Fuel valve rebuild and manual mistake WARNING

[Bruce Fenstermacher]

Repair was done by owner/licensed mechanic - discovery of error by NTSB. Just trying to understand if misassembly would influence operation.

bev5887 could you please elaborate what you mean by "- discovery of error by NTSB."

[blueldr]

Miles,
You seem to have come up with a rather astute answer to the method used by Cessna for the installation the fuel selector valve. I never thought of suspension by a thread. I always assumed that Cessna had secretly discovered a method of suspension through nullification of gravitational force and the services of one of their crew of three handed assemblers.

[bev5887]

Thanks for the thoughtful replies. We were just trying to make sense of findings of NTSB for N28RP. Earlier post seemed cautious about someone getting in trouble. While the valve may have been misassembled, agree that it was probably no better or worse than before repair.

[Bruce Fenstermacher]

Just looked at that report. Interesting, actually pretty unbelievable,that Cessna hasn't corrected this IPC mistake at least up to the 172F model, which this aircraft was, as reported by the NTSB.

The Report says the balls were riding high and wouldn't open that side of the valve. I'll bet that was because of the damage not the two washers. As I recall the cam will only go so far into the valve body and the cam would have to line up with the balls when it is seated. The washers on top couldn't push the cam lower into the valve body allowing the balls to ride high. It's been a while but that is how I remember it. I seriously doubt the presents of two washers had any thing to do with the operation of the valve.

I also doubt whether the valve,working or not, had anything to do with this accident. If I read the report correctly this guy decended from 7000ft at 6000ft a minute to impacted. That is no where close to an engine out glide.

[falco]

Ever since I first had to do maintenance on my fuel valve, I've had this picture in my mind of the very beginning of the Cessna 170 assembly line being a fuel valve hanging by a thread from the ceiling, then the airplane being build around it.

Miles

Ford had the same sort of procedure... start with the heater core.

[GAHorn]

I just rebuilt my selector as it was leaking. It is a job to get the valve out as it is hard to get a wrench on the bottom nuts holding the valve in place. I called Sac sky ranch and he didn't have the o-rings but gave me a number to try. El Cajon Aircraft Supply 1 800 888-3227. He had all the parts and yes, I could buy only one of the gaskets. His price for the gasket was $3.94. I ordered double the parts to have a spare set. The price with postage ($1.00 as I asked him to just put it in an envelope and send it US mail) came to $14.40. So one set was only $6.70. I ordered on Friday and had them Monday in the mail. This was April 2004. It is nice to have a valve that doesn't leak.
GARY 1909C

El Cajon is not quite so agreeable these days. I called them and they will NOT ship except via Fed Ex...AND their price now for the GASKET ALONE is now $29.00 !! (Cessna's direct price is $23.) They offer only Nitrile O-rings, no Viton. (Aircraft Spruce does not handle the gasket OR the shaft seal O-ring.)

The best bet is to contact McFarlane and obtain their Viton o-ring complet kit (link posted previously) for only $27 plus $3.98 Priority Mail.) http://www.mcfarlane-aviation.com/Produ ... r=FSS-KT-2&

[GAHorn]

I have read this thread and would appreciate any thoughts. What could be the potential effect, in flight, of a misassembled fuel selector valve? What would happen if two washers were installed on the cam shaft below the o-ring that seals up against the valve cover when the valve is assembled? What happens if the o-ring is loose within the valve body due to mis-assembly? Thanks

As for your question of two washers on the cam shaft (#8) under the o-ring. I do not recall what the tolerance is of the cam shaft from the bottom of the valve to the top plate. This tolerance would also change depending on the thickness of the gasket (#10). If a thick gasket was used two washers might be required to get o-ring to seal at the top of the shaft.

What would be the effect of two washers if used? Well first I'd think that if two were used it means that using one wouldn't seal and I'd have to find the reason why. Second once satisfied why two washers were required and having inserted two washers the valve would either work properly meaning the cam would turn and the shaft wouldn't leak or the cam would be to tight and not turn and the valve would not function being stuck in what ever position it was assembled. I'd think this malfunction would be pretty obvious before one got to far.

I won't comment on whether it would be proper to install two washers in this position other than to say if the cam turns and doesn't leak there would probably be no further consequence. I'd probably exhaust all attempts at getting a thinner gasket for the top plate before installing two washers however.

Just looked at that report. Interesting, actually pretty unbelievable,that Cessna hasn't corrected this IPC mistake at least up to the 172F model, which this aircraft was, as reported by the NTSB.

