MV Agusta Forum banner

1 - 20 of 55 Posts

·
Registered
Joined
·
200 Posts
Discussion Starter #1
My low fuel like spontaneously stopped working a while back. No flickering, no erratic behavior, just one day it never lit up again.

The manual has a simple test for checking the fuel sender. I've attached a screengrab of it below, and you can also see a simplified execution of that test on THIS THREAD, where SpainF4 ran the test but only built his circuit through pins 3 and 4, and it worked.

The problem with this test is that if it is negative--that is, the light doesn't come on--that doesn't tell you whether the thermistor is bad, or whether there's just a connection issue somewhere between the 4-pin fuel connector and the thermistor itself. MV doesn't care, because they intend for you to replace all of it.

A new fuel sender is north of $250 in the US. You can buy an entire fuel plate, including sender and pump, for around $150-175 on ebay, but then you're never quite sure how far away THAT fuel sender is from giving up the ghost.

That seems like a lot of money if the issue could be fixed by cleaning up/repairing one of the connections inside, or replacing just the thermistor at the tip, which I've found similar ones for less than $40 around the web. I wanted to find out whether fixing the unit itself was possible instead of buying a whole new one. This could also open the door to being able to raise the thermistor higher with a spacer or washers of some kind, so when the low fuel light comes on, you have a little more warning before empty.

So far, my conclusion is that it would be possible if there was a way to remove the resin sealant. That would give you access to fix the wiring connections, and also remove the thermistor entirely and test it by itself, which you can do just with a multimeter reading resistance while you dip it in/remove it from liquid.

I don't know how to do that, I've never worked with the stuff. Anyone? Maybe it liquifies in boiling water or something?
 

Attachments

·
Registered
Joined
·
97 Posts
From the 910S manual from Donsy's site pin 3 and 4 are for the sensor and 1 and 2 are powering the pump. If I'm looking at the wrong manual please correct me.

To test that the connection between the thermistor and the plug is ok or not you can carefully shave the insulation from the white and black wire in your pic without breaking the wire and test continuity from there to the plug.
If there's no continuity there's your problem, then test the resistance of the thermistor by dipping it between a glass of water with ice cubes and then a glass of very hot water while connected to your previous test points
and you should see the resistance rapidly fall as the temp gets higher.

You probably already searched on how they work but http://www.resistorguide.com/ntc-thermistor/ a 10K thermistor is most common because they have a working range between -20c to +150c

I guess you already checked continuity from the mating plug to the dash and that's ok.

Google 'de-potting electronic circuits' about how to remove the resin, it's epoxy so it won't dissolve in common solvents, paint stripper with dichloromethane also called methylene chloride will soften the top layer
then you have to scrape it away a layer at a time, don't know if it will wreck that tube or not. Glasses, gloves and a well ventilated area are mandatory

They only safe way to remove it is via mechanical means, If the problem was a broken wire you could dremell out all the resin, fix the connections and re-pot it.
The caveat being you would have to find fuel proof & vibration resistant epoxy and if it fails later you have fuel leaking straight on top of a hot motor.

P.S.
how did your dash LED fix go?
 

·
Registered
Joined
·
200 Posts
Discussion Starter #3
From the 910S manual from Donsy's site pin 3 and 4 are for the sensor and 1 and 2 are powering the pump. If I'm looking at the wrong manual please correct me.
Thanks for the thoughtful response. That is the correct manual; same wiring as the 750 S that I'm working with. And that would make sense for the wiring to be the way it is diagramed in the manual, BUT you get continuity to pin 2 from both the ground on the fuel pump connector AND the ground on the thermistor, and no continuity to pin 4 from the thermistor ground. The Knurl had similar results when he checked his. If it wasn't all epoxied in so well, I'd assume there's a short, but because of how the piece is assembled, I think it's more likely some screwy MV wiring...

To test that the connection between the thermistor and the plug is ok or not you can carefully shave the insulation from the white and black wire in your pic without breaking the wire and test continuity from there to the plug.
If there's no continuity there's your problem, then test the resistance of the thermistor by dipping it between a glass of water with ice cubes and then a glass of very hot water while connected to your previous test points
and you should see the resistance rapidly fall as the temp gets higher.
That's essentially what I was working towards, but stopped because there's no way to disassemble it to fix independent components (unless you drill out the epoxy). That's the crux of the decision: If you can repair the wiring inside, it is worth verifying whether the fault is the wiring or the thermistor. If you can't, it doesn't really matter what the culprit is, the thing is broken and you have to replace.

They only safe way to remove it is via mechanical means, If the problem was a broken wire you could dremell out all the resin, fix the connections and re-pot it.
The caveat being you would have to find fuel proof & vibration resistant epoxy and if it fails later you have fuel leaking straight on top of a hot motor.
I haven't decided what I'm going to do yet. I feel like it ought to be fixable and it's really just my obstinance pushing me to take a crack at it. I am considering drilling it out, which would trash all the wiring in the process, and testing the thermistor by itself/rewiring/re-potting. My concern there would be what you stated about a fuel leak, of course. There's a lot of room in that tube in which you can fill it with epoxy and get a nice tight seal. It's tempting, but also time consuming, and I may just cave and decide the time it would take me to repair is more valuable than the $150 I'd spend on ebay for a whole fuel plate assembly.

