33 PD fuel gauge problem

[Pandanom]

OK, I installed the sending unit I bought from Robert's Motors and it worked perfectly! It fits 1928-1938 and cost me 85.00.

Jim

[Pandanom]

Robert's Auto Parts sells a sending unit for 1928-1938. I ordered one today and will let you know if and how it works. Has anyone else tried it?

Jim

[TodFitch]

Huh?? I'll need to see the math on how the 390 ohm resistor fixes this.? As to "empty", there is still gas in the tank when the sender float will no longer float, just an observation.? Best of luck, I throughly enjoy these conversations as my car is still in pieces!? Just look for the thread on temp gauges for proof!? ?

See http://en.wikipedia.org/wiki/Resistor#Series_and_parallel_circuits for how to calculate the equivalent resistance of two resistors in parallel.

For the empty case: 180*390/(180+390) = 123.16 ohms.

For the full case: 10*390/(10+390) = 9.75 ohms.

[notdavidspade]

Huh?  I'll need to see the math on how the 390 ohm resistor fixes this.  As to "empty", there is still gas in the tank when the sender float will no longer float, just an observation.  Best of luck, I throughly enjoy these conversations as my car is still in pieces!  Just look for the thread on temp gauges for proof!   

[TodFitch]

Sigh. Now that I have the sending unit in and a full tank of gas I see that I can get a universal style sending unit similar to the one I used but with a 180 to 10 ohm range designed for systems using VDO gauges:

http://www.egauges.com/vdo_indS.asp?Sender=10_180ohms&PN=226-001

or

http://www.egauges.com/vdo_indS.asp?Sender=10_180ohms&PN=226-002

With a 390 ohm resistor in parallel with that it would read between 123 ohms (empty) and 10 ohms (full) which would be pretty close to what the car really wants. Oh well. Next time. . .

[TodFitch]

I don't know if it will help, but my sending unit measured 128 ohms empty and less than 3 ohms full.

Thanks for the capacitor tip, I'll have to check that out.

I went back to my notes and my empty reading on the dash gauge unit on my PD was actually 120 ohm while the full was 8 ohms. So much for my memory. It would seem that your 1935 and my 1933 use the same range of values which makes me think those could have been the standard values (at least for Plymouth) prior to the introduction of the two wire thermostatic type gauge. We can summarize:

Comment	Empty	Full
JimCno's 1935 PJ Sending Unit	128	3
Tod's 1933 PD Dash unit	120	8
Designed Range (Best Guess)	120 to 130 ohms	3 to 5 ohms

For what it is worth, here is the data I took by hooking various resistors onto my dash gauge (with the sending unit wire detatched):

Ohms	Reading
120	Empty
94	1/8
60.5	1/4
47	1/3
37	1/2
23.5	3/4
16	7/8
8	Full

The unit I just installed does not have sufficient range to indicate both full and empty. I can have it accurate a one point on the scale or another but not at all readings. I think I will go ahead and add a 33 ohms resister in series with the sender and make the range be 123 ohms at empty and 33 ohms at full. Having it only show 3/4 when full does not bother me too much. Having it show 1/8 tank when it is dry does bother me. :)

Interesting thought about adding a capacitor to damp out the variations as gas sloshes around the tank. I don't think I'll do that as I am used to the needle wandering about and I know that as long as it moves a little like that there must be some gas in the tank. When the needle is totally stationary near empty I know it is time to refill.

[JimCno]

I don't know if it will help, but my sending unit measured 128 ohms empty and less than 3 ohms full.

Thanks for the capacitor tip, I'll have to check that out.

[TodFitch]

With a number of 1/2 watt resistors put in series and/or parallel between the tank side of the gauge and ground I have determined that my PD gauge reads full with between 0 and 10 ohms and empty at about 100 to 110 ohms. There is some stickiness in my dash unit and I can only get a limited number of resistance values with the parts I have on hand, so those numbers are approximate.

The fuel gauge sending unit uses a standard five hole mounting pattern.

Looking around it seems that recent (since about the 1950s) Chrysler and Ford sending units range from 78 ohms (empty) to 10 ohms (full), Stewart-Warner sending units range from 240 ohm empty to 33 ohm full. The closest match in terms of range is the GM unit (0 ohms empty, 90 ohms full) but it would read backwards.

Of course you can send big bucks to Atwater-Kent and get a unit that is claimed to be duplicate of original. But this is an item that can't be seen once installed, so why spend the big bucks if there is a cheaper way.

Initially I rejected using the GM unit as it would read backward. However an article in my local region's newsletter by Bob Amos noted that the JC Whitney universal fit sending units for Chrysler could have the resistance unit swapped (he was working on a 1950s car that needs the 78 to 10 ohm range). So I took a look at http://www.jcwhitney.com/autoparts/Product/Pr-p_Product.CATENTRY_ID:2005714/c-10101/Nty-1/p-2005714/Ntx-mode+matchallpartial/N-10101/tf-Browse/s-10101/Ntk-AllTextSearchGroup?Ntt=FUEL+LEVEL+FLOAT and decided to give the GM unit a try.