The Report says the balls were riding high and wouldn't open that side of the valve. I'll bet that was because of the damage not the two washers. As I recall the cam will only go so far into the valve body and the cam would have to line up with the balls when it is seated. The washers on top couldn't push the cam lower into the valve body allowing the balls to ride high. It's been a while but that is how I remember it. I seriously doubt the presents of two washers had any thing to do with the operation of the valve.

I also doubt whether the valve,working or not, had anything to do with this accident. If I read the report correctly this guy decended from 7000ft at 6000ft a minute to impacted. That is no where close to an engine out glide.

Well, Well, Well,.... This little discussion suddenly had interest to me because as I took my valve apart...it had TWO WASHERS under the camshaft O-ring. One was the one called for in the IPC, and the other a smaller one.

I'm trying to figure out what to do with it as I type. (I have my valve on the bench being re-assembled now.)

[GAHorn]

OK, in an effort to get these online before I lose them, here are pics I took of my fuel selector valve rebuild. I will be editing this msg as I develop commentary I hope will be found useful.

Firstly, I want to Thank Bruce for instigating this thread, as I think it is likely that many of our members/owners will eventually discover to their chagrine that this is a project they will probably suffer.

As already mentioned in this thread (Thanks to all who contributed and Thanks for allowing me to plagiarize a bit), the Illustrated Parts Catalog, B-model, Fig 64 has an error in the check-ball retainer/O-ring relationship. ALL THESE PICS MAY BE ENLARGED BY CLICKING ON THEM. FURTHER ENLARGEMENT MAY EXIST BY A SECOND CLICK.

This IPC Illustration is WRONG, showing retainer/O-ring reversed, and NO CAM BUSHING.

The IPC has the retainer (Item 17) shown as next to the large check-ball (Item 18) instead of the O-ring (Item 16). Ok, so that's now well-known that they are actually reversed so that the O-ring is NEXT TO the check ball in order to provide a soft sealing surface for the ball. The retainer merely provides a seat in which the O-ring resides.
As already mentioned by Gary, DO NOT ATTEMPT TO REMOVE THE RETAINER from the valve-body. Instead, use a dental pick or other sharp tool to remove the O-ring from the retainer and valve body. Toss that O-ring in the trash and replace it with new.
When reinstalling the check-ball O-ring...DO NOT USE ANY TOOLS which might injure the new O-ring. Instead, use a drop of clean engine oil on your finger and pick up the O-ring and place it into position in the retainer USING ONLY YOUR FINGER. One side of the O-ring will enter easily (with the O-ring becoming momentarily deformed) then your finger can easily press the protruding side of the O-ring into position.
In fact, use clean engine oil when installing all the O-rings in this valve.

NEWLY DISCOVERED POSSIBLE-ERROR in the IPC versus the Service Manual Illustrations: Compare the IPC Fig 64 with this Service Manual illustration (Fig.13-9, Item 22) of the same valve (also used in 172/175 aircraft.)

This Service Manual illustration is CORRECT...but no part number supplied for the (perhaps necessary) bushing.

See that BUSHING BENEATH THE CAM? (Item 22) It MAY be necessary in your valve...it may not...but ... That bushing is totally missing in the IPC illustration. It is used to elevate the cam so as to properly engage the smaller fuel valve balls in their outermost circumference. (I suspect a missing cam bushing might be the reason for the NTSB report*** previously mentioned in this thread.) The bushing was found ERRONEOUSLY PLACED in my valve on TOP of the cam (between the cam and the washer just beneath the shaft O-ring. Was this what the NTSB thought was a second washer in the accident aircraft? We'll likely never know.)

Notice the bushing improperly placed beneath the AN960C416L shaft-seal-retaining washer. IT SHOULD HAVE BEEN BENEATH THE CAM!