P.S.
how did your dash LED fix go?
Still going. I practiced soldering/de-soldering on a little circuitboard to get more confident with my technique and to zero in on a temp setting that works. I'm fighting a war on multiple fronts: fuel sender, dash LED, I was troubleshooting alternator+battery but wrapped that up, getting my microtec set up with a lambda sensor to autotune on a closed loop...
The dash is the lowest priority because it's so accessible that I can finish up everything else and reassemble the bike and get it rolling, and come back to the dash as I have time for it. Ideally, I'll have it all done this week, but we'll see what life has to say about that
 

·
Super Moderator
Joined
·
22,354 Posts
Because of the way thermistors work, they need heat to produce continuity. The circuit heats up with applied voltage and no neat sink. When a temperature limit is reached, the circuit closes via action of the thermistor and then continuity is gained and your light comes on.

This is why the circuit shows open with a standard continuity test.

The thermistor stays cool (no continuity) while immersed in fuel. When the fuel drops below the thermistor, the heatsink is removed and continuity is gained.
 

·
Administrator
Joined
·
19,858 Posts
I seriously doubt a wiring failure has occurred inside the potting material. After all, how could it? the wires are immobilized and sealed. No way to flex and bend, no way to wear and short or break.
I would suggest cutting the thermistor off with some lead wire attached. You can then test continuity and insulation of the wiring to the connector, and test the thermistor independently with ease.
You can then solder in a short length of similar gauge wire with heat shrink tubing to reconnect the thermistor or a replacement.

Don't forget continuity and insulation from the pump connector back up to the dash.
 

·
Registered
Joined
·
200 Posts
Discussion Starter #6
Because of the way thermistors work, they need heat to produce continuity. The circuit heats up with applied voltage and no neat sink. When a temperature limit is reached, the circuit closes via action of the thermistor and then continuity is gained and your light comes on.

This is why the circuit shows open with a standard continuity test.

The thermistor stays cool (no continuity) while immersed in fuel. When the fuel drops below the thermistor, the heatsink is removed and continuity is gained.
And based on the diagram for the test with the test light, we know that pin 4 is the side of the connection that gets opened/closed, and pin 3 should have continuity to the positive lead up the center of the thermistor (I haven't exposed any of that wire to check). What I'm not understanding is why the negative side of the thermistor also has continuity to pin 2

I seriously doubt a wiring failure has occurred inside the potting material. After all, how could it? the wires are immobilized and sealed. No way to flex and bend, no way to wear and short or break.
Yeah, when you put it like that, it seems painfully obvious that there was no need to suspect the connections inside. Lesson learned.

I would suggest cutting the thermistor off with some lead wire attached. You can then test continuity and insulation of the wiring to the connector, and test the thermistor independently with ease.
You can then solder in a short length of similar gauge wire with heat shrink tubing to reconnect the thermistor or a replacement.
I think Ed's got the fix here. Then it would just be a matter of finding a similar thermistor, and plenty abound on the internet. Is heat shrink submersible in gas, is there a special kind you would need to buy?

It might be difficult to get the new wire and heat shrink coiled back into the plastic housing. There's a little vent hole on the side, and perhaps you could use a small pin to hook the wire inside the tube and try to work it down in, pulling the thermistor back to its perch on top

Don't forget continuity and insulation from the pump connector back up to the dash.
Right. One thing at a time. The sensor failed the test on its own, so fixing that has to happen.
 

·
Super Moderator
Joined
·
22,354 Posts
What I'm not understanding is why the negative side of the thermistor also has continuity to pin 2
As pin number 2 is the ground for the fuel pump, perhaps they are bridged together as shown:

F4 312R Fuel Sensor Circuit.jpg
 

·
Registered
Joined
·
97 Posts
Because of the way thermistors work, they need heat to produce continuity. The circuit heats up with applied voltage and no neat sink. When a temperature limit is reached, the circuit closes via action of the thermistor and then continuity is gained and your light comes on.

This is why the circuit shows open with a standard continuity test.
If you are measuring resistance across the thermistor when isolated it should show high resistance at room temp (e.g. ~10K) not infinite if its OK,
This would show if the thermistor is good or has gone open circuit and needs replacing.
I agree that there would be a constant current source when its in circuit as a liquid level sensor. The link I posted earlier describes the self heating effect.

The tests posted earlier would show connection between the plug of the sender and the thermistor are good or not, then testing the connection between the harness plug and the dash LED should eliminate bad connections there.

Wires do fail in potted enclosures as anyone who owned a 1970 -> 1990 Ducati will tell you but its unlikely now days but easy to test in this situation.
 

