Fuel pump thermistor readings

[Xicobombas]

It seems my fuel pump thermistor is busted, need to get a new one to put the reserve light working.
Since there are more husky owners with the same problem, I need a good samaritan with a working fuel light reserve to measure the resistance with and without gas to see if this thermistor fits the requirements.

https://www.aliexpress.com/item/Aut...32548087148.html?spm=2114.40010708.4.9.ELIKtw

Best way to do it, is to take the fuel tank off, grab a multimeter and read the resistance (ohms) between the blue and white pins of the plug (picture attached).
Do this with gas above the reserve line and without gas.

Thank you.

 

[EUROJulian]

I have plans to go into fuel tank to judge how previous owner did fuel pump replacement, but cannot commit to how soon. I will do that time permitting in the first phase of restoration or second. I have plans to have bike running by this coming weekend and still bunch of work left including fork seals. I would take reading from removed tank as is but I do not actually know if my reserve light works or not. I just rode it around the house couple of times, topping off tank while coming back...

 

[Trenchcoat85]

Xicobombas-
The manual specs the thermistor as 1300 ohms at 20°C (68°F). The thermistors I bought of ebay (5 for less the $1) are the same as this pretty much, except they have radial leads instead of axial leads (1990 ohms at 58°F and 210 ohms at 137°F.... semi-linear in between *ahem*)

Speaking of which, your thermistor lead may have broken inside the can and might be repairable (you will have to open the can regardless). Gently pry up the rim and see if the fiber plate is still connected to the intact lead (or the lower lead is still soldered to the can). If not, solder a new section of wire on and maybe epoxy the thermistor to the side of the can to reduce movement (leaving space for gas to pass through and leaving the thermistor body exposed of course).

I took a hiatus from experimenting when I concluded that the design has a flaw: the can is cantilevered out on a stiff steel bracket- which is prone to break right at the mount arm (look very, very closely for a hairline crack). I have several redesigns in mind but haven't really gone forward on any... because my bike doesn't even use your type of fuel pump/thermistor (eventually, I'll regain interest though).

The ebay thermistors are here:
http://www.ebay.com/itm/5-x-1K-OHM-...EE-SHIPPING-/251118501039?hash=item3a77d43caf





oh yeah- I think it was you that was asking: that rubber skirt (not in you picture) is a splash-guard thingy I think; and keeps gas around the bottom of the fuel pump.

 

[Trenchcoat85]

BTW, Xicobombas... your ohm test will not really work except for the fact that it will give the continuity (and resistance). The "wet" and "dry" readings will be close (assuming the gasoline is the same temperature as the air) because the thermistor depends on the current going through it as the source of heat (IOW: no current- no heat produced; thus no change in resistance). Meters, especially digital multi-meters, use insanely small currents to measure resistance.

 

[Xicobombas]

BTW, Xicobombas... your ohm test will not really work except for the fact that it will give the continuity (and resistance). The "wet" and "dry" readings will be close (assuming the gasoline is the same temperature as the air) because the thermistor depends on the current going through it as the source of heat (IOW: no current- no heat produced; thus no change in resistance). Meters, especially digital multi-meters, use insanely small currents to measure resistance.

You surprise me... your experience and experimentation's are bless to this forum and it's readers.

I know my thermistor is dead or some lead is not making contact because I tested the continuity of the 2 wires attached to the thermistor, black and orange to the blue and white on outside plug and they are good.
Checked the resistance on both places and my digital multi-meter reads ZERO, so either is dead or something might be broken inside the can. I'll give it a try later on.

Will take a photo of the rubber skirt too.
Thanks

 

[Xicobombas]

Here is a photo of my pump with a nice black rubber skirt.

Opened the thermistor can and one contact is broken ... dunno if I broke it when was removing it or it was already broken.
Never seen a sensor which such a bad look, the head seems cooked... Can't weld it, the tin doesn't stick, the thermistor material is weird, and the readings are crazy, in the center reads 1300 ohms but if I move a bit gives all kind of readings.

The can diameter is 10mm, your thermister has a head of 6,5mm. It doesn't left much room for the gasoline to run freely.

Someone borrowed me a thermistor of 10K ohms, gonna try in the bike to see if the reserve light comes alive.

