View Full Version : 48 Willys Pickup - Correct Temperature Sensor

01-04-2016, 03:49 PM
Another mystery in the ongoing truck restoration saga.

About 1974 or so I replaced the original 134 L-head engine with a 134 F-head engine. Since that time, my temperature gauge has read backwards - a cold engine starts out at 212 and as it warms up, it goes to mid-scale, about 160 degrees.

In the restoration of Ham's 48 pickup, we have had a sequence of events that has left me dazed and confused with the temperature gauge.

In August 2015 I had the instrument cluster overhauled; all gauges checked and calibrated. When we put the engine back together, we ordered KWAS temperature switch P/N 905926 for "6-volt King Seeley gauges". That's what I need for the cluster on my truck - the gauges are King Seeleys. But - the temperature gauge still reads backwards.

So, let's look at this problem a little closely. The sensor in the engine (allegedly the 6 -volt, 905926 King Seely sensor) measures about 350 ohms at room temperature. In a cup of really hot water- not boiling but close, it measures about 10 ohms. The hotter it gets, the lower the resistance. Both the sensor currently in the engine and the one that had been in there since 1974 agree closely. Technically, these are both negative temperature coefficient sensors - probably the same part.

So, today I went to the truck with my box of fixed and variable resistors to try to figure out just what it took to make the gauge run correctly.

Switch OFF. Gauge reads just a little past full scale to the right, a little more than 212 degrees. Then, I took the wire off the sensor and replaced the sensor with a 100 ohm variable resistor. Not wanting to burn out anything, I started with the full 100-ohms in series with the temperature gauge. The needle moved hardly any, maybe 1/3 a needle width.

I carefully decreased the value of the series resistor and as I did, the temperature gauge read lower and lower.

A rough resistance vs. temperature gauge readings is

100 degrees 19.5 ohms
130 degrees 22.6 ohms
160 degrees 32.3 ohms
190 degrees 47 ohms

Hmmm .... these readings indicate that the temperature gauge needs a positive temperature coefficient sensor to make it read right. I wondered; "Do I have a bad gauge - did I damage it at some time in the dim dark past?"

So, I called Bob's Speedometer Service, the folks who overhauled the gauge cluster. I spoke to their gauge guru and told him the vehicle I was working on.

His first question "Is it the temperature gauge?"


His answer, "Your gauge should rest at 212 or above with the ignition OFF, then come down to colder readings when you turn on the ignition". OK, I hadn't damaged the gauge back in '74 or so.

He continued "The old King Seeley gauges on the late '40 Fords and the Willys gauges used a different sensor. It was about 10-ohms or so when it was cold and it went up in resistance as the temperature rose". That matches what I measured on my test setup.

As stated earlier, KWAS calls out the 905926 sensor for 6-volt, 46-58 Jeep and Willys with King Seely gauges. The package the sensor in the engine came from has a P/N 905926 label on it, so I can only assume that this is a 905926 sensor.

Is there anyone out there who can:

A: Verify or deny that this is really the right part for the 6-volt gauges, or

B: Measure the resistance of your properly working sensor to let me know what the resistance of a cold, correct, sensor should be, or

C: Give me the part number what you know to be the correct sensor.

Here are a couple of pictures of the sensor I have - the one that doesn't work with the gauges. The thread that screws into the head is a 1/2"-NPT as far as I can tell. (I don't know what the "240" stamped on the sensor hex means).

Thanks guys.

09-07-2018, 05:30 PM
I have the exact same issue. My original sender says KS with a circle around it and 2. Must be a King Seeley #2. It reads 12.93 kOhms cold and then the resistance drops as it gets hot. I ordered a new sender from KW and it's the same thing. They were super cool and refunded me when I called to tell them it wasn't the correct application! So, I too am looking for the correct sender that starts with little resistance and gains resistance as it heats up. I checked the gauge operation prior to ordering the sender and it functions as you explained and as I've read they should. Full hot key off, full cold key on engine cold then gradually moves toward hot as engine heats up. Mine is full hot key off, full hot key on and then moves from hot to cold as engine heats up.
Please let me know if you found something that worked for you!

06-09-2019, 08:25 PM
I hope someone who is knowledgeable on this topic will response to these comments, I too have the same issue(s).

06-10-2019, 08:06 AM
I hope someone who is knowledgeable on this topic will response to these comments, I too have the same issue(s).

I don't know if I am knowledgable or not, but here goes.

Temperature gauges on the '48 trucks, station wagons and sedan deliveries continue to be a source of frustration. There is confusion because many vendors of sensors will classify their sensors only as 6-volt or 12-volt sensors; “Works with 6-volt King-Seeley gauges” is a typical statement. The problem arises because there are two classes of 6-volt King-Seeley gauges, pre 1950 ˝ and post-1950 ˝.

