Online Community of Zenith Builders and Flyers
I've had a recent problem with one of my fuel level indications and wanted to see if anyone else has seen this. I have the STOL 750 with the Zenith-supplied VDO/Volkswagen resistive fuel level senders mounted per plans. The senders are wired to my RDAC which then displays the fuel levels on my MGL XTreme EMS. This has worked great for a number of hours.
Recently, however, my "Fuel level #2" (right tank) began to indicate a few gallons lower than what I knew was in the tank. As the tank got low, it then would indicate "0" or empty even though it was not.
After re-checking all my wiring connections, I contacted Matt at MGL and he suggested re-calibrating the EMS with the sender. I did this and it seemed to be working perfectly again. I flew about 2 hours today and the first 3 legs of the flight it worked normally. However, when I fired up for the 4th leg and final flight home, it indicated several gallons low again and eventually fell on off to "0" or empty when I knew there was still several gallons of fuel in the tank. (I have a fuel totalizer and could subtract the fuel in the left tank from the total and knew how much I had burned out of the right tank.)
I'm now wondering if this might be a defective sender? In all of my time on this forum, I've never heard of one going bad that was working properly initially.
Just wondered if anyone else has ever experienced it? If the sender does turn out to be defective, is the VDO/Volkswagen sender that Zenith uses the best quality sender available?
Wouldn't doubt it. My feelings on these pieces of junk is clear. I'd say replace it (PITA that it would be) or get a fuel flow meter and use the EMS to read the tanks as the fuel flows. Lots of guys are doing this with the MGL EMS's and say it is way more accurate than the senders... I hate these things... :(
I did a little research and as you suggest, I could set-up two "virtual tanks" with two fuel flow senders. I've already got a "red cube" fuel flow sensor on my final fuel line and it is amazingly accurate! Guess I could get another fuel flow sensor and move them to my left and right fuel lines feeding into my selector valve. Since the VDO senders are not leaking, I'd just leave them in place, disconnected.
I'm going to give serious consideration to that versus having to drill out the rivets, remove the top wing root skin, and replace the sender! However, I'd like to definitively diagnose if the sender is bad or not if I can find out how to do that!
Yeah, you'd have to set the EMS to read one tank with 30 gallons unless you added a second flow meter. The cross feeding doesn't seem to be an issue. When I fill my tank I have to make sure the valve is on "off" or the tanks will cross feed faster than I can fill them.
I topped off one tank once and by the time I got to the second tank it was half full just from the cross feed! :-0 Could have saved some money myself by just getting an off/on tank valve. I thought I'd have use of the left/right to balance the plane in flight but turns out that was a non issue.
If you go the route of taking the wing apart for the fix, I'd switch the senders to the belite pressure ones. Heard nothing but good things about them and you can "T" them in the line near the tank too boot!!! That's what I plan to do, but it will involve getting the massive 2" hole in the tanks welded or patched! Still the worse setup I've seen yet for fuel senders... >:-(
It does apparently! :D
The Sonex builders have used them. Only issue apparently is the sensor uses 0 to 5v for max fuel and MGL apparently only goes to 0-4v, that causes issues reading a full tank. I'd fire an e-mail to Belite, I'm sure they have a fix now for this. Here is the info from a Sonex user from their site... Looks like he has a similar setup to you and I. I'm guessing this could be fixed easily. I was planning on looking more into it when I decided to switch out my senders WHEN (notice not IF) they start leaking...
I have a Belite fuel sender (measures fuel weight) and used it with my MGL xtreme. The problem I found was the Belite produces a voltage of 0 to 5 V. The MGL likes 0 to 4V. So when full, the MGL displays ERROR. Once the level goes down below 4V, it reads fine. I also use the MGL calculated fuel level and the Red Cube flow transducer. The levels are very close in flight once I get below the 4V level.
As mentioned further down in that thread, seems a simple resistor/voltage divider (or a diode?) could reduce the range to 0 to 4v. This is defintely worth looking into - but I first want to be sure my sender is the culprit!
John, I'm assuming you want to just prove it is the sender and not your EMS. Can't you just reverse the wires from the sensors so you right and left tanks are now reversed to the EMS. If you have any connectors between the sensor and EMS, I would check that too.
You're correct, I want to be positive it's the sender before I (a) get into drilling out that top wing root skin to get at the sender, or (b) consider any other senders or fuel flow systems!
Your suggestion is a great idea! Duh! Why didn't I think of that? ;>)
It'll be very simple to just switch the fuel sensor wires at the RDAC (the RDAC is the sensor interface for the EMS). However, I'm pretty certain I'll have to re-calibrate both tanks when I do that, but it's not difficult, just involves filling them and then draining 2.4 gallons at a time to enter the calibration points on the EMS. (When I switch the wires, I might get lucky and see the fuel level drop off abnormally low on the suspect sender even without calibration - to me, that would be evidence enough and I could switch the sender wires back and not have to recalibrate the "good" sender.) But, you're correct, if the problem follows the sender wires, then the RDAC and EMS are "innocent!" (There is no fuel sender wiring between the RDAC interface and the EMS - just the two CAN bus wires, so don't think there could be a problem there.)
I did check the wiring integrity at the RDAC and at the sender (as best I could through the access panel under the wing root!). I know there is at least one other variable - the sender is relying on a voltage to energize it. As I recall, I have the same voltage source supply both senders, so seems that if one works OK and the other erratically, likely the voltage is OK. Since the sender is kaput again, I might try to compare the voltages on the sender wires to make sure they're similar.
Thanks for the idea on switching the wires!
I tried to edit my reply but it hung...You could check the sensor itself by disconnecting from EMS, connect wire to a non-digital VOM and monitor the resistance for awhile. A non-digital (dial type) VOM is slower reacting and smoother for checking something like this. Might even want to run a new wire to test like this.
Well, went over to the hangar yesterday to do some diagnostics and of course, the sender was acting perfectly again! Ha! Intermittent problems are the worst to deal with! I didn't switch the sender leads to the RDAC as I was afraid it would mess up my calibration and I need the plane for a flight tomorrow. I did try slowly de-fueling the tank to see if there was a spot or range where the problem began in the sender, but it seemed to function OK over the full range from full to empty.
To have at least a reference to start from, I did go ahead and open the access panels and check the voltage at the senders and found them nearly identical (about 1.3v - my senders are powered by 4.3v reduced to 1.3 by an in-line resistor). Also, I checked the resistance of the senders at about 1/2 tanks and it was in the 45-50 ohm range - I didn't expect them to be absolutely identical since the arms may be slightly different in their bends, etc. and that's compensated by calibration at the EFIS. With full tanks, one measured 3 ohms and the other 5 ohms.
So now I least have an idea of what "normal" voltages and resistances look like. I've got some flying to do tomorrow, and if it acts-up again, I should be able to determine if it's definitely the sender by re-checking.
I'm wondering if perhaps there's something mechanically wrong with the sender - loose wiper arm that sweeps the resistor windings, etc. - and it gets "shaken loose" by flying. Guess I'll find out tomorrow!
Seems like you may have a poor ground connection from the tank/sender to airframe or back to your ground buss. Since the voltage rises at the input of the RDAC as the VDO resistance increases (towards empty tank) a poor ground would cause a similar indication.