fuel injector seal 2

[brashear]

One of the fuel injectors in my car developed a leak on the inside valve. This got me thinking - why is the needle seal metal-to-metal? With the different types of fuel resistant polymers and elastomers out there today (ex. PTFE, PFA, Delrin, Viton rubber) , wouldn't it be more reliable and less costly to make the needle seat on something softer?

 

[dgallup]

Assuming you are talking about port injection, most injectors produce in the last decade+ use stainless steel balls to make the seal. These are available with incredible precision for a couple of pennies. There are some natural gas injectors (Bosch NGI2 for one) with a molded on elastomer to achieve the lowest possible leak rate. These molded components cost an order of magnitude more than the easily mass produced balls and do not last as long.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[patprimmer]

Also I think EFI type injectors cycle rapidly and I suspect the elastic recovery rate would be a real issue.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[brashear]

In looking at some designs, the ball is attached to a stem, which is a lot more expensive than just having a precision bearing type ball. I really don't see how having a polymer seat would be more expensive than a metal one. You should be able to machine a polymer seat using the same process as a metal seat, except with less stringent tolerance and surface finish requirements on the ball.

The only reason I can think of is that a fuel injector will see something like 1E9 cycles over a decade and using a polymer or elastomer seal there would be issues with creep and compression set. This would probably affect the flow rate of the injector because I'm guessing the ball valve lift is only a few thous.

 

[dgallup]

You can buy the ball as a commodity item and resistance weld (or other process) it to the rest of the armature structure. This is simple, robust and efficient. The elastomer has to be insert compression molded and then cured. This is much slower and more expensive. You are correct that the expected life of the injectors is on the order of a billion cycles and the lift is on the order of 100 microns. This can be achieved in either case but the elastomer is much more expensive than the metal ball.


----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[brashear]

That makes sense - thanks for the explanation. What I don't understand then is why it costs me $120 to replace a fuel injector when I can't spot anything else in the design that would make it expensive. I assumed that the ball and seat were the most costly components but if it can be made for less than a dollar, where is the rest of the cost going? The coil should be able to be wounded for very cheap and most other parts are molded plastic.

 

[bradrs]

The reason it costs $120 might be because you are buying it from the wrong source

Some injectors are VERY expensive, even from the OEM. The already mentioned Bosch NG injectors are a good example. But even those are relatively cheap if you buy thousands of them. For most parts though, it is all about the markup between the injector manufacturer, and the OEM, their dealer, etc. It is more on the order of $10 from the manufacturer, in high enough volumes.

 

[dgallup]

As a rule of thumb you are paying 10 times at the parts counter what the OEM paid for any part.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[ivymike]

That multiplier varies a fair bit depending on the size of the engine and availability of "will fit" parts, of course.

I've seen some OEM injectors sold at retail for only about 1.5X the OEM cost...but buying just one of them would cost as much as replacing 25 of yours.

 

[brashear]

That explains it. How often do fuel injectors leak? It seems like it would be difficult to achieve an absolute leak-free seal with metal-to-metal unless there was some sort of significant deformation like in pipe threads and compression fittings. Is there some sort of mechanism that prevents fuel from dripping out of the injector when the engine is off such as a check valve on the pump?

 

[dgallup]

Everything leaks. It's just a matter of degree. Just like there has never been a physical round hole or a flat plane. These are mathematical concepts that are not physically realizable. If you have sensitive enough instruments you can measure the leakage of any valve.

There is no shut off valve, the rail is supposed to maintain pressure when the engine shuts down. As a rule, leaky fuel injectors are quite rare these days, the primary failure mode is contamination. A soft seal may or may not be less susceptible to failure from contamination. Hard contamination may damage a soft seal but eventually work it's way out of a hard seal.

From my experience at least 75% of the automotive fuel injectors that are replaced have no failure at all. Some of these may have been replaced because one was bad and the labor charge is basically the same to replace them all. Others are replaced because the mechanic (parts changer) has no idea what is wrong and is just changing one thing after another. A lot are replaced because someone is trying to make a boat payment.

I'm curious, what was the problem that was supposedly caused by your leaky injector and how did they diagnose it?

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[brashear]

Well the leakage rate must be small enough where the engine can sit there for a year or longer without the engine being flooded with fuel. My car was running a bit rough at low speeds and it was time for the routine checkup anyway. I don't know how they diagnose it was a leaky injector but that's what they told me. Maybe there was just a vacuum leak or something.

 

[patprimmer]

The fuel pump turns off when the ignition is turned off.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[brashear]

If there's no check valve or a positive shut off valve and the tank is higher than the injector, fuel

 

[brashear]

fuel will leak into the engine

 

[dgallup]

If the car sits for a year a leaky injector will be the least of it's problems.

I know of no cars with the tank above the engine.

The amount of leak we are talking about here is miniscule. Typically for a port injected engine the spec is less than 1 sccm nitrogen with 4 bar differential pressure. Obviously it will be a lot less on gasoline. So much less it gets really hard to measure which is why all the testing is done on nitrogen.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[Tmoose]

gunk at the tip can mess up spray patterns.

http://www.allstates.com/fuel injector asnu testing injectors.jpg

I have some Bosch LH FI info that says some modest amount of injector leakage is still OK. Maybe a few drops per minute.

If left long enough the residual pressure would drop, and the pump would have to run a few seconds to fill and pressurize the system again, resulting in slightly delayed starting. The Fuel pressure regulator and residual pressure valve near the pump can leak, and create similar symptom.

 

[patprimmer]

The rings do not form as good a seal as the injectors.

Before 1920 most cars had the tank higher than the engine, however I don't recall any being injected.

If you think about the volume of fuel under pressure, the potential size of the leaks and the volume of the cylinder and the potential leaks from the cylinder it really becomes very obvious that this is a non issue.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules