EAE Update-Carbon Fiber Plenum

[tbuck]

My life must be pretty damn dull that I have come back here for some entertainment, lol!

The only thing I would interject in this ridiculous thread, you may have the best **insert description here** ever made, but if you have shitty service you will never get any of my money. Don't care if it costs a single $ or thousands, same thing applies.

Good day.

[ygoz]

Originally Posted by Chrisyphus

Come on. Calling him a customer is not real. He has been combative since his first comment.

No, he has not. He asked legitimate questions, in response to empty/BS marketing statements.

Challenging why your product is good, with the background it has is reasonable, sadly for this project, its not starting form 0 like a new company, its starting from -100, because of the previously bad reputation - that is something that you should have taken into account when taking this project on.

And again, while Abdul claims to have many successful products many disagree.

The most successful one, the actuators are not Abduls at all, and it is weird he tries to take credit for it as if its his R&D or Product.

Actuators are loved and needed because:

* They are much cheaper than OEM - (due to rebuild's pricing)
* They improve on the original design and last - (once again, rebuild's R&D and work - nothing to do with Abdul)

Abdul made sure availability is good in the US - and so far all of the issues/complaints were from this domain. no one complained they are pricey or do not work
Acknowledge that area has improved - but the reputation lingers.

No one said Abdul is shady.

I said the Dyno charts were shady, because they had marketing BS about gains (35hp - common, don't give the raw number, that shit is misleading).

And the number fudging with stock dyno runs ending before the redline - To which Abdul conveniently did not even comment.

On top of that - the original marketing of EAE was - "we designed this for bling, and look at that - by mere coincidence its the best performing plenum on the market" - Of course it is

[e92dud]

Originally Posted by Chrisyphus

That’s a really good question & I can see why they chose to go that route.

If we had bonded the Aluminum to the carbon, it would have been a minor concern across 100*c you can expect the material to expand ~2.3 mm

Our prepreg, like most, on the other hand is formulated for a near perfect zero CTE.

Lengthwise on our flanges we could expect to see about a third of that across 100*c this is sort of a worst case scenario, obviously, but it is concerning. For this reason we decided that instead of bonding the aluminum we use a gasket and fasteners to secure the flanges to the carbon.

Thanks for the response, seems like they both have their advantages.

[M3SQRD]

Originally Posted by Chrisyphus

If we had bonded the Aluminum to the carbon, it would have been a minor concern across 100*c you can expect the material to expand ~2.3 mm

U = alpha* DelT*Uo

Uo = 36” = 914.4 mm, for assumed plenum width (should be distance between F-R sided trumpets or the long length of the lexan window)

U = 2.3 mm (quoted thermal deformation)

DelT = T - Tref = 100 C, T =120 C. Tref = 20 C (room temperature)

So CTE is alpha = U / (DelT*Uo)
= 2.3 / (120*914.4) = 251.6 ppm/C
Note: CTE would be even higher if a correct, shorter, length was used in the calculation

Which material has a CTE of 251.6 ppm/C? CF CTE is (likely negative and) on the order of 0.1 ppm/C. You’d have a potential CTE mismatch problem between this material, the structural adhesive, and the CF, if the CTE or deformation at 120 C are correct.

[JDiaz]

This thread is why I stay the F... away from forums. Everyone has the right to question a product and ask for details. But stop being blasphemous about fellow members. It doesn't take a genius to see that Abdul is a very standup person and has a great reputation across multiple BMW subgroups. I have known him from the M5baord since my days of owning an E39 M5. Till this day, I don't know of anyone who has dealt with him that had anything negative to say other than the fact that sometimes takes a few days to get back to you. From what I understand, this was not his fulltime thing when he first started so that's understandable.

Yes, he is not the one who invented Rebuild Ta's. But he was damn well the person who put his money where his mouth was and provided that service for us here in North America. I have done 2-3 transactions with him over the years. I Bought TAs from him back in the day when I still had my M3 and had an amazing experience. Also bought G30/F90 taillights from him and I think one other thing. Can't remember off the top of my head.