This will get a little long, but bear with me. The King-Seeley gauge is a thermal gauge – it works a lot like an old fashioned thermostat for your furnace. Inside it has a bi-metal strip. That strip is heated by current flowing through it. That current is controlled by the sensor on the cylinder head that measures the water temperature. As the strip heats up, it bends and moves the needle on the gauge.

In the 1946 to 1950-˝ vehicles, the temperature gauge is actually mounted upside down. The fuel and oil pressure gauges are mounted right-side-up. As a result, as their sensors for gas level and oil pressure cause them to heat up, the needle moves upscale (Full or 50 PSI). The temperature gauge has to heat up to read colder!

The original sensor for the L-134 engine used in the early trucks operated a lot like a thermostat as well. It consisted of a set of contacts mounted on a bimetal strip wrapped in a heater wire. When things were cold, the contacts were closed; current flowed through the heater and the gauge. As the heater warmed the strip, it would bend and open the contacts. Now, while the contacts were closed, they allowed current to flow through the temperature gauge and move the needle. Since the gauge is upside down, as the gauge heated up, it moved to the 100-degree end of the scale. With a cold engine, the only heat source for the sensor was its own heater and it took a while for it to warm up enough to open the contacts.

But, as the engine warms up, the contacts are closed for a shorter period of time, the bi-metal strip in the gauge does not get as hot and the needle on the gauge drifts to the right, 212-degrees.

Then, in mid-1950 things changed. The electrical system in the vehicles stayed at 6-volts, but the F-134 engine was introduced and the instrument cluster styling changed from the square cluster to the “clock and bar” cluster. On the cluster, the temperature gauge became a “right-side-up” gauge and the sensor on the engine changed from the switch and contact sensor to a temperature sensitive resistor (thermistor) sensor. The sensor in the F-134 is also much smaller – you can’t mount an L-134 sensor in an F-134 block. With the new gauge orientation, things operated backwards. As the key was turned on, the temperature gauge no longer swept across the dial from right to left, it just wiggled a little off the left side 100-degree mark.

My truck has had the L-134 replaced with an F-134 – the gauge read backwards - and after looking for several years, here is what I found:

A. If you have an early vehicle with the L-134 flathead engine, there are temperature sensors for early Ford V-8’s that might (and I say might) kind of work. They claim to be the thermostat/switch style of sensor because the early Fords also used an upside down gauge. These sensors should at least move the pointer in the correct direction, but I have no idea if they read anywhere close to correct as far as temperature goes. This might be a good solution, I don’t know.

B. Kaiser-Willys has an adapter that uses the new sensors and makes the old gauges read correctly:


There is a lot of information in the description of this adapter that talks about the King-Seeley gauges as well. Using a new sensor in the L-134 with this adapter would be a solution, but if you have a 12-volt to 6-volt voltage reducer, it needs to be a voltage regulator, not one of the on-off switched reducers.

Other stuff:

Sensor Resistances:

212 degrees 15 ohms
185 degrees 22 ohms
160 degrees 35 ohms
100 degrees 140 ohms -- Below 100 degrees the resistance rises very rapidly.

If the gauge doesn’t move with the switch on, check for a defective sensor , a broken wire or a defective gauge. A temperature gauge should measure about 12-ohms across its terminals.

06-11-2019, 09:46 PM
Thank you, I now have a path forward that I can work too. I am sure there are still challenges and issues ahead but the breath and scope of those issues have been narrowed considerable. No matter the outcome I find this forum helpful and always direct in solving challenges that come up with any rebuild project.

06-12-2019, 10:13 AM
A quick check to see if the gauge is working is:

A. Make sure that there is power applied to one terminal of the gauge

B. With a clip or test lead attached to the other terminal, tap the other end of the lead to ground (chassis). Tap ... tap... tap - don't hook it up directly - too much stress. If the gauge is good, after several taps it will start to move to the cold side. If it's open - not unusual from overvoltage or corrosion, it can be fixed. Bob's Speedometer fixes these gauges - check their website.

07-17-2019, 03:24 PM
My meter is fine, it was the King Seeley temperature sender. I found a Ford sender, part number 01A-10990. It's for a 1930 to 1950 Ford six volt. I am pleased with it. My thermostat is for 180 degrees. Once it opens and my meter stays around 180. That close enough for me. With the bad sender the meter would just peg pass 212, that was not good. With my Ford sender it does not peg. Some say Ford used King Seeley senders back then, some say Ford made and used their own, no matter my new sender works with my King Seeley temp meter. Life is good. Now I am off to address my fuel gauge "issue."