What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Electrical Sender Grounding and Fuel Tank Sealing question

precession

Well Known Member
Apologies in advance if this is a dumb question or making a mountain out of a mole hill. I'm not the original builder, but I do have the plans/drawings for the RV-4, and in addition to having read them and the Vans SBs, have spent quite a while reading threads on here about fuel tank sealing, sender grounding, etc.

So I'm getting ready to install a new fuel sender onto the RV-4 fuel tank access plate (T708), after complying w/ tank SBs (anti-rotation bracket, etc.). I will also be re-attaching & re-sealing the tank access plate to the tank rib.

Question: A couple of posts here recommend using a lock washer on one of the screws that attach the sender to the access plate. The idea being that the sender must be grounded to the access plate, and the presence of tank sealant around all the attaching screws, and between the sender and the tank access plate, might impede an electrical connection. Presumably, the installation of a washer will provide a good connection.

As originally built, the plane had cork gaskets on both the sender and access plate, but now I’m planning on just using Vans tank sealant, no cork gaskets. It’s my understanding that is Vans current recommendation.

Of course, once the sender is grounded to the access plate, then obviously the access plate will need to be grounded to the wing rib in order for the sender to actually be grounded, and I think I also read one post saying to use a lock washer on one of the attaching screws of the access plate, for the same purpose.

So I’d like to hear if people think this is a good idea for accomplishing the necessary electrical connection. I’m wondering why Vans doesn’t give any guidance on it either way (to my knowledge at least).

Actually, since when screwing down the access plate apparently I’m supposed to leave a small gap between the access plate and the tank rib (gap filled with sealant of course, but wide enough to allow a knife to be inserted between the two plates in case the access plate later has to be removed), I’m wondering if it wouldn’t make sense to place washers on the screw shafts BETWEEN the access plate and the tank rib. That way, I could go ahead and tighten the screws right down, and the lock washers underneath the plate would provide me with a small but uniform thickness gap between the two plates.

Comments or recommendations would be appreciated.
 
Last edited:
The screws make a solid ground connection as they bite into the nut plates on the sender to rib and tank to wing. The screw head will ground the sender just rough up the plate a bit where the head rests. I don?t think any other ground is necessary. There are all kinds of theories on ground loops and ground differentials with airframe grounds but I would be surprised to hear anybody has had an issue.
This is why you are probably not seeing anything referenced in the plans, but remember, early RV?s like the 4 where not exactly step by step detailed instructions.
 
Most RV float senders I?ve seen are just screwed in, no special effort to ground the sender. They seem to work.

I choose however to put a #22 wire (with lug) under one of the screw heads, then proseal on top of the screw. The #22 wire goes to the common ground on the firewall.

Carl
 
Question: A couple of posts here recommend using a lock washer on one of the screws that attach the sender to the access plate. The idea being that the sender must be grounded to the access plate, and the presence of tank sealant around all the attaching screws, and between the sender and the tank access plate, might impede an electrical connection. Presumably, the installation of a washer will provide a good connection.

I?m wondering if it wouldn?t make sense to place washers on the screw shafts BETWEEN the access plate and the tank rib. That way, I could go ahead and tighten the screws right down, and the lock washers underneath the plate would provide me with a small but uniform thickness gap between the two plates.

Good questions, because a failure in the sender ground path between the flange of the sender and the screw head is the most common cause of fuel level indicator problems.

The more recent vintage plans call for a toothed lock washer under at least one screw head to help assure good ground continuity.

Adding washers between the sender flange and the cover plate could be done as extra insurance. It was considered for the plans for just the reason you mentioned (aid in removal) but filing sanding a bevel on the edge of the sender flange was chosen instead.
 
How I did it

In theory the airframe metal will carry the return path. Experience is that over time resistance builds up and can open circuit.

Best practice is a separate ground to the sending unit. See Scott's post for recommended Van's method.

My RV-6 has been flying over 20-year with the same setup I am using on my RV-8 build.

Here is a photo of my RV-8 left wing tank sending unit ground.

2018-05-04_15.20.45.jpg
 
The problem that a lot of people run into when using the recommended / kit supplied screws is that the heads are large enough, that if sealant squeezes out the screw hole before inserting the screw (common), there will be a very thin layer of sealant trapped between the screw head and the sender flange.

This is very good for the prevention of leaks at the screws.
It is very bad when screw head contact with the sender flange is needed for a good ground connection.

That is the benefit of the toothed lock washer. It will provide a good electrical bond even if there is sealant present.

Adding a separate ground wire is another way to mitigate the problem but you still need to assure that there is not a film of sealant between the sender flange and the ring terminal.
If I was using a separate wire, I would still use a toothed lock washer.
 
Last edited:
Scott plus Gary = 200% of people I trust.
I have an intermittent sender. I will clean up one of the screws, clean, reseal and add the star washer. I?ll report back if it is a ground issue.
 
In theory the airframe metal will carry the return path. Experience is that over time resistance builds up and can open circuit.

