65 GP I think I may have a defective new fuel sending unit

[joelj]

Let me know if I am on the right track. I installed a new sending unit on my 65 Grand Prix (the old one would not read above 1/4 tank). The new unit read about 3/8 tank then fell to empty with about 5 gallons of gas on a completely empty tank to start with. I then put another 5 gallons or so in the tank thinking that the reserve capacity of the tank may be affecting  the gage. Still the needle is on E. I then grounded the feed wire to the sender and the gage went to full. Re-connect and it goes to E. With about 10 gallons of fuel my gage should be reading at just under ¹/2 tank right? Before I  call and try to get a replacement sending unit I would appreciate your input. Am I missing something? Thanks.

[Bloo]

3 minutes ago, joelj said:

I then grounded the feed wire to the sender and the gage went to full. Re-connect and it goes to E.

That's backwards, so something is really wrong. Shorting the wire to ground should have made the gauge go empty.

 

Around 1965, GM sending units and their gauges changed from 0-30 Ohms to 0-90 Ohms. The year of changeover probably varied a little by make/model (I think) so I am not sure which standard your 1965 GP gauge should be on. Both standards have 0 ohms (wire shorted to ground) as "E", so either way something is screwed up if grounding the wire makes the gauge go full. Disconnecting the wire should make the gauge go full, and most likely peg. Shorting the wire to ground should make the gauge go empty. There's probably something wrong with the gauge. Is there any chance it could be hooked up backwards? I don't know what the effect of that would be. I doubt it explains backwards but maybe.

 

To check the sender, use a multimeter set on Ohms. Basically any will do, even the 6 dollar Harbor Freight ones that they used to give away free will work fine for this. With the sending unit out, Zero the ohmmeter buy shorting the leads and pushing the "delta" button. Then hook one wire to the sender terminal and one wire to the sender frame. With the float hanging all the way down, the resistance should be extremely low, like preferably less than an Ohm. True zero is physically impossible, but it should get real close. With the float all the way up it needs to measure at least 30 ohms more (or maybe 90 ohms more) depending on which standard the 65 GP used.

 

If your meter doesn't have a delta button, short the leads and write down the number you get. This is the resistance of the test leads. Subtract it from the readings you get from the sender. For instance, if you get 2.1 Ohms with the meter leads shorted, and 2.8 ohms when measuring the sender with the float hanging down, then 2.8 - 2.1 = 0.7 real Ohms at the sender.

 

I guess I should point out that if they sent you the wrong year sender, you would have the wrong amount of sweep, for instance a 0-30 Ohm sender connected to a 0-90 Ohm gauge would only be able to go about 1/3 of the way up. That doesn't explain backwards though.

 

[TTracing]

Here is a chart I had from working on several different jobs     Keep in mind ya never know what the last guy did 🙂

 

[Bloo]

There's some mistakes in that (Mopar for sure), but it's about right for GM cars. The trouble is a 1965 car is right on the edge. Should it be 0-30 or 0-90?

Edited June 25, 2023 by Bloo (see edit history)

[joelj]

11 hours ago, Bloo said:

That's backwards, so something is really wrong. Shorting the wire to ground should have made the gauge go empty.

 

Around 1965, GM sending units and their gauges changed from 0-30 Ohms to 0-90 Ohms. The year of changeover probably varied a little by make/model (I think) so I am not sure which standard your 1965 GP gauge should be on. Both standards have 0 ohms (wire shorted to ground) as "E", so either way something is screwed up if grounding the wire makes the gauge go full. Disconnecting the wire should make the gauge go full, and most likely peg. Shorting the wire to ground should make the gauge go empty. There's probably something wrong with the gauge. Is there any chance it could be hooked up backwards? I don't know what the effect of that would be. I doubt it explains backwards but maybe.

 

To check the sender, use a multimeter set on Ohms. Basically any will do, even the 6 dollar Harbor Freight ones that they used to give away free will work fine for this. With the sending unit out, Zero the ohmmeter buy shorting the leads and pushing the "delta" button. Then hook one wire to the sender terminal and one wire to the sender frame. With the float hanging all the way down, the resistance should be extremely low, like preferably less than an Ohm. True zero is physically impossible, but it should get real close. With the float all the way up it needs to measure at least 30 ohms more (or maybe 90 ohms more) depending on which standard the 65 GP used.

