Stuck discriminator valve causes near-catastrophic fuel tank failure

[Grant]

At last year's Sebring race we noticed our discriminator valve once got stuck closed. The car had some winter blend fuel in it that was boiling, so we figured it created enough pressure to make the ball stick to the top, but this wouldn't be a problem going forward with the right fuel. That was wrong.

 

I recall some discussion here as to whether or not natural fuel tank pressure (with no venting) can cause the fuel tank to permanent deform, increasing volume. Well it definitely can, deforming the the tank skin closer to the driveshaft and differential flange. Then when under additional pressure the driveshaft and differential can cut a hole in your metal fuel tank, causing a massive gasoline leak all over your exhaust. You'd think these tanks would be designed to withstand stuck venting systems, but I guess not necessarily.

 

This fuel tank to diff clearance is quite a bit less than stock (NC Miata):

 

 

The hole:

 

 

I'm very lucky this happened on a cooldown lap, when nothing was hot enough to ignite.

 

We're going back to the stock tank venting system. It doesn't fill very quickly, but it never gets stuck, never pisses any fuel out on to the track (our discriminator valve sometimes would). We had even tried two discriminator valves, one 10 ORB Harmon Racing one that would get stuck with no pressure in it at all. The one we actually ran was a -12 AN Fuel Safe piece, that would at least not get stuck all on its own.

 

With a new stock tank we measured our capacity at 15.1 gallons, just under the 12.7 + 2.5 = 15.2 limit. So my guess is we ran a few races with capacity well in excess of this.

[turbogrill]

Wow! Thanks for sharing. That looks mega scary. I can only get 14.5 in my tank, maybe measuring error on my end (stock filler tube).

So stock venting is that tiny vent hose next to the filler tube? And the cap it self has a valve for vacuum and high pressure?

Can't you use both? So during fueling you use the discriminator valve and get quicker fueling but if it get stuck the tank will normalize it's pressure using stock?

 

[Grant]

3 minutes ago, turbogrill said:

Wow! Thanks for sharing. That looks mega scary. I can only get 14.5 in my tank, maybe measuring error on my end (stock filler tube).

So stock venting is that tiny vent hose next to the filler tube? And the cap it self has a valve for vacuum and high pressure?

Can't you use both? So during fueling you use the discriminator valve and get quicker fueling but if it get stuck the tank will normalize it's pressure using stock?

We have a fairly small surge tank and larger filler neck, which bumps up volume.

 

Stock venting is the hose on the passenger site of the tank with a quick connect fitting. That leads to a network of floater valves at various points, but they aren't as high as they could be. So there is probably still some air left in the tank if you only use those vents.

 

You could use both, but then I'd still have my discriminator valve occasionally pissing fuel out. I'd rather just get rid of it.

[Grufton]

14 minutes ago, Grant said:

With a new stock tank we measured our capacity at 15.1 gallons, just under the 12.7 + 2.5 = 15.2 limit.

 

I think your math is off.  Should be 12.7 + 2.0 = 14.7

[Grant]

Just now, Grufton said:

 

I think your math is off.  Should be 12.7 + 2.0 = 14.7

2L of surge tank adds 0.53 gallons of capacity which does not count towards the stock + 2 limit.

[Grufton]

7 minutes ago, Grant said:

2L of surge tank adds 0.53 gallons of capacity which does not count towards the stock + 2 limit.

 

I agree the capacity of the surge tank is not included in the limit, perhaps I misunderstood how the measurement was done.  If you drain the tank empty, and measure how much goes back in to fill it, it would need to be no more than 14.7 to comply with 9.10.2

[gavro]

1 hour ago, Grant said:

2L of surge tank adds 0.53 gallons of capacity which does not count towards the stock + 2 limit.

 

Not sure I'm correctly understanding what you're saying here, but you only get that 0.53 of added capacity if you actually have a surge tank. If you are running the stock tank by itself your capacity is stock plus 2 gallons.

[Black Magic]

His surge volume could be in the tank (shove a basket in the tank)

[Grufton]

23 minutes ago, Black Magic said:

His surge volume could be in the tank (shove a basket in the tank)

Seems common on cells, is this ever done with stock tanks?  Either way, the tank still needs to comply with 9.10.2 / 9.10.2.1 = 14.7 gallons.

[mender]

As mentioned,  shouldn't the stock gas cap vent excess pressure?

 

 

My Civic tank started to deform at 4 psi. Some research I did showed some gas caps designed to start venting at 5 spi.

Edited October 25, 2022 by mender

[Team Infiniti]

Holy crap that was close to disaster.

[turbogrill]

I guess when he filled the NEW stock non ballooned tank he filled 15.1 gallons:

- 1.9 gallons extra in main tank

- 0.5 gallons in surge tank.

So 0.1 margin.

 

However with the (involuntary) ballooned tank he was probably over the limit. This assumes the increased volume was more than 0.1

 

The NC gas cap has a bunch of stuff in it that I always thought was a vent. The passager side hose goes to carbon canister box?

 

Edited October 25, 2022 by turbogrill

[Grant]

6 hours ago, mender said:

As mentioned,  shouldn't the stock gas cap vent excess pressure?