Except for the generous thickness of the operating lobes of my cam, it's possible that my valve could have suffered a similar failure, only needing sufficient vibration to eventually wear the cam to descend below the circumference of the smaller balls.
SUGGESTION: There is not available any detailed overhaul/rebuild manual for these valves. I suggest that a digital micrometer be used to measure the depth of the cam's upper ring versus the lower, thickened surface of the top-plate, where the shaft seal resides. The thickness of the cam-bushing will be determined by the need to keep the operating lobes of the cam directly at the largest circumference of the small balls (which in-turn press the larger check-balls away from the retainer o-ring seat, to allow fuel to flow thru the valve.
The bushing which was mis-placed in my valve appeared to be slightly rough, which led me to believe it was locally manufactured. (Which made me wonder even more as to why it was even utilized inside the valve. I think I now know. I believe the previous rebuilder of that valve thought the shaft seal o-ring needed more support due to the increased distance between the cam upper-ring and the washer (item 8 in the IPC), otherwise the shaft seal o-ring might be allowed to fall below it's residence in the top-plate. Up/Down movement of the cam might eventually dislodge that shaft seal O-ring unless the washer(s) beneath the shaft seal adequately took up the available space between the cam upper-ring and the seal.)
My bushing measured .050". My washer (item 8 ) measured .030" for a total thickness of .080". The total distance between the top-plate and the cam upper-ring was .082", therefore .002" of free-play exists in my valve. YOUR valve may differ....therefore I suggest that after assembling the cam on it's bushing to adjust the cam-lobes relationship with the small balls... use whatever additional AN960C416L washers necessary to take up the distance between the cam top-ring and the base of the top-plate. (DO NOT forget to consider that the cork gasket (item 10) which will then provide sufficient clearance to avoid binding of the fully-assembled valve.) IF YOUR VALVE REQUIRES A BUSHING to elevate the cam to fully-contact the balls ... you will not find a part number for it. But a common AN960-416L (replace the hyphen with a "C" for a preferable stainless version) washer or two will suffice. They typically measure .025 to .029 and your digital caliper can assist you in determining the necessary thicknesses or number of washers req'd. Additionally....keeping in mind that not all valves may require such a bushing... you can visually inspect the contact between the cam lobes and balls to confirm correct engagement prior to final assembly.

Finally,...I used fuel lube to lubricate the cam-shaft between the shaft seal o-ring and the shaft to insure against premature failure due to friction. No other lubrication (other than the aforementioned engine oil on o-rings) was used. No gasket compound or sealant is used.

Removing the valve from the airplane required 2 hours. RE-installing the valve was harder than the removal. It took 3.5 hours assisted by some choice vocabulary. Rebuilding the valve itself was a walk in the park compared to the labor of removing/re-installing that valve. The pics will illustrate why. The two nuts which secure the AN3 mounting bolts (IPC Fig 63, Item 25) not shown in this msg-thread are the DEVIL to remove and reinstall. The "closeup attach nut grrrr" pic only displays (barely) the right/aft nut.

The red arrow points to the right/aft mounting nut. It cannot be accessed with an open-end wrench because of the bulkhead just aft (out of view.) The left/aft mount nut is completely invisible and must be manipulated blindly.

It's a BEAR. That nut is so close to the valve body that an open-end wrench cannot simply be slid over it, but must approach it from the right-rear of the valve...problem is...there's a bulkhead just aft that prevents getting a wrench in position, ...hence my twisted-up, bent-over-double little wrench I made (see special tools.)
But the real challenge is it's partner on the aft/port side of the valve which is totally non-viewable and requires a Midget-Superman to access, remove and install, using the "special tools" I made for the purpose. The control cables which pass above/beneath/around this valve do not help.

I did this alone (except for my properly-certificated supervisory mechanic, of course), but if you have a helper to open the tunnel plate (beneath the flap handle) and use a 1/4-drive socket set with extensions and universal-driver... it might be a bit easier.
I personally like the nut-plate arrangement that has been shown here previously by DaveF, but I think that it may not be necessary to actually fasten them to anything, if all one wishes is for a nut that doesn't require a wrench (special tool) to hold it from turning. All that is probably necessary is to use two-lug MS21069/NAS697 (10-32) ...or MS21075/NAS1068 (10-32) nutplates. The rivet-lugs alone will resist turning when they contact the valve body. (However being the originality-NUT that I am, I kept the AN365-1032 fiber lock nuts. Dang, I hated them in this job!)

Addt'l NOTES: The valve top-plate bolts BENEATH the fuel mount bracket in the airframe. Be CERTAIN to support the ball-and-spring housing containing the roll-pin while removing/re-installing the roll pin. Do not merely use a hammer and punch to drive it in/out or you will possibly stress the shaft seal and cam. Suggestion: Lay the housing on a vise or block to drive the pin out with a punch. When re-installing the pin, place the entire housing and pin inside the vise jaws and squeeze it back into position by closing the vice jaws upon it...or using a C-clamp to accomplish the same act.
Lubricate the housing/spring/ball with graphite-based grease, or with Lubriplate and dry graphite mixed with each other.

As mentioned by Mike, the BEST SOURCE for the fuel selector valve complete rebuild kit is:
http://www.mcfarlane-aviation.com/Produ ... r=FSS-KT-2&
It's street-price is about $28. Cessna charges that just for the gasket.

McFarlane also offers an overhaul kit for a bunch more should any of your hardware be unserviceable.
And they'll do all the overhaul/rebuild on YOUR valve for $500 appox (the easiest of this entire task, by the way. You still have to do all the hard work of removing/reinstalling the valve.)
And finally, they'll even exchange your valve for a rebuilt/overhauled one for only $1250.00 exchange. Right.