 

If your meter doesn't have a delta button, short the leads and write down the number you get. This is the resistance of the test leads. Subtract it from the readings you get from the sender. For instance, if you get 2.1 Ohms with the meter leads shorted, and 2.8 ohms when measuring the sender with the float hanging down, then 2.8 - 2.1 = 0.7 real Ohms at the sender.

 

I guess I should point out that if they sent you the wrong year sender, you would have the wrong amount of sweep, for instance a 0-30 Ohm sender connected to a 0-90 Ohm gauge would only be able to go about 1/3 of the way up. That doesn't explain backwards though.

 

Sorry, the backwards issue was my mistake. I missed a step when I was trying to describe the problem. When I disconnected the feed wire the gage went to full. Grounded, it goes to empty. Re-connected it reads empty. Thanks for your help.

[TTracing]

Bloo   you are right didn't notice that last mopar  happened to be a dart I worked on was   73-78  to  10ish at full

[Century Eight]

On 6/25/2023 at 7:52 AM, joelj said:

Sorry, the backwards issue was my mistake. I missed a step when I was trying to describe the problem. When I disconnected the feed wire the gage went to full. Grounded, it goes to empty. Re-connected it reads empty. Thanks for your help.

Joelj. Did you ever fix this?  I have had the very same scenario twice on two Buicks.  The sending units they sell today are mediocre.  The floats have microscopic pin holes and will work at first but then fill up with gas and sink to the bottom.  Just get a new float for a few bucks and put it on the sending unit.  In my case, that was the fix.   That will probably fix it.

[joelj]

15 hours ago, Century Eight said:

Joelj. Did you ever fix this?  I have had the very same scenario twice on two Buicks.  The sending units they sell today are mediocre.  The floats have microscopic pin holes and will work at first but then fill up with gas and sink to the bottom.  Just get a new float for a few bucks and put it on the sending unit.  In my case, that was the fix.   That will probably fix it.

No, I never did get around to fixing this issue. I am reluctant  to buy another identical sending unit, go through the trouble of installing it and have the same results. The float on my replacment unit is a hallow white plastic. I am curious if that is the same as the one that you had problems with. If so,  are the replacements the same or different material? Thanks for your input, it gives me another idea on what to look for when I finally get around to removing the tank again and see what's going on.

[Century Eight]

Joelj. I was ready to get another sending unit but I started hearing about floats that failed (or sank).  So we just got another float, a friend had several sitting around so we tried one of those before going for the sending unit.  Mine were brass, both the bad and the good.  We got lucky and got a good one.  I don’t know about the plastic ones, but I suppose they could have a hole too.  I know they have cork floats but you should varnish those first, to keep them from getting as water-logged, only with gas instead.  Parts stores like NAPA used to carry them and still may, but you can get them from places like Bob’s or CARS etc.  

 

You should also check if the sending unit works, either by checking the ohms, as mentioned in other posts, or you can do it manually off the car. You can use a jumper wire and connect that to the wire that goes to the gauge plus a ground. Then work it up and down. If it works you can also put it in the tank while the tank is still off the car and check it and bend the arm slightly to get the needle where you want it.  You can experiment so the float shows empty before you actually run out of gas.  Who cares if it doesn’t show full, but you don't want the tank to run out while the gauge shows gas in it.  

 

I find the floats are under $10 while the complete units are maybe close to $90?  I have done it enough times that it’s just more or less a pain in the … but not really that difficult and I am 75.  It’s just nice to know it’s all going to work before you put it all back together.  And then it’s really nice to know how much (or little) gas you have in the tank.  

 

Good luck.

 

Phil

[Century Eight]

Joelj

On 6/25/2023 at 7:52 AM, joelj said:

Sorry, the backwards issue was my mistake. I missed a step when I was trying to describe the problem. When I disconnected the feed wire the gage went to full. Grounded, it goes to empty. Re-connected it reads empty. Thanks for your help.

Joelj

I just found a great tutorial on bad fuel pumps.  I was re -reading an old Hemmings before I threw it out, and came across this short but great article on defective fuel pumps. I can take a picture of it and email it to you, or snail-mail it.  I found it on page 38 of the May ‘21 Hemmings in the advertising section sponsored by “Chevy of the 40’s”.  You may find it on their website , if they have one.  The article is entitled “Life’s a Gas” and describes poor grounds, stuck gages, gage empty, gage full, gage read wrong etc.

 

It tells you simple diagnosis’s , and simple fixes and is short, and direct to the point.

 

Phil