Effects of high eth content on fuel level sensor

Do you happen to have pictures of the failure mode experienced? Is it the wiring or is it the actual contacts?

 

Not yet. i havent pulled it out. I'll post some when i do.

 

I recently pulled the failed OEM (3rd of 4th now?) sensor out and found the EZ turn goop completely gone. Same failure mode though, with the positive wire and contact pad corroded. I suspect I did not completely encase the positive wire with solder. I received the ford replacement sensor (cant recall from which car) and cut the terminals off again. This time I really tinned the shit out of the wires and dripped a good amount of solder on the board. The wires were excessively encased in solder this time. However, with different resistance values the reported level on the gas gauge is inaccurate. This was expected of course.

It would be a whole lot cheaper if the actual DENSO circuit board could be purchased as they can be removed from the white plastic holder mount. The only indication of a PN on the board itself is the numbers printed on the side of the board. I used the denso cross reference search on their website and cant seem to find anything.

I plan on purchasing another OEM level sensor and soldering the wires similarly, but also completely covering them with the 3M 2216 epoxy we have here at work.

 

Quick little update. The current Ford sensor is still holding up at almost 5 months since replacing. So far, tinning the wires and encapsulating in solder (in excess) directly to the level sensor board has proved the best results. Fuel level reported on the gauge is still off due to different resistance levels. When and if this sensor fails, I will follow the same process with an OEM sensor.

Sent from tapioca

 

How much is the fuel level off?

 

Empty is at the half tank mark. I just splash it with 2.5gal of 91 and 2.25 of e85 every time I drive it lol

Sent from tapioca

 

Why not use a 3 wire variable resistor to dial in the gauge? Would be a cheap and easy install.

Had to use one on my old Mitsubishi with an SAFC 2 to get the RPM to read correctly. Costs less than 5 dollars.

 

Even so it will be a huge pain in the ass to get right; cars have logic built in to the fuel level gauge so you don't see the effect of fuel sloshing around all the time. He's probably going to have to do some key off/on stuff or actually wait a while for it to correct after an adjustment is made using that method. Be better to inline resistor it at the appropriate value and be done with it IMO, or just wait a while until he can mod the appropriate gen sender to be e-resistant since he's gonna have to go in there anyways to do any method.

 

I found a possible solution which is using the Rockauto Delphi whole assembly. The description sounds very promising. If you google Delphi there are mixed reviews. I'm thinking it should work fine. Full description here: https://www.rockauto.com/en/moreinfo.php?pk=8733236&cc=1502914&pt=10147
It's $334 + shipping (Fedex ground 10.99 for me) which is the lowest I found as other places it's like $400+.

 

Any and all solutions are appreciated. That sensor on the delphi unit looks identical to the oem mazda sensor. Are you experiencing a similar condition with failure of the fuel level sensor? Honestly, my solder trick has been holding up. I'm still on the same sensor from a different ford vehicle. I think it's more about the connectors on the level sensor it self corroding. So cutting them off, tinning the wires, and soldering them to the board has worked so far.

 

I'm glad the solder has worked for you! I guess this info could also be useful to others. I'm not experiencing problems yet as I'm on 93 but will be at least e30 in the future. I heard all gen 2's have this problem. My guess is they used cheaper metal for the sensor such as mild steel that easily rusts instead of stainless. It would be nice to see a test of gen1 vs gen 2 sensor material. The Delphi actually looks pretty different. It comes soldered, the tracks look slightly different and it uses more resilient material according to them
OEM vs Delphi

Which Ford sensor are you using? ps603, AE5Z-9A299-H
It seems this Delphi unit, although more expensive, is a simple drop in solution and I'm guessing the fuel level reading will be correct and not empty at half tank mark It was listed on Rockauto for Ms3.

 

@drew I cant recall the PN i used for the ford sensor. I want to say it was from a mercury MKZ or something similar. The delphi sensor on that pump basket assembly may work because the wires are directly soldered to the board. Pricey for an uncertainty in my opinion. The oem level sensor can be had for 70 bucks i think. Just take the path i took and you should be good

 

Picked up another oem level sensor and performed the same process with soldering wires directly to board. I overpot the shit out of it with 3M 2216 epoxy. Supposedly this stuff should hold up to ethanol/jet fuel says a coworker of mine.

Sent from tapioca

 

Pulled out the level sensor from the basket. You can see the solder was holding up well. Remember, this technique made the sensor last more than 6 months. Instead of a couple weeks! So I'm really curious about the 2216 epoxy providing that much more protection.

Sent from tapioca

 

Well the new OEM sensor died in a short few days. Looks like my Epoxy technique did not work. I suspect the OEM sensor just isn't cut out for ethanol/moisture. The MKZ sensor i pulled out would have kept on just fine. Drew101010 suggested a comparable sensor for a yaris that will arrive today.

 

I wonder if when you figure this out you can sell kits or something through RaiderFab.

 

Happy to just contribute to the community with research and data

 

Stumbled upon this thread/post earlier:
https://mazdaspeeds.org/index.php?threads/gen1-fuel-pump-assembly-in-gen2.3208/

Note that the genwon setups don't have issues with e85.

 

The solution was right there the whole time

 

I'm pretty sure he's looking into non basket replacements. You know, less expensive options.