Online Community of Zenith Builders and Flyers
From time-to-time there have been reports of the VDO fuel senders leaking or even the central electrode falling out! These senders were OEM for VW's and intended to be top-mounted on the fuel tank. When side-mounted in the Zenith tanks, they are continuously immersed in fuel under slight pressure from the weight of the fuel. My VDO senders are at least 10 years old, installed for 8 years, and have never given a problem ... till now!
I recently flew to Mexico, MO for the Zenith Homecoming and when I returned, began my annual. I opened the access covers on the bottom of my STOL 750's wings and the senders were bone-dry without evidence of leaking, nor was there even a hint of a whiff of gas fumes. Since all was good, I buttoned that area up! However, I had left the tanks about 1/4 full on both sides when I started the annual. Afterwards, I topped off both tanks. The next day, I noticed a drop of fuel on the hangar floor near the aft edge of the wing near the root. Looking up, I saw a drop of fuel hanging from the inboard flaperon bracket. I opened the access panel again and the sender was very wet with fuel! It also appeared the white insulator was slightly shifted or cocked from its usual position. I touched the insulator and a piece about the size of pea crumbled off! Yikes! It was obvious the insulator was deteriorated and I immediately set about draining the tank with visions of the insulator popping out and dumping a tank of fuel into the wing root! Fortunately, that didn't happen.
Here's a pic of the crumbled insulator: (Ignore the AN3 bolt - I had already removed one machine screw and just used the bolt to serve as a "handle" when I manipulated the sender out of the hole.)
Bad enough, right? No, it gets worse! After draining the tank, I was walking around the plane and saw a drop of fuel on the floor in a similar location on the right! I opened the right access panel and the right sender was leaking, too! The insulator didn't seem as deteriorated as the left - it didn't crumble - but it did feel slightly loose and could be slightly rotated.
Obviously, the fuel top-off triggered the leaks. What are the odds that I could have just completed an 800 nm round trip with 3 refuelings and the senders wait to leak till I get home - and both senders at that! I felt the angels were truly with me on that last trip!
So, I have drilled out the rivets enough to fold back the top wing root covers and remove the VDO senders. I plan to install the Stewart Warner senders the Van's RV builders have used for years. They appear to be of a much higher quality and use a "thick film" resistor that is superior to the VDO's wire-wound resistor. It's a big plus that they mount in an absolutely identical manner as the VDO's (they even use the same screws and mounting ring). One of our members, Mark Pensenstadler, has an excellent YouTube channel, Kitplane Enthusiast, and has a detailed video on these senders here.
After researching this somewhat, I am going to mount them the way Van's recommends - no gasket and use ProSeal instead. I'll also ProSeal the screws to prevent leaks. Obviously, one needs to dry fit and test the resistance, etc., before final installation as it would be very difficult to remove a sender later and no way am I going to drill out that wing root skin again! Ha! Although these senders seem very reliable, If a sender ever failed, but was not leaking, I'd be very tempted to simply leave it and cut an access panel in the top skin over the tank and mount a sender there!
I also plan to use my flexible "snake" video camera to verify the float action during a dry fit test. You want the float to both touch the top and bottom of the tank to get the maximum measuring range. I saw a post elsewhere where a builder dry-fit the sender and left a screw out and passed a piece of safety wire through and tied it to the float wire. He could then verify the float had full range of motion by hearing it tap against the top and bottom skins as he pulled the safety wire. Obviously, I could also verify this by observing through my "snake" video camera.
Just thought I'd put this out to remind everyone to check those VDO senders very carefully! If yours is leaking, I would recommend to definitely drain the tank below the level of the sender before probing around that white insulator!
Update - Today I did fill the tanks, checked for leaks, and calibrated the senders as I had planned. Thankfully, there were no leaks - Just to be extra-cautious, I plan to let the tanks sit full overnight and re-inspect the senders tomorrow.
One glitch did come up, however. The left tank sender seemed to get stuck when the tank was near-full. I could wiggle the wing, slosh the fuel, and it seemed to unstick. I also found that I could simply press the tank wall above the sender - the wall would oil-can or pop inwards and this would free the sender. I was puzzled by this, having been very careful to measure for clearance, etc., but I think I've figured out what happened. When filling the tank, the skin the sender is mounted into oil-cans outward. This slightly tilts the sender, lifting the float arm - apparently enough that the angled portion can contact the top of the tank and stick. I've hopefully solved this by wedging a rubber ring of thick-walled hose (about 2" in diameter) into the space between the rib and tank wall above the sender. This supports the tank wall and doesn't let it oil-can or pop outward. I put some rubber cement on this ring of hose to ensure it stays in position. Tomorrow, I'll drain the tank and check to be sure the float doesn't stick again.