·
Super Moderator
Joined
·
22,354 Posts
The wires in my wife's 2007 Ducati Monster 695 failed in the potted enclosure...so it can occur even on more modern bikes.
 

·
Registered
Joined
·
170 Posts
I have been looking at this thread as I also have the same problem and in addition my carbon tank is leaking at the rear mounting spigot, I have sent my tank away in the hope that the leak can be repaired. In the meantime I have purchased a 1000 tank as a spare and I intended to install my 750 pump into the 1000 tank should the carbon one be unrepairable . Unfortunately the mountings are different on the two tanks, the 750 uses bolts to hold the pump plate and the 1000 uses nuts onto pre installed studs. So the question is can the 750 plate mount to the 1000 tank and before someone says try mounting it and see I sent the 750 pump and plate away with the carbon tank and it probably will not be back until the end of summer. Glenn
 

·
Super Moderator
Joined
·
22,354 Posts
There should be no reason that the pump lates are not interchangeable.

Or to word it differently: Yes, the pump plate should fit.
 

·
Registered
Joined
·
200 Posts
Discussion Starter #12
So the question is can the 750 plate mount to the 1000 tank and before someone says try mounting it and see I sent the 750 pump and plate away with the carbon tank and it probably will not be back until the end of summer.
Hey Glenn--

I think you answered your own question there, but you can go to StarTwinTrading.com and look at the parts fiche that includes the carbon tank you have for your 750, and for the 1000 from which you purchased the spare tank, and compare the part numbers of the fuel plates. While looking at fuel plate+pump assemblies on ebay as an option to replace my fuel sender unit, I noticed that there was variation in the plates and you're likely going to need to get a plate that matches the 1000 tank.

If your fuel sender is bad, as mine was, I'll be posting some insights this weekend as to what I think a good troubleshoot/repair process would be for that based the mistakes I have made and learned from.
 

·
Registered
Joined
·
170 Posts
I do have a plate that will fit the 1000 tank but wanted to use the pump from the 750 without having to decant from one plate to another if possible. Glenn
 

·
Registered
Joined
·
170 Posts
I checked the pumps for both the 750 and 1000 and both of them show the newer style plate to be used with the studs and nuts so the conclusion is that you should be able to use the older plate on the newer mounting but will confirm when I get my 750 pump back in a few weeks. Glenn
 

·
Registered
Joined
·
83 Posts
I would assume they all have common specs, such as those listed here:

SANGSHIN ELECOM CO., LTD.

I may just go down to a junk yard and pull one out of a car. The unit in my dad's 31 year old Toyota 4runner has never failed.

Seems like the biggest issue is how to mount it. The original MV unit is pretty uniquely shaped to fit on the end of the plastic rod. I'll let you know how it goes.
 

·
Super Moderator
Joined
·
22,354 Posts
The placement of the thermister, and its calibration, are what provide the correct fuel level indication.
 

·
Registered
Joined
·
83 Posts
The placement of the thermister, and its calibration, are what provide the correct fuel level indication.
Placement isn't critical, if it's a few mm higher or lower just means it will turn the light on sooner or later. Better than no light. As for calibration, there is none. It's a simple thermistor so it just heats up when the fuel is no longer cooling it and it makes continuity activating the light circuit.
 

·
Super Moderator
Joined
·
22,354 Posts
Placement isn't critical, if it's a few mm higher or lower just means it will turn the light on sooner or later. Better than no light. As for calibration, there is none. It's a simple thermistor so it just heats up when the fuel is no longer cooling it and it makes continuity activating the light circuit.
Um, I do understand how thermistors work. They are designed for differing temperature ranges and applications. For instance, if you install one in your fuel tank that is designed for an application in a high temperature Scram Instrument Level system for a Nuclear Power Plant, well, you get the point.

If you are worried about the level being high or low relative to distance, why not just skip it and use mileage...
 

·
Registered
Joined
·
200 Posts
Discussion Starter #20
They are designed for differing temperature ranges and applications
I imagine as long as the thermistor you get is intended for fuel, you'll be fine. There are tons of cheap options out there that look like they'll work,

If you are worried about the level being high or low relative to distance, why not just skip it and use mileage...
This was my approach for a long time while I waited around for a good deal on a new fuel sender unit. The reason I finally decided I wanted the dummy light back was because the gas mileage I get varies greatly by the type of riding I'm doing. When I knew I burned all/most of the tank riding canyons, I had a good idea of the total distance I can expect from the tank. When my riding was more mixed--sometimes highway to commute, sometimes opting to cut through the park, sometimes filtering through traffic at low speeds for a while--it was harder to track, and I'd resort to shining a light into my tank before setting off on a ride to sort out whether I had to get gas en route, or could wait until I was on my way back or something. I never ran out doing this, but it was annoying, and had me stopping to fill up more frequently than I needed to be.

There's also something to be said for just wanting things to work, rather than "living with it." At least I know for me, that's a motivating factor
 
1 - 20 of 55 Posts
Top