I figured that I could put a 10 ohm resistor in series with the unit to change the range to 100 empty, 10 full if needed but the first step was to simply install it with the resistance unit flipped vertically.

I had, by measuring with a stick, about 2.5 inches of gas when I dropped the tank to replace the sending unit. Given the dimensions of the tank, that is about 3 gallons of gas. When installed the dash gauge read about 1/8 tank. Just took the car and filled it to the brim. As full as it can go. Again measured the depth with a stick and found 8 inches of fuel. Which should be 15.1 gallons by my calculations (15 gallons by the manual). It took 12.5 gallons to fill it, so there is a slight error in the measurements somewhere. But the dash needle is sitting nicely on full at the moment.

I'll measure the fuel depth at various gauge readings and see how the dash unit responds with this sender but so far it is looking good. I'd rather have empty be accurate than have full be accurate, so I may add a 10 or 22 ohm resistor in series later.

I don't know the sender characteristics of other pre-War Plymouths so the above may only work for the PD. And it certainly will not work for the two wire sending units they started using in the late 1930s through the P-15 era. That said, there is the possibility it will work for the single wire sender applications prior to the introduction of the two wire thermostatic gauges.

[FourDoor]

If you tack a minature 10 mfd./50 volt electrolytic capacitor from the gas gauge to ground on those old one wire sytems you can slow down those violent swings. Put it on the side coming from the gas tank. Since these old Plymouths are POSITIVE ground you will put the lead of the cap with the 'line' on it to the gas gauge or the lead with the + sign to ground/chassis/body.

[JimCno]

I dropped the tank of my '35 this past winter. While it was off I took the sending unit apart, cleaned up all the components (it's a very simple process) and then put it back together. My fuel gauge now works though it has a tendency to jump around as the fuel is sloshing about. I might be wrong, but I assume that is because it is the old one wire system.

[TodFitch]

I am going through a issue with my fuel gauge at this time and have a experiment on a new sending unit I will be trying in the next week or two (really busy and I have about 1/2 tank of gas in the car that I want to run down before I drop the tank).

If you disconnect the wire to the tank from the dash unit you can hook up a jumper and test the dash unit. My dash unit reads full with about 8 Ohms resistance to ground. And it reads empty with about 100 Ohms of resistance. Those are not standard ranges for any modern universal sender, but the good news is that the 5 hole pattern on the sending unit is a standard. I'll get back to you when I see if my experiment works.

In the mean time, you can pick up some resisters from a place like Radio Shack and hook them up to a jumper connected to the tank side of the dash unit and see if the needle goes through an appropriate range when you ground the other end of the resisters. If so then your dash unit is okay.

As pointed out by an earlier reply a very frequent issue is with the tank not being properly bonded to the frame (bad ground). With a volt-ohm meter you can check the resistance between the tank sending unit and a good clean ground someplace on the frame. On my two door sedan you can just reach the sending unit when the tank is installed by reaching up over the rear axle. I assume other body styles will be the same.

While you are under the car you can check the resistance of the wire between the dash unit and the tank unit. If you have any reading at all that will probably be okay.

The last element is the tank sending unit. But you have to drop the tank to get that out which means having a relatively empty tank of gas (this is where I am at). Based on a write up in my local region newsletter I am going to see if I can modify a relatively inexpensive ($36) modern sending unit to work. The alternative is to order a reproduction gauge for big $ from Atwater-Kent. I will let you know how mine turns out when I get it done in the next week or two.

[33plycoupe]

Pandanom,
Thanks for responding. I will try the jumper wire suggestion. NO, I haven' had the sending unit out of the car. If the jumper wire doesn't work, is there a possible fix at the sending unit? Thanks again for the info.

[Pandanom]

You could do a little troubleshooting for a bad ground while the tank is full and the gage is reading less than full.

Make a jumper wire out of #14 wire and some alligator clips. Install the temporary jumper from the gas tank to the frame and/or the body of the car and see if the gage changes. Try it too on one of the screws holding the sending unit in .

Have you had the sending unit out of the tank?

Good Luck,

Jim

[33plycoupe]

The fuel gauge on my 33 PD registers a little over a half  tank after I filled the tank 5 gals at a time. (I took out a second mortgage on my home for the gas.) Is the problem in the gauge or perhaps at the tank. The gauge in the car is one I picked up on EBay because the original worked okay until the gauge reached 1/4 tank. That is where it stopped. Any suggestions on how to fix this gauge to be accurate? Is it possible? Again, thanks in advance for any advice.