 

[Trenchcoat85]

Here is a photo of my pump with a nice black rubber skirt.

Opened the thermistor can and one contact is broken ... dunno if I broke it when was removing it or it was already broken.
Never seen a sensor which such a bad look, the head seems cooked... Can't weld it, the tin doesn't stick, the thermistor material is weird, and the readings are crazy, in the center reads 1300 ohms but if I move a bit gives all kind of readings.

The can diameter is 10mm, your thermister has a head of 6,5mm. It doesn't left much room for the gasoline to run freely.

Someone borrowed me a thermistor of 10K ohms, gonna try in the bike to see if the reserve light comes alive.

skirt: yep, I got the same one.

thermistor: "...seems cooked"- yeah, it is totally burnt. I'm surprised you can measure the resistance at all. It probably went into thermal runaway (got a little hot-> resistance drops-> more current-> got hotter: repeat until it smokes). Maybe check the other inline resistor (110ohms, 68ohms, or 7xxohms depending on the bike) which is near the regulator usually inside the loom (wtf!). It is just soldered onto a wire without support and easily breaks.

I cooked my thermistor by hooking it straight to 12v and watched the resistance slowly (and then exponentially fast) try to get to zero ohms. oops. well, I bought it to experiment, I guess.

btw, consider supporting the new thermistor with a drop of epoxy to the can sidewall (leaving the majority of the thermistor body clear).

can diameter/ thermistor clearance: absolutely no problem at all, less is better. All the gasoline comes in and out those 0.5mm holes in the can. The liquid gas (ugh) conducts the heat to the can and the can conducts the heat into the outside gas. The thermistor body is 5.4mm diameter, 3.2mm thick, and with the leads splayed to go to the top & bottom of the can: 5.9mm. I measured 9.4mm for the inside diameter of the can- no worries there. But I'd use epoxy for support and insulation, because the bare leads are not as thick as the stock thermistor. hell, put a heat shrink tube on the top lead too.

Before doing all these mods, maybe set up an experiment (without using the can or gas- maybe water?) with the thermistor hooked into your bike's circuit. Submerge and slowly raise the thermistor (live circuit) and make sure your LED works as it should. Don't hold it in your hands- it could go to 1000° in a second or two if something is wrong.

You might be able to emulate the circuit by using 2 different resistors at 2 different times. If you go this route, I'd probably try 1200ohms (tank full) and maybe 330ohms (tank empty? it's about 120°F or 50°C... which liquid gasoline could get I suppose. shit- maybe a smaller resistor). Also, I am not happy with my measurements of the thermistor I did last November- I used hot water but I also used an IR thermometer instead of my digital (conductive) thermometer. The low emissivity of stuff like water may have come into play.

Inspect the can mount strap for a hairline crack closest to the silver bar that mounts on the fuel pump holder.

hey, take pictures, take notes etc... you're going to be teaching me here soon.

good luck.

 

[Xicobombas]

Trenchcoat85

Here is well explained the entire procedure.

http://www.aeu86.org/forum/Thread-How-to-Repair-low-fuel-warning-sender?pid=151308

Ordered the thermistors, unfortunately the radial ones, axial could only find in USA and had to pay 18€ of shipping costs . Installation will begin next week. Pictures will be provided.
If the led reserve doesn't work then I'll check the inline resistor and bulb. One problem at a time.

 

[Xicobombas]

Here is what I did to replace the busted reserve sensor.

First of all, the thermistors. A friend of mine has an electrical store and ordered this 2 for 1€.





Using the multi-meter.



Since it should have 1000 ohm at 25º C, I believe it's working well by having 1068 ohm at 24º C.
The manual specs says the thermistor has 1300 ohm at 20º C, this thermistor might read that, I did not tested it at that temperature but after welding I read pretty low resistance values, so, higher the temperature lower the resistance and vice versa.

Now the welding process.
Removing the tin, it's quite hard to remove it from the bottom of the can and from the top of that connector, I tried to used the welding vacuum but with no success, so I used a tiny drill and lot's of patience, both holes were tin clear.



I realized only 1 of the thermistor connectors can touch the can because it messes up with the readings.