Gas caps usually have check valves that allow air in, to accommodate the fuel pump removing volume from the tank. None of the caps I've tested allowed anything out.

 

We have a surge tank. A 100% stock, never-pressurized tank + surge tank + larger fuel neck = 15.1 gallons. However the expanded tank was certainly well above this, and not legal by 2022 rules. I don't know how long it'd been ballooned.

[Robmink]

In the stock configuration, all gasoline tank venting is done through the charcoal canister (as equipped in the US since the early 70's), both in and out.  Any venting through the cap is strictly for safety relief purposes, and the fuel tank can deform before the cap can relieve the excess pressure or vacuum.  Most plastic tanks do not want to see more than 15 mbar (0.2 psi) of vacuum, or more than 70 mbar (1 psi) of pressure. 

 

The vent system in a stock tank is quite complex and yes it does leave some airspace above the fuel level when full to allow for heat expansion.  The lower of the vent valves in the tank is the fuel fill cutoff, and the upper one is to allow venting of vacuum and pressure once the tank is full.  All of the vents act as discriminator valves, not allowing liquid fuel to pass through and also shutting off the vents in the event of a roll-over.

[mender]

18 hours ago, Team Infiniti said:

Holy crap that was close to disaster.

Maybe there are some good reasons behind going to a fuel cell...

[Team Infiniti]

55 minutes ago, mender said:

Maybe there are some good reasons behind going to a fuel cell...

I agree, but, it was a hybrid fuel cell venting solution on a stock tank, while that wasn’t the problem, the discriminator failure was.

 

I still run the stock discriminator valve inside the tank plumbed to the original set up under the hood (minus charcoal canister) with a coil of hose, fortunately it’s worked well in our application.

[Racer28173]

@Grant  After seeing what @Robmink posted above, I’d be interested in knowing what the specs were for the discriminator valve you were using.  Those pressures he listed are minuscule, in my opinion 

[mender]

6 hours ago, Team Infiniti said:

I agree, but, it was a hybrid fuel cell venting solution on a stock tank, while that wasn’t the problem, the discriminator failure was.

 

I still run the stock discriminator valve inside the tank plumbed to the original set up under the hood (minus charcoal canister) with a coil of hose, fortunately it’s worked well in our application.

I now have a tall and narrow catch can on the end of the vent line to the stock in-tank discriminatory valve.

[Grant]

9 hours ago, Racer28173 said:

@Grant  After seeing what @Robmink posted above, I’d be interested in knowing what the specs were for the discriminator valve you were using.  Those pressures he listed are minuscule, in my opinion 

It was a 12 AN fuel safe unit. It's a light, floating ball on top of a heavy, sinking ball, with a tapered seal at the top. The difference between it and the OE valves, aside from size, appears to be the OEM's use of springs to hold the float away from the seal when they aren't floating.

[frankrehnelt]

The same thing can happen to an e30. It happen to us at Daytona a few years back. The drive shaft almost rubbed a hole in the tank. Our vent was not stock and honestly it was a mess. If I’m not mistaken Docs car had its expansion tank (the plastic jug in the right rear wheel well) break open and caused a fire at NCM. It can also happen using a fuel cell if the discriminator valve “float” ball expands and jams shut as Grant has pointed out. The first symptom this has happened is the fill rate slows way down. 

Edited October 26, 2022 by frankrehnelt

[collinskl1]

4 hours ago, Grant said:

It was a 12 AN fuel safe unit. It's a light, floating ball on top of a heavy, sinking ball, with a tapered seal at the top. The difference between it and the OE valves, aside from size, appears to be the OEM's use of springs to hold the float away from the seal when they aren't floating.

 

This one?

 

https://www.pegasusautoracing.com/productdetails.asp?RecID=7265

[Grant]

5 hours ago, collinskl1 said:

 

This one?

 

https://www.pegasusautoracing.com/productdetails.asp?RecID=7265

Yes

[veris]

I'm running one of these: https://www.summitracing.com/parts/abt-re185

 

It is clear so you can see if it is stuck. 

 

 

 

 

[drgrumpus]

On 10/26/2022 at 3:20 AM, Grant said:

It was a 12 AN fuel safe unit. It's a light, floating ball on top of a heavy, sinking ball, with a tapered seal at the top. The difference between it and the OE valves, aside from size, appears to be the OEM's use of springs to hold the float away from the seal when they aren't floating.

Interesting, we had some trouble with the Fuel Safe 12AN DV. I disassembled it for inspection and found the lighter floating ball on the bottom and the heavier smaller ball on top.  FuelSafe confirmed this arrangement but you found yours in the opposite orientation.  Which is correct? 

[collinskl1]

4 hours ago, drgrumpus said:

Interesting, we had some trouble with the Fuel Safe 12AN DV. I disassembled it for inspection and found the lighter floating ball on the bottom and the heavier smaller ball on top.  FuelSafe confirmed this arrangement but you found yours in the opposite orientation.  Which is correct? 

Mine is also floating on the bottom and heavy on the top. I've found pictures online of different units with similar function which are also floating on bottom.