Best practice is a separate ground to the sending unit. See Scott's post for recommended Van's method.

My RV-6 has been flying over 20-year with the same setup I am using on my RV-8 build.

Here is a photo of my RV-8 left wing tank sending unit ground.

2018-05-04_15.20.45.jpg

The ground wire is on a lug like this one. Yes there is good electrical contact using an ohm meter.

Picture%20013.jpg
 
The ground wire is on a lug like this one. Yes there is good electrical contact using an ohm meter.

Nicely done.

Note that Gary's solder lug installation is also environmentally stable. Long term, it is unlikely to add resistance due to any reasonable level of corrosion.
 
Wow, what a great set of replies, even including beautiful pictures no less. My thanks to all who took the time to reply, this really helps. (Glad I sent Doug my yearly contrib too - paid off in one post.)

I was even wondering what type of lock washer, and Scott confirmed star-type - I guess because those little points will always poke through any sealant and contact the underside of the screw head above and plate below.

I wasn't aware that current plans call for "filing sanding a bevel on the edge of the sender flange." I assume that means filing a bevel on the underside outer edge of the sender plate (the side that contacts the access plate) so in the future you can insert a screw driver into the bevel to pry the sender plate off if necessary(?). I think I'll do that rather than place washers underneath the sender flange/plate.

The only part I'm still having a problem with is the idea of not fully tightening down the fastening screws on the both the access plate and sender plates. I'm hoping someone can maybe restate what the current instructions are for exactly how much you are supposed to tighten the fastening screws after you have put down your sealant - i.e., how much of a gap you are to leave between the two plates - because it is really going against my grain to not tighten down the plates flush and have nice, tight, and torqued screws. ??

Reasons for the question: When cork gaskets were used, you obviously had the cork that you would feel creating compression when you tightened your screws. You wouldn't fully tighten your screws (because you wouldn't want to overly compress the cork), but you would at least feel the cork creating some compression, and that compression, along with the sealant, presumably provided some force keeping the screws from vibrating loose. But now you are not using cork, only sealant, so it seems that if you don't tighten the screws to the point that the two plates are flush, there will really be nothing against which to create any compression on the screws - only the viscous sealant, which isn't going to provide any real resistance. So (1) I will be creating a gap between the plates that probably will not be of uniform thickness, because I'll just be eyeballing it, and (2) the only thing to prevent the UNtightened screws from vibrating loose will be the grip of the nut plates and the sealant that will harden around them?? Is that enough? And besides, don't you need some pressure on the screw or screws with the star washer under them to insure firm contact from the underside of the screw head to the washer to the plate below it?

Maybe my instincts are just wrong on this point, and I haven't seen the current instructions, so thanks in advance for any additional comment.
 
Last edited:
I never heard of bevels or leaving screws loose.

I built a lot of tanks with no leaks with a straightforward approach:
- I use the cork gasket on the inspection plate. I add pro seal to the cork gasket on both sides, as well as the tank opening and plate. My thought was that if I ever had to disassemble a razor blade cutting the gasket would work. Never had to do this so I cannot comment on how well it would work.
- For the fuel sender I first used the rubber gasket, again with proseal on all mating surfaces. The screws tighten enough for the gasket to compress. Of late I?ve not used the rubber gasket (per the Van?s instructions). No leaks with either approach.

The only tank I?ve seen leak at the fuel sender is one the buildier made using only the rubber gasket - and it was leak free for 13 years. Pulled the sender, threw away the old gasket and put proseal on the mating surfaces - all is well.

Bottom line, this is fairly straight forward. Press on.

Carl
 
I would take the "screws loose" comment as -

Do not torque down and just tighten the screws until you have a 'squish' of Proseal around the entire circumference of the sender/access plate.

Not quite the same as "loose", especially after the Proseal sets. :)
 
I would take the "screws loose" comment as -

Do not torque down and just tighten the screws until you have a 'squish' of Proseal around the entire circumference of the sender/access plate.

Not quite the same as "loose", especially after the Proseal sets. :)

Exactly

Makes the cover or sender much easier to remove.
 
Okay guys, thanks for the additional replies, they are much appreciated.

Based on them, I will go ahead and attach both the sender and access plate using just the tank sealant as my gasket (i.e., won't use the rubber on the sender plate, or cork on the access plate, even though I bought the cork and appreciate Carl's description of how to apply proseal when using cork).

I will tighten the screws until they are close to, but not totally, down and I get a hopefully roughly uniform amount of squished out sealant around the perimeter of the plates. I'll have one or more star washers on both plates to achieve the electrical connection.

Thanks again for the excellent help and alleviating my concerns. Really need to get this bird back in the air.
 