 

[Xicobombas]

So I folded the thermistor to look like an axial one.



Started welding the bottom to position the thermistor in the middle of the can.
Then weld the top and leave a bit on the outside to attach it to the black wire of the pump. The bottom can be cut off.



Welded to the black wire of the pump.



I used heat-shrinkable sleeve to insulate the weld and the black wire.
Then weld the can to the pump and with it the orange wire.



Eh voilá, all well welded and secured.
Keep clear all 3 holes on the top and bottom to ensure the gas runs free.
The readings on the blue and white wires of the connector give 1000 ohm at 25º C.
Next step, to check the reserve led.

 

[EUROJulian]

Crafty

 

[Trenchcoat85]

Great job.

Make the LED test short; you've already lost one thermistor. I would test in stages.

 

[Trenchcoat85]

You know, I'd buy some 1/2 watt 330ohm resistors from your buddy (this a nice friend to have, btw) and stick one in the female sockets where the blue and the white wires go (no soldering, no tank, no pump) and turn the key on. If the LED did not come on instantly, add another resistor in (lowering the resistance to 165ohms) and try again. No LED? fix that problem before sacrificing your thermistor.

cheap. easy. fast.

And if the LED did come on, I'd feel better about plugging in the fuel pump plate (without the tank) and watching the LED come on a bit slower- hopefully. After that, a dry tank test & then a wet tank test... and then ride.

Of course, the 2nd part of this test is to do the same thing, before the tank test, with a bigger resistor (1.2k ohm?) and make sure the LED goes off (OFF).

I have a great amount of respect for ignition sources in a gas tank (IOW: scared shitless). I know from an intellectual standpoint that the atmosphere in the tank usually can't support combustion or an explosion, but try telling that to my inner brain- it ain't working.

This is all based on the premise that your thermistor was not the cause of the failure; something in the circuit outside the tank failed... and caused the thermistor to go into thermal runaway.

maybe.

 

[Trenchcoat85]

Hey Xicobombas- if you still have your thermistor/pump out, add a fillet of solder in the area of the holding rod.

 

[Xicobombas]

Hey Xicobombas- if you still have your thermistor/pump out, add a fillet of solder in the area of the holding rod.
View attachment 77542

It's a good advice indeed. I removed the can thru the top and nothing was touched except desoldering the orange wire. Putting all back together I added extra solder, but all seemed to me pretty tied and consistent. One never knows when the gas takes it better at the holding rod but when it does, I know where to start repairing.

 

[Xicobombas]

You know, I'd buy some 1/2 watt 330ohm resistors from your buddy (this a nice friend to have, btw) and stick one in the female sockets where the blue and the white wires go (no soldering, no tank, no pump) and turn the key on. If the LED did not come on instantly, add another resistor in (lowering the resistance to 165ohms) and try again. No LED? fix that problem before sacrificing your thermistor.

I'll test that... Have some resistors from my arduino kits, and try with a 1K ohm and a 100 ohm resistors.

Of course, the 2nd part of this test is to do the same thing, before the tank test, with a bigger resistor (1.2k ohm?) and make sure the LED goes off (OFF).

Aren't you mistaken ? With a 1K ohm resistor the LED should come ON. And with a 100 ohm the LED should go OFF.

 

[Trenchcoat85]

The hotter the thermistor gets, the lower the resistance. Liquid keeps it cooler, air lets it get hotter.

 

[Xicobombas]

The hotter the thermistor gets the lower the resistance. Liquid keeps it cooler, air lets it get hotter.

Yes it's true, so I am right. When thermistor reads 1K ohm resistance there's no fuel in the tank, the LED should come ON.
If the liquid (gas) makes it cooler, then with a 100 ohm resistance, the LED should go OFF.

Don't have time today to test that, but tomorrow definitely.

 

[Trenchcoat85]

I am interested to see your test results.

 

[Trenchcoat85]

Yes it's true, so I am right. When thermistor reads 1K ohm resistance there's no fuel in the tank, the LED should come ON.
If the liquid (gas) makes it cooler, then with a 100 ohm resistance, the LED should go OFF.

Don't have time today to test that, but tomorrow definitely.

Hey, anything new to report?