FINAL NOTE: Triple antibiotic cream will help the backs of your hands heal from the scratches.

Here's what you see when you remove the inspection cover beneath the valve...not nearly sufficient to access and work on this valve.

This is the view obtained by sticking my camera up thru the forward inspection cover...NOT a view available to the naked eye.

Check-Ball retainer O-ring in position.

Check-Ball retainer with O-ring removed (by dental pick.)

Notice the bushing improperly placed beneath the AN960C416L shaft-seal-retaining washer. IT SHOULD HAVE BEEN BENEATH THE CAM!

Looking up at the tunnel/flap handle from beneath. Not much access.

Another view of the tunnel from below. What a rat's nest to work in.

Cheap chineese ignition wrenches were bent to hold the nut from spinning while turning the bolt from above. The dental pick was used to remove the check-ball retainer o-ring. Nutplates are possibly more convenient to use in lieu of the original fiber-lock nuts. Note the valve top-plate is mounted BENEATH the mounting bracket.

***NTSB report excerpt, C-172F accident: "The fuel selector valve was sent to the NTSB Materials Laboratory for further examination.
According to the Materials Laboratory Factual Report, the damage to the cam shaft was consistent
with mechanical overload. Two washers were observed on the cam shaft below the o-ring that would
seal up against the valve cover when the valve is assembled. The illustrated parts catalog for the
airplane indicated only one washer in the illustration and accompanying parts list. The additional
washer would locate the actuating cam slightly lower within the valve body. (bold emphasis =mine, gh)
The actuating cam displayed a black colored smear on its lower portion, consistent with material transfer from the
erroneous presence of a loose o-ring within the valve body. The illustrated parts manual
incorrectly depicts the o-ring installed ahead of the retainer. To correctly assemble the valve,
the retainer should be installed ahead of the o-ring. The o-ring found in the valve body was
examined under stereo microscope and no discernible wear was observed. The condition of the valve's
internal components and the loose o-ring found within the valve body did not provide adequate
information to establish whether or not the valve was functioning properly prior to the accident."

I do not believe the NTSB is correct to say that two washers will relocate the cam lower within
the valve body. This would depend upon whether or not the cam was already resting on the bottom
of the valve body (as it was in mine) and also would depend on whether or not the two washers actually
had sufficient combined thicknesses to take up more-than the available space between the cam top-ring
and the valve top-plate (mounting plate or valve-cover as they like to call it.)
Simply because two washers exist below the o-ring that seal the "valve cover" does not automatically
force the cam lower in the valve body, expecially in light of the ordinarily-considerable space in that area
of the design. (My own valve had operated so-assembled for many years without any complication whatsoever
and with adequate engagement between cam and ball(s).
In fact, the housing spring-and-ball actually pull UP on the cam after the valve is assembled, and as long as the
two washers do not exceed the dimensions of the distance between cam top-ring and "valve cover" then the cam
cannot be forced "lower in the valve body" than the bottom of the valve body allows. No mention of that, or
the lower cam bushing is made by NTSB.
However, this is an important point to consider when inspecting your valve. Be CERTAIN YOUR CAM FULLY
ENGAGES the small engagement balls in the Left/Both/Right positions...before final assembly.

[jrenwick]

I hope you'll publish this in the 170 News, George!

[Bruce Fenstermacher]

Another good catch on the bushing.

Looking at the IPC and Service Manual pages it looks like the bushing would be pressed into the bottom of the valve body and truly be a bushing not a spacer which might be a better description in this case.

You might want to develop your wording on that bushing. You mention and picture it in the wrong place and while you describe why the bushing should be under the cam to align the cam lobes to the balls, you don't actually say directly it should be under the cam as depicted in the service manual. I knew from looking at the pictures what you were describing but read the text several times to follow your description and what I was seeing and that seems to me to be a missing piece.

Hopefully it won't distract from the main intent of this thread but It should also be noted for those that might be looking at their aircraft but don't have the fuel sample valve shown in one of Georges pictures sticking through the belly skin, that that valve is not a stock item. It replaces the drain plug at the bottom of the valve. This common fuel sample valve and STC are available at about $25, last I checked for B models. Some have suggested the valve could be placed their as a minor alteration. (While we are talking B models in this thread since A models have a different fuel shut off valve the placement of such a fuel sample valve is accomplished at a different place and a hole will have to be cut in the belly to accommodate it.) A fuel sample valve in this location is a great idea since this is the lowest point for water to collect in the fuel system while parked in a three point attitude.