Last edited:
For what it's worth, I've mentioned this product:
https://www.permatex.com/products/gasketing/permatex-permashield-fuel-resistant-gasket-dressing-flange-sealant/

in several prior threads on sealing the access cover and sender. I've had it in service on my -4 for a number of years, and never had a leak. I *have* had to re-open a tank to fix additional *proseal* leaks, and the really nice thing about this stuff is that it never hardens. Gentle pressure is all that's needed to separate the parts later (even years later). It's impervious to both av and mogas. My technique is to spread a thin layer on both surfaces, as directed, and then a small drop on each screw hole (enough to cover the hole) prior to inserting the screw. Tighten screws to normal torque; the sealant fills any microscopic voids but allows normal torquing of the screws. I haven't done it, but you should be able to use the star washer & eyelet for external ground wire, if desired.

Available at most auto supply stores.

Charlie
 
You will need a ground cable from the sender to the tank-structure.
Dont ask how I know ... It took a while to find this out. It was wheater related, high humidity it was ok, dry it was not grounded. If you where measuring ... then it was nealy ok ... I could only see, that I could not measure it twice the same Ohm value ... Then after a ground strap was added, no problem anymore.
 
Scott plus Gary = 200% of people I trust.
I have an intermittent sender. I will clean up one of the screws, clean, reseal and add the star washer. I?ll report back if it is a ground issue.

It?s all good until it isn?t. I promised I would report back.
I apologize for my assumptions that a separate ground isn?t necessary. After more than 700 hours trouble free, things changed. I added a temporary ground for a long trip as I didn?t want to take out a screw, seal it, and then worry about a leak while the airplane was sitting on the ramp for a few days.
The temporary ground solved the issue. No more odd sender issues. A permanent, separate ground will be added this month during my CI.

Thanks Scott and Gary. As usual, ur right!
 
The instructions on my 1990 RV4 kit are sketchy. Is the recommendation to NOT use the rubber seal with the sender - just use proseal? I am just about at that step. This is a great thread.

Also, those toothed lugs are cool - where does one get those?

thanks for all replies on this great thread. I dunno what I would do without VAF.
 
Erratic fuel sender

I have been having issues with my right fuel tank sender on my RV 14A. The sender sometimes read fuel then it tells me it needs calibrating then reads fuel. Have been thinking the float is hanging up on inside rib? But perhaps the sender isn't grounded properly? Thoughts?
 
More than one way to skin this cat. However, I can't see any downside to putting in a return wire just to make things sure. I built my -6A without and the Blue Mountain EFIS did not register, even though I had ground continuity and could see the changes on a multimeter as the sender was moved. So I put in a return, which also didn't work until I isolated the sender using shoulder washers under the screws (the sender had the rubber or whatever gasket under it). Now working since 2008 no problem. Also, my end plates are in using the cork gasket and fuel lube with no problem since new. I removed the right side once to install the return from the fuel purge shortly after I did my first flight but otherwise have not touched them. It is my opinion only that gasket failures are due to improper installation but the proseal method works and is fairly foolproof so I have no issue with those that use it.
 
Fuel sender grounded?

I just installed the Fuel Sender units in both wings and I have now found the center screw is grounded on both wings. Before I used the pro seal the center was not grounded. I did not test the readings as the float was moved from empty to full positions. What did I do to cause both units to ground the center screw when I used the pro seal?
Thanks
 
I just installed the Fuel Sender units in both wings and I have now found the center screw is grounded on both wings. Before I used the pro seal the center was not grounded. I did not test the readings as the float was moved from empty to full positions. What did I do to cause both units to ground the center screw when I used the pro seal?
Thanks

That’s exactly what you want, you want continuity to ground on the center screw.
 
Dumb question: can I remove one of the screws on the fuel sender in order to put a ground wire connector on it without emptying the fuel tanks? And do I have to put pro seal on the screw when I replace it?
 
Dumb question: can I remove one of the screws on the fuel sender in order to put a ground wire connector on it without emptying the fuel tanks? And do I have to put pro seal on the screw when I replace it?

No, and yes.

Longer version: If you pull a screw with the tank full, it is going to leak. Then you may have a problem re-sealing the screw hole, because the proseal you use to seal the screw isn't going to set properly if it is bathed in fuel.

You need to drain the tanks. The important thing is to provide a good ground between the airplane and whatever containers you use. I suggest doing this process outside.
 
A follow up question - to make sure grounding was my problem I held a temporary grounding wire from the sender plate to the airframe - in various places. I expected the gauges would move to the correct indication. Each tank had about 7 gallons. Nothing changed on either side. Then I removed the wires from the center screws of the sending units, again expecting to see gauges move to full or empty and again no change with either side. So now I'm wondering if the problem is the ground. I can hear the floats moving when I rock the wing so I don't think they are stuck and the gauge indications do change on the G3X, but they are not even close to accurate. I don't want to drain tanks and pull screws if it's not the problem. Thoughts?
 
Resistance between screw terminal on the sender and the sender flange should be about 30 ohms when full, and about 240 ohms when empty.
With that info you should be able to do some measurements and determine whether it is the sender, the sender ground to the airframe, or maybe a corrupted look-up table in the EFIS (I have seen it happen before).
 
Back
Top