20v Retainer development...

[fast_a2_20v]

Hey guys, I've been working on these 20v drop in retainers for a little while now, and have gotten Don R. and Bob Q. a little bit involved in their development as well.

They are going to be cut from 7075 aluminum. This makes these a full 30% lighter then a set of Titanium retainers.

The trick has been to add some material to ensure strength, but without adding material in places where it doesn't do any good.

Here is an OEM retainer for comparisons sake:




This is where I started, material is added to the top of the retainer to ensure strength against the spring pressure. Material is also added around the keepers to make sure the keepers do not pull through the retainer. Note how the top of the retainer is concave...





At this point we decided to do some stress analysis... You can see the stress concentrations in this short movie, blue being 0 and red being the highest. All are well below half of the materials yield point. (Min safety factor 2.08)

http://www.integrated-engineering.org/gallery2/main.php?g2_view=core.DownloadItem&g2_itemId=65&g2_GALLERYSID=d17af6f966fb6fcaf91ebf53d9af6726

As you can see in the movie, the extra material along the top of the retainer is clearly not necessary. A bit more testing and I ended up close to stock on the "TOP" of the retainer, and it is no longer concave at all:





In this case, the FEA stress concentration analysis allowed us to shave off a full 10% of mass with absolutely no effect on the minimum safety factor achieved, in this case, 2.08.

Final analysis:

http://www.integrated-engineering.org/gallery2/main.php?g2_view=core.DownloadItem&g2_itemId=68&g2_GALLERYSID=d17af6f966fb6fcaf91ebf53d9af6726

The files have now been sent to the machinist to have a prototype set cut and hard anodized. Will update when that is complete. 

[Don®]

This is what I like to see...innovation  Thanks for your efforts Pete!

[Wolk's_Wagon]

Looks good so far.

I'm glad you are going to hard anodize them, other wise I would not suggest Alum. for that application. Although the yield strengths of the higher grade Alum. look to be as strong as some 300 series SS. My experience is it gets beat up too easily, dings, dents and abrasion wear take their toll. I have seen some spring seats made out of Alum. that just get chewed up by the flattened cut and ground end of the spring. Hard anodize will help but it is realy only .0006" deep surface treatment, the matl. is still soft underneath, and it can ding and scrape off easier than you would think.

[Jeronkie]

Guys like you make good cars better cars, keep up the good work

And if it works and I am sure you will make it, I want some   
 

[fast_a2_20v]

Looks good so far.

I'm glad you are going to hard anodize them, other wise I would not suggest Alum. for that application. Although the yield strengths of the higher grade Alum. look to be as strong as some 300 series SS. My experience is it gets beat up too easily, dings, dents and abrasion wear take their toll. I have seen some spring seats made out of Alum. that just get chewed up by the flattened cut and ground end of the spring. Hard anodize will help but it is realy only .0006" deep surface treatment, the matl. is still soft underneath, and it can ding and scrape off easier than you would think.

This is where the finishing of the valve spring is very important. I was recently considering having my own valve springs made for 20v as well, and did quite a bit of talking to some valve spring companies. After they are wound, there is a lot of options you have with how fancy you want to get on finishing the spring.

Most of the 1.8t springs are already quite nicely finished because they have to sit directly on the cast aluminum head, in the machined pocket, and odviously they can't chew up the head as its not replaceable haha.

That said, the 90 degree joint between the spring seat and the center of the retainer is a stress riser. Its quite critical to fit a radius in this region, but we are limited by the fact that the springs are ground... If the radius is too large, the sharp corner of the ground off spring will dig into the aluminum and put back in the stress riser we are looking to eliminate in the first place. Not much we can do here except compromise, and just go with the biggest radius that will fit.

These are a fairly aggressive race part. On a motor being revved way out frequently, in street use, I would probably recommend pulling them out, along with the springs, and checking / replacing the springs as neccessary and inspecting the retainers.

[bobqzzi]

Thanks for sharing the development.  Properly finished aluminum is okay a retainer material.  Seems to work better on smaller stuff than large V8 stuff.

[enginerd]

Don't forget that these are going to fail in fatigue, not from yielding.

Steel can only sustain 80,000 psi tensile in fatigue application while the yield strength can be +340,000 psi. (only 23% of the yield strength)

Aluminum does not have an endurance limit and will always fail from fatigue after xxxxxx cycles. You can design around living for xxxx million cycles, however an engine gets to millions of cycles very quickly.

I did a quickie search on "aluminum endurance limit" to show an example. It is a web site about bicycle frames, but gives the basics of what I am talking about. I apologize if you already know all of this, my intent is not to be condescending, but to share. Your comment regarding 50% yield makes me believe that you are not considering this.

http://bobbrowncycles.netfirms.com/eng.htm#section1


Bottom of the page shows a stress vs number of cycles diagram. - Note that the aluminum never reaches infinity for cycles like the steel does. He has 26 ksi for some reason we use 80ksi for steel in our calcs. But the example is still sound, just not the exact numbers.



You should pick a stress level that will last for a good amount of time. Maybe 200 hours on an engine @ 7000 RPM = 200*7000*60 = 84 million / 2 (head is 1/2 erpm) = 42million cycles. Now I know people have different drive cycles, etc etc and you might not spend all day at 7000, but If I were testing this I would aim high.

For the aluminum on this particular diagram (yours will be different)
42 million cycles = (42 * 10^6 )  occurs ~ 25 KSI it is a log scale btw just not shown.  So this would be your design strength for a 200 hour race engine. And that has no margin for safety factor, and does not really consider the full range stress.

Safety factors for fatigue have a range stress and a peak stress component contribution.

So basically be careful with aluminum in high cycle fatigue applications. If you change them out every season, or only put 2 hours on a year it's another animal.

[fast_a2_20v]

 

Great advice and tech info... I have the yield point set to 17ksi in solidworks, wheras the 7075 t651 we are using is actually 73 ksi yp, gave us the mentioned safety factors of ~2 on the intake retainer and ~1.5 on the exhaust retainer (calculated using the low yp)...  The fatigue strength at 500M cycles for this material is 23ksi (qouted from matweb), but I haven't gotten solidworks to accept any changes to material strength yet, so 17 will have to do. 

I certainly wouldn't recommend anyone drive a million miles on them, but from what I have been told by the valve spring guys, I wouldn't go a million miles on most aftermarket valve springs either. 


Its a good thing too, because there really isn't a way to add any more material where it is most needed, at least not while still using an oem "size / shape" valve spring. Its just a really nasty stress riser created by the spring seat / step.

[fast_a2_20v]

Update: Machinist decided to do my prototype set for free. That seems to translate to.... SOOOOOOOOOOOOOOO SLOWWW!!

[transient]

if you're bored, would you want to make the crank oil pump gear for an MKiv?

[fast_a2_20v]

Got the prototype set some time ago. Putting togheter the rest of the motor. 

[Dizzy]

Slacker.