Double valve springs
 

So I've just had a built with no expense spared engine fail on me due to a dropped inlet valve and I have absolutely no idea why.

The car was running over 600WHP and was just 10,000kms old which were mostly low speed road miles.

Car was a sleeved EJ257 with Cosworth rods and pistons, cosworth cams 278 10.7mm lift on the inlets and 274 10mm lift on the exhaust.

Inlet valves were all brand new when built.
Supertech Niterided Inlet valves
Supertech Inconel Exhaust valves
Supertech Double valve springs (82LB seat pressure)
Supertech titanium retainers.
Rev limit was set to 7800 and although I've hit the limiter a few times it has never been over revved to a missed gear etc.

I even took the precaution of running a slightly thicker head gasket to give me a bit more piston to valve clearance. 1.1mm instead of stock 0.6mm ones.

So I'm stumped as to why this has happened because it really shouldn't have.

One school of thought is that the springs im using are quite high pressure and maybe not suitable for use at extended mileage in a road car. Thus causing the valve to fatigue.

From the picture below you can see the valve isn't bent it looks like it just broke off.

So I'd like to hear from anyone running double valve springs.
What make of valves and springs, what spring rate and importantly what mileage.

Would also like to hear from anyone who has suffered a similar failure.

This is the second supertech valve fail on me but the first time (6 years ago) I put down to some faulty machining I had done. This may not be the case now.

Looking forward to your replies.

http://www.scoob.co.uk/blackhawk/failureapril16/4.jpg

I've been looking into building a engine and have read AS MUCH info as I can.

No I've read a few times of valve failures due to the valves touching each other whilst timing up! Inlet and exhaust obviously.

It obviously only takes a small fracture or imperfection.

Have you fully stripped it yet Daz ?


If not the clues will be there for the failure, by your dyno runs you put up on 22B directly before failure, to me it never lost a valve head instantly, it would of lost a lot more power & never recovered to make a 2nd run, as you said it felt like a slight misfire, is the spark plug damaged by the failure, if not is the electrode intact ?


If its piston contact it will show in other pistons & valves, anything missing in the cold side of the turbo ?.


The Supertech double spring ratings are dependent on the bottom platform, I presume you have the thicker of the two fitted, with certain cams the springs do come very close to coil binding on the thicker platform.


I have had a couple of Supertech failures, but not as a result of the valve itself, & these were exhaust, not inlet, one was an over rev, 11,000rpm, the other was a bucket failure & the valve stuck open, & the piston hit it.


As I mentioned I have had instances of damage to the valves in manufacture & bad packaging, from valves hitting valves during delivery etc, a good ding on the taper of the valve head will certainly throw up a stress riser, & will eventually break from it.

Very sorry to hear about this daz hope u get to the bottom of it .

Hi Paul,

I'm always grateful for your input so thanks for taking the time.

The spark plug was hammered by the valve see pic below.
Yes I have the thicker of the seats fitted. It's never once had a high RPM buzz due to a missed gear or anything like that.
Limiter was set to a conservative 7800rpm. It had slightly thicker head gaskets and the whole valve train was brand new including all the buckets so why this has happened is a complete mystery to me.

Both hotside and cold side of turbo are fine with no sign of debris going through it. The failure itself actually happened at idle which I am very thankful for and no doubt saved me from much worse damage.

My school of thought now is to go up to a 1.5 gasket and change all the inlet valves to ferrea ones.

I'll get the plasticine out this time and properly measure the PTV clearance.

But if it's not caused from hitting the piston ( which it shouldnt be) then I have no clue why it failed.

http://www.scoob.co.uk/blackhawk/failureapril16/1.jpg

Valve springs too strong? If they are closing too quickly, could they have snapped the head off like pulling a dandelion flower through your fingers? All it would take would be one flaw in the valve stem.....

I sent Supertech an email. About that and here is As expected it's a bit of a non reply.

I guess the only positive part is that they confirm spring pressure is not a factor.

I think my plan is to replace the inlet valves with Ferrea ones. I'm going to stick with the exhaust valves as they are inconel and the problem only seems to be inlet valves. I'm also thinking of going to up to a 1.5mm gaskets. But I'm going to get the plasticine out and properly check ptv clearance before I decide to go that route.

Anyway here's the reply from one of Supertech's engineers.

Hi Darren,

It is almost impossible to know what happened from just a photo. *The clean cut that you describe is typically a fatigue failure. This happens when the valve is flexing when closing against the seat. This typically happens when the guide is not aligned with the seat or when the valve is bent.

This type of work by the valve, by fatigue, generates a small crack that will advance until it completely breaks producing what looks like a clean cut.

*

In your case, for some reason, that valve may have been slightly bent at some point and lasted just some time until it fatigued and cracked.

