up to a lot of $$$ cost in a production run.

My second suspect is the gauge, but it was reading OK prior to the sensor stalk cracking off. Would be a heckuva coincidence if both had gone out simultaneously. Unless it was something related to a voltage spike which damaged the gauge when the sensor broke?

IIRC, I'll need two resistors in the circuit. The 12V lead would Tee between the two resistors, one resistor going to ground, and the other going to the gauge.

User avatar
andyofcolumbusmerkur
Level 4
Level 4
Posts: 325
Joined: Mon Jul 15, 2013 7:58 pm

Re: Coolant Temperature Sender

Post by andyofcolumbusmerkur » Thu Sep 26, 2019 8:40 pm

I don't know. I put a replacement laptop keyboard in a dell inspiron 15r 5520. The only way to get the correct version of keyboard, a few different oem vendors/suppliers it seems, was to enter the service tag number though the database beforehand. Just getting one from the exact same model wasn't enough. So it didn't even power up let alone work correctly. I can't say who is correct or anything but wouldn't doubt an international company like ford would have various parts with different design parameters yet along tolerances. So spend the money for the new nos part or be like me and be happy that the temp gauge works correct in 3 out of 5 vehicles...and that ain't bad.
You're only as good as the last car you build.

john keefe
Level 3
Level 3
Posts: 250
Joined: Sat Sep 02, 2006 1:13 pm

Re: Coolant Temperature Sender

Post by john keefe » Fri Sep 27, 2019 1:18 pm

Thanks Andy,
The problem is that I did purchase 2 new Motorcraft sensors, and they're both reading the same. They happen to be green ring ones from Europe, as I found nothing in the U.S. correct for this 2.9L. The U.S. listed sensors all seem to be the smaller, 5.0L or Bronco types.

As far as I could research, it appears all the differently colored rings are just Euro manufacturer indicators. They are supposed all the same for this (2.9L) application. Unless someone has some secret source of NOS U.S. Motorcraft sensors for this specific 2.9L that's not going to be correspondingly highly over-priced based on scarcity , then I can live with what I have.

But, I'd rather that the needle settles in the middle of the "Normal" range, than on the high side. Hence, the consideration for the twin resistor.

my8950
Level 5
Level 5
Posts: 784
Joined: Wed Jan 19, 2005 11:07 pm
Location: Detroit

Re: Coolant Temperature Sender

Post by my8950 » Fri Sep 27, 2019 1:22 pm

john keefe wrote:
Fri Sep 27, 2019 1:18 pm
Thanks Andy,
The problem is that I did purchase 2 new Motorcraft sensors, and they're both reading the same. They happen to be green ring ones from Europe, as I found nothing in the U.S. correct for this 2.9L. The U.S. listed sensors all seem to be the smaller, 5.0L or Bronco types.

As far as I could research, it appears all the differently colored rings are just Euro manufacturer indicators. They are supposed all the same for this (2.9L) application. Unless someone has some secret source of NOS U.S. Motorcraft sensors for this specific 2.9L that's not going to be correspondingly highly over-priced based on scarcity , then I can live with what I have.

But, I'd rather that the needle settles in the middle of the "Normal" range, than on the high side. Hence, the consideration for the twin resistor.
Maybe some superglue and place the needle where you like?

User avatar
andyofcolumbusmerkur
Level 4
Level 4
Posts: 325
Joined: Mon Jul 15, 2013 7:58 pm

Re: Coolant Temperature Sender

Post by andyofcolumbusmerkur » Sun Sep 29, 2019 11:46 am

So red ring is pretty accurate but green one is possibly good also? I'm not sure why I read though this entire post I don't even have a Scorpio. Seems like Ford engineers favored high current draw, higher voltage, and something measuring resistance at expense of sender life. I think you guys may be onto something with the r=V/I or is it r=l/V? You could use the principle to change what any cluster "sees" in case you can't calibrate the gauge and can not find an out of production sender. I have decided to add some more auto-meter parts to both Merkurs since I'm tired of gauges lying to me or not moving at all. And possibly super glue idea.
You're only as good as the last car you build.

john keefe
Level 3
Level 3
Posts: 250
Joined: Sat Sep 02, 2006 1:13 pm

Re: Coolant Temperature Sender

Post by john keefe » Wed Oct 02, 2019 12:35 pm

I guess superglueing the needle would be the easiest solution.

For my 5.0XR, I installed a few Autometer gauges for accuracy. Centered the tach here, and installed LED bulbs for the turn signal blinker as it ends up getting kinda' blocked on the right side, not to mention the stock bulb is too weak to grab your attention in that position:


Have had this RS500 RHD cluster for a while, but don't want to cut it up to make a similar center tach setup. Really nice condition, though probably best suited for someone with dyslexia... If anyone's interested, I could put it in the XR for sale section, or you can PM me. Don't know why it's not posting the pic like with the centered tach,
KIMG0207.JPG
RS500 RHD Cluster
(717.13 KiB) Not downloaded yet
.
Attachments
012A5270.JPG
Centered Tach
012A5270.JPG (596.92 KiB) Viewed 872 times

dimswits
Level 3
Level 3
Posts: 236
Joined: Wed Jan 07, 2004 8:29 pm

Re: Coolant Temperature Sender

Post by dimswits » Wed Nov 13, 2019 4:12 pm

Guys,

The Merkurs use gauges manufactured by VDO. Many VDO gauges of the era used sensors with the same profile, that is ~250 ohms at the "bottom" end of whatever scale (like cold water/oil or empty gas tank), and ~20 ohms at the top end of the scale. The coolant temp and fuel fill gauges in both Merkurs fit this profile. It also happens that this is VDOs aftermarket profile for 250 degree gauges.

I would bet that the VDO 323900 sender (OEM No 323-801-005-001D) would drive either merkur temp gauge perfectly. This sensor is about $25 and readily available. In the scorpio it will screw right in, but require a modification to the wire. In the XR it would need a 1/8 to 1/4 NPT bushing and the same wire mod.

The curve for this sensor is on page 10 of this document: https://www.vdo-gauges.com/media/instru ... ersion.pdf

I no longer have a stock Scorpio to run this experiment, but I bet it works.
Dimitri in NYC
88 Scorpio 2.5 Turbo
89 Scorpio 4.6 DOHC
02 E46 330Ci
17 MKC

brokencase
Level 4
Level 4
Posts: 383
Joined: Sun Dec 06, 2015 12:18 pm
Location: PA

Re: Coolant Temperature Sender

Post by brokencase » Wed Nov 13, 2019 10:46 pm

Exactly Dimitri. Like I said above..
"I say this because "19 ohms hot and 250 cold" = GE55 thermistor material.
It's not like they had a lot of options to choose from, especially at this low resistance level."

If the gauge is off it is because it is old and has drifted and needs to be tweaked or there is loose wiring/connection somewhere.

Post Reply