I had no problems on the right, but I glued a rubber support into the same position as the one on the left just to ensure the tank wall can't oil-can outward like the left did. I don't know if this oil-canning is unique to my tanks, but it is likely a lot of these tanks do this since both of my tanks exhibit this to some degree.
As a consequence, I would recommend bending the 65 degree angle via a rounded arc rather than a sharp bend so as to lower the possibility of the float arm sticking, or make the angle in 2 or 3 small bends rather than one. This would have the effect of slightly shortening the overall length of the proximal portion of the float arm and ensure more clearance.
One think I didn't do that I wished I had done now is dry-fit the SW sender and leave one of the top screws out and thread a safety wire through and attach it to the float arm. Then, one could pull on the safety wire and inspect for clearance with a snake video camera. Of course, you might have to pull gently outward on the sender (you could use an AN3 bolt in the threaded central electrode hole as a handle) to see if the skin was going to oil-can outward and still have sufficient clearance.
Fingers crossed for tomorrow's re-check of the float arm!
Thanks for the info, John. That is my next project.
Well, sitting around waiting for the ChemSeal to cure is just about as exciting as watching paint dry! Ha! To further add insult to injury, it seems most builders add a few days to the cure time before testing for leaks. After all, if there is any possibility of incomplete mixing of the 2-part sealant, or the temps aren't optimal for the cure, waiting is cheap insurance! I felt good about the ChemSeal being mixed well as it did appear to be a very consistent gray color (looks like JB Weld!). However, our daytime temps are in the 70's (and the insulated hangar gets a little solar boost through the translucent fiberglass doors - temps get up to 83 or so lately) but at night, we're now going down into the 40's in East Tennessee, so rather than the 72 hours the B2 sealant calls for, I'm going to give it some extra time, although it does appear and feels to have set up firmly as of today. Each night, I've placed my "Hornet" explosion proof heater near one or the other sender to keep it warm to try to ensure a good cure.
So, since I'm twiddling my thumbs right now, I decided to re-insert my "snake" video camera and inspect the finger screens for debris to see if they needed removal and cleaning. I was impressed at how clean the tanks are - not surprising since I have a filter on my fueler trailer so the gas gets a final filtering before entering the tank - I've noticed that my in-line fuel filter in the aircraft is always clean, too! I inspected the screens and they were perfectly clean - here's one of them:
So, still waiting a bit on the sealant to cure - guess I've run out of excuses to start cleaning up the hangar! Ha!
John, Just be careful with the sealant getting into your tank. A guy in my EAA chapter had a Sonex and some sealant broke loose, clogged his finger screen, made the engine come and go and trying to make an open field hit a tree and was killed. The previous owner had put sealant in the tank. There must have been a lot though. I like the picture inside your tank.
Yes, the interior of my tanks were extremely clean and I specifically closely inspected the interior of the senders with my snake video camera after clamping them into place with the sealant. I wanted to be sure there was no excessive squeeze-out of the sealant into the tanks, and there was hardly any sealant visible at all!
Finally got everything buttoned-up and re-riveted the top wing root skins. The SW senders have remained bone-dry since filling the tanks, so no problems there. Today I flew for nearly an hour - long enough that I could burn some fuel off each tank and see how the senders worked. They worked just fine! The MGL XTreme's EMS has an electronic filter to reduce fluctuation of the indicated fuel levels - I assume it more or less averages the readings the higher you turn up the filter. I turned the filter "off" and could see the levels fluctuate quite a bit, indicating the float arms are swinging freely. All in all, I"m quite pleased with the switch-over to the SW senders, especially since at least one person told me they had the same senders and they were working great in their plane after 25 years! :>)
Today I flew again and when I got back, the fuel flow totalizer (which has always been extremely accurate) indicated I had burned 8.1 gallons. When I checked the MGL EMS's fuel level indicators for the left and right tanks, the amount that I had used from the two tanks totaled exactly 8.1 gallons! At least so far, the calibration of the senders is uncannily accurate! Fingers crossed that this accuracy will continue as the fuel levels decrease to near-empty.