[lowNslow]

George, great article and great pictures. What kind of camera are you using for those close ups? The pictures I try to take with my little Olympia digital come out looking like crap. I used a flat piece of steel stock about 1/2" wide and cut a notch in it to fit over the bolt head, fit into the tight space nicely and took only about 20 minutes to take valve out and 60 minutes to put it back in.

[GAHorn]

George, great article and great pictures. What kind of camera are you using for those close ups? ....

My comopolitan daughter laughs at my "clunker" camera. It's a Sony "Mavica" and uses 3.5" hard-floppys for data-storage. The autofocus is exceptional and can show the individual threads in fabric from a half-inch distance.

As an example, here's the zipper on my logbook carry-case:

[170C]

George, I concur with Bruce that this subject would be a good one and helpful to all if you would do an article for the 170 NEWS. I am curious which fuel drain valve you have in the bottom of your fuel selector valve. It would appear to be a CAV-170H-4 or CAV-160H-4 (Spruce pg.163). Wondering which one would be correct for late model B's & early 172's? I suppose putting one in would be similiar to removing & installing one on the fuel tank itself. One would need to be quick on the trigger to prevent a loss of fuel, but this installation would not be as easily accessed as the wing fuel drains. Yea, I know, do it when the tanks are empty, but when do we let the fuel get that low or want to have to drain gallons & gallons of fuel in order to install the drain valve (assuming the fuel valve itself would not prevent fuel from flowing out the bottom of the valve).

On a lighter note, I agree with LowNSlow that your photos are really good. Don't know how you are doing them-----maybe you can enlighten us You must have been one of those guys in junior high that put mirrors on the toes of your shoes to look up the girls dresses

[GAHorn]

.... You mention and picture it in the wrong place and while you describe why the bushing should be under the cam to align the cam lobes to the balls, you don't actually say directly it should be under the cam as depicted in the service manual. ....

I went back and CAPITALIZED the caption on that photo to better emphasize the point. Thanks.

That bushing is not pressed down into the lower valve-body, but merely sips onto the lower shaft of the cam, thereby limiting the depth to which the cam descends into the lower valve body. A common AN960 washer would suffice for that bushing if one were needed and not available...particularly interesting since no part number exists in the IPC for it.

[GAHorn]

George, I concur with Bruce that this subject would be a good one and helpful to all if you would do an article for the 170 NEWS. I am curious which fuel drain valve you have in the bottom of your fuel selector valve. It would appear to be a CAV-170H-4 or CAV-160H-4 (Spruce pg.163). Wondering which one would be correct for late model B's & early 172's? I suppose putting one in would be similiar to removing & installing one on the fuel tank itself. One would need to be quick on the trigger to prevent a loss of fuel, but this installation would not be as easily accessed as the wing fuel drains. Yea, I know, do it when the tanks are empty, but when do we let the fuel get that low or want to have to drain gallons & gallons of fuel in order to install the drain valve (assuming the fuel valve itself would not prevent fuel from flowing out the bottom of the valve)....

Keep in mind that the plug in the bottom of the fuel valve is 1/8" NPT. The SAF-AIR drain valve in mine is a CAV-110H-4, and the short type similar to what goes in the wing sumps is a CAV-110.

Installation is a snap as far as draining fuel concerns go....when the fuel valve is turned OFF...no fuel enters the valve. (The most you'll get is whatever drains from inside the valve...about an ounce or so.

WARNING: The pipe threads are tapered. This means that if you use a big wrench and your gorilla installs the valve...you'll likely FRACTURE that $6,000.00 fuel selector valve.

Here's 'nother warning: Don't EVER use teflon tape on fuel fittings or valves. (The teflon will shred and let particles into your fuel stream, possibly clogging your carburetor.) If you use anything at all, use Permatex No 2 ...and then ONLY on the middle two threads of the new SAF-AIR drain valve. Do not overtighten!...only "snug" . Remember, it will be safety-wired and this is only a GRAVITY feed fuel system, you aren't dealing with a lot of pressure.

While I believe this is a simple, minor alteration requiring only a logbook entry.... if you lay awake at night about such things you can go get the FULL MONTY if you buy the $50 kit from C-mods http://www.c-mods.com/ , and have your A&P/IA install it, fill out the STC paperwork, and submit a Form 337 , which shouldn't total much more than a C-note, to install that $12 drain valve. They did donate a kit to TIC170A at the Tehachapi convention, so they are a supporter. <Edit 6/2/17, C-mods has indicated to the Assoc'n they intend to close down and offered their STC to the Assoc'n. I believe the Board's consensus was not to take the offer, as this is a minor alteration requiring only a log entry.)