Aaron check with Jim DiGorgio in the New England region. Jim did a complete restoration on his 34.

Thanks Ed. I did send Jim an email. The place where it mounts appears to have the same pattern as the sending unit. My old water sending unit was just in pieces. I will probably block it off for now but I thought it was interesting how similar it was to the fuel tank.

Aaron,

Jim and I worked on his gauges, and managed to get his radiator water level sending unit working intermittently. Al Brintnall said that he had only been able to get occasional operation of the '34 Hudson water level sending unit in the car he owned and drove for many years. I have never attempted to get my '35 Hudson unit working.

Per

Per, any idea on the ohm range for the water level sending unit?

Aaron, these were bi-metal, not resistance units. They worked on the temperature of the water affecting the time the bi-metal contacts took to open and close. A resistance my work, but not reliably. The resistance of the later model senders ('51 onwards) was 10-90- ohms.

Aaron,

I suggest trying a rheostat with the resistance range Geoff mentions. These make-and-break senders operate a heavily damped gauge. Heavily damped means "slow to respond". That means the gauge is responding to the percentage of time the make-and-break is in the "make" position. You can achieve the same result if you can devise a rheostat similar to what is in a more modern gas tank sender. This is to attempt to read the water level in the top tank. You will need to install a series resistor so the gauge doesn't burn out when the rheostat is at the zero end of its operating range.

However, you might prefer to measure the temperature of the water in the top tank. In this case, make an adaptor plate to mount a temperature sending unit like what is in 1951 stepdowns in the top tank. In this case you should add a low-ohm resistor so the gauge doesn't go off scale. You can get the system to give a reliable reading, but the exact reading will vary depending on whether the car is charging or idling. I have such a setup in my '49, which has the same type of dash gauge because the '49 engine had a make-and-break temperature sender. However I have a '51 engine in the car. I found that a 20 ohm resister connected to the sender works well, and gives me a usable, reliable indication (although the gauge reads backwards) as I drive down the highway. I have gone about 20,000 miles since installing this.

Per

Per

I believe the water level sending unit mounted in radiator, as used in the U.S. , is a float type.
It is interesting to hear of bi-metal use in the 30's in other countries.

Yes, it is a float type, but still uses bi-metal. The float lever just alters the tension on the bi-metal strip.

The fuel and water sending units are identical except for the length and orientation of the float rod. There’s an excellent write-up on operation of these units in the January 1935 Terraplane Hudson Service Bulletin, available on Alex Burr’s website.

That article states the average current draw at “empty” and “full”. A simple Ohms Law calculation indicates a 150 to 30 ohm replacement unit will work.

Thank you all,, just got out to barn to look at one, the float is what I recalled, fun to learn and be refreshed.

Aaron: these are some photos of the sending unit needed for the 1934 H8 radiator!

Here is an example of the water sending unit.

Hii had a nòs 1àñď sòlď ita while ago came out òf a collection of parts ì got at an aùtìòñ ì think ì hàve1mòŕè ñòs complete èçèpt çòrķ fĺòat missing