*

I can assure you that the stiffness of the spring has nothing to do with this failure. The stiffest the spring, the better for the valve.* The valve can withstand a lot of spring pressure.* It is much worse to have a soft spring that may float the valve because when the valve floats, the dynamic impact of the valve against the seat will stress the valve 5-10 times more than the stiffest spring you can put.

*

We are not experienced Subaru tuners so we are unable to explain why, but your problem may be related to tuning issues. *Some years ago there have been some discussions in some Subaru forums regarding this issue, happening with valves, involving different manufacturers, so it was not related to the valve specifically. I also remember that some of these issues involved some Cosworth cams (maybe some others but I don’t remember), and the other thing was that it was almost always on Cylinder # 2, which brings a lot of coincidences with your failure, and may indicate a tuning issue at some point.

I believe that there are specific procedures from Subaru and I believe from Cosworth too, to set up the cams.

*

We are no trying to take away any responsibility here. We care and work hard to have the best products in the world so if this were a Subaru valve issue we would have fixed it long time ago because we sell these valves every single day around the world, much more than anybody else, for a reason.

*

Fell free to contact me with any questions.

*

Best regards.

Willy Tagliavini
Supertech Performance Inc.
3550 Charter Park Dr., San Jose, CA 95136
Ph: 408-448-2001
Fax:408-448-3700
willy@supertechperformance.com
www.supertechperformance.com

Do you know how much additional lift the cosworth cams provide and can you rule out coil bind as a start?


Really sorry to hear about the failure. poop.

Not massive lift. I haven't had to machine the heads to accommodate them put it that way .

Inlets is 10.7mm lift exhaust is 10mm

Daz, if you go to thicker gaskets, won't that affect the compression ratio, leading to possible det?

I had a track car once running throttle bodies, lumpy cams (9K rev limit) and double valve springs. After 2000 miles, it dropped a valve, which completely destroyed the head! Although it was impossible to prove, the most likely explanation was that due to the tight fit, contact between the inner and outer spring over time caused a spring failure. I was also told by the suppliers (SBD Racing) that I shouldn't have expected to run them for more than one season or 1,000 miles before doing a strip down - something they singularly failed to point out at the time of purchase :Suspiciou More genrally, I'm still curious as to why double springs are recommended; in the Honda tuning community, there are plenty of great durable engines using high lift cams (14mm+) and revving to well over 9K for which single springs seem to work perfectly well

The cam controls how quickly the valve seats, nothing to do with the springs, if the springs were to strong it will result in to higher cam nose pressure, resulting in either premature cam / bucket wear.

Daz, you should not need to go with a thicker gasket, I have never had any interference issues with Subaru engines, even with a 0.78 gasket, & 12.5mm of lift, you need to dummy build it with very light springs that you can push down with finger pressure, you can then measure your clearance usually at 13 to 15 degrees before & after tdc, usually its the inlet valve chasing down the piston after tdc that gets the nearest to touching.

Well, yes and no. A weak valve spring may not be able to seat the valve even when the cam allows it to at high revs, valve float. Conversely, a super strong one will shut it as soon as the cam allows, but maybe snapping it shut that fast causes stresses and strains?

I think Paul already explained that....

Where?

Post 12

Nah.

Think about it.....the stronger the springs, the faster the valve will close..or what's the point of stronger springs?

Oh so you're the expert now?

What would Paul finch know about it anyway? :confused:

Now now girls.

I think you need to think about it !


What controls the springs, the cam, the valve cannot move without contact with the cam, the reason you select a stronger spring to go with a faster opening & closing ramp on a cam, with increased rpm range to keep the valve under control, & maintain a acceptable seat pressure with increased boost pressure acting upon it.

:thumb:

And how, pray, does it keep the valve under control? By maintaining contact, so that the valve shuts as fast as it can.

Stronger springs, shut quicker, end of.

You really don't get it do you, how does the valve shut quicker than the cam lets it ?

What shuts the valve? Not the cam, the spring. The cam allows it to shut, controls when it can shut, but it's spring pressure that pulls it shut. No?

If no spring, it would hang open.

if a spring, it's pulled shut.

if a STRONG spring, pulled shut quicker.

Simple really.

How can it shut quicker if its against a cam, if no cam it will shut quicker, but it cannot, as its against the cam, the only time its free to move under no cam control is the last few thou of the valve clearance.

If the cam doesn't allow it to shut quicker how on earth can it?

I think you maybe confused with valve bounce and valve float.

Dude just shhhhh.

Paul knows more about engines than you and I ever will.

Anyway got the head off today and this is what I was presented with.

http://www.scoob.co.uk/blackhawk/failureapril16/5.jpg
http://www.scoob.co.uk/blackhawk/failureapril16/6.jpg

**** :( , that's a right mess !