Abdul I'm happy you're still in the business and are on to new projects. I wish you all the luck.

[Chrisyphus]

Originally Posted by M3SQRD

U = alpha* DelT*Uo

Uo = 36” = 914.4 mm, for assumed plenum width (should be distance between F-R sided trumpets or the long length of the lexan window)

U = 2.3 mm (quoted thermal deformation)

DelT = T - Tref = 100 C, T =120 C. Tref = 20 C (room temperature)

So CTE is alpha = U / (DelT*Uo)
= 2.3 / (120*914.4) = 251.6 ppm/C
Note: CTE would be even higher if a correct, shorter, length was used in the calculation

Which material has a CTE of 251.6 ppm/C? CF CTE is (likely negative and) on the order of 0.1 ppm/C. You’d have a potential CTE mismatch problem between this material, the structural adhesive, and the CF, if the CTE or deformation at 120 C are correct.

lol who’s gonna tell him?

[M3SQRD]

Originally Posted by Chrisyphus

lol who’s gonna tell him?

I don’t know…Who’s the one that would have trouble dealing with a window in the side of a wall?

[Chrisyphus]

Originally Posted by Chrisyphus

lol who’s gonna tell him?

Okay I will.

Originally Posted by M3SQRD

U = alpha* DelT*Uo

Uo = 36” = 914.4 mm, for assumed plenum width (should be distance between F-R sided trumpets or the long length of the lexan window)

Okay, fine let’s pull random numbers out of our ass instead of nice round figures

U = 2.3 mm (quoted thermal deformation)
So far so good, although you didn’t mention that’s across 100c… can’t imagine that’ll be a problem later.
DelT = T - Tref = 100 C, T =120 C. Tref = 20 C (room temperature)

I already defined delta T when I answered their question as 100, but okay. Technically it’s Tref1 -Tref2 =deltaT, but okay I’m sure I’m being pedantic here.

So CTE is alpha = U / (DelT*Uo)
= 2.3 / (120*914.4) = 251.6 ppm/C
Oh you goofed. deltaT, as I defined is 100, but you had to get sexy with it.
Note: CTE would be even higher if a correct, shorter, length was used in the calculation

Which material has a CTE of 251.6 ppm/C? CF CTE is (likely negative and) on the order of 0.1 ppm/C. You’d have a potential CTE mismatch problem between this material, the structural adhesive, and the CF, if the CTE or deformation at 120 C are correct.

you really typed all of this out and thought “yeah! This part is made from carbon, aluminum, and lexan and it seems probable that one of those would expand a full two and a half centimeters per hundred degrees?”


Further, your math as written is wrong but the solution your calculator gave you is… also wrong. You lost a decimal point somewhere. It’s almost like you forgot I said 2.3mm over 100c shit.


To correct your math the formula would be:
2.3=a*914*100

2.3mm/91,400= 25.1ųm/c or as you put it, 25.1ppm



But wait there’s more, because that’s not accurate. Why? Because you redefined the length without a commensurate redefinition of the expansion across that length at 100deltaT

Let’s do that so we can determine which material I was talking about:
91.4% of 2.3 is 2.102 soooooooo

2.102/914*100= 23ųm per degree Celsius

https://www.gabrian.com/6061-aluminum-properties/




Man I am too slow with forum formatting lol

[reidetailing]

Originally Posted by Abdul

Thanks a lot! The plenum does look and sound amazing.

I received your preorder for the valve covers. I also can’t see a reason why it won’t be. But let me check with the engineering team and get back with you.

Yes, please keep me posted on those fuel rails, Abdul.

[herrubermensch]

Unfortunate devolution of this thread, gentlemen.

For the record: Partee Racing so far has generated two carbon fiber products: carbon fiber velocity stacks for the E39 M5 and the plenum that is the subject of this thread.

The velocity stacks have been a huge success and even more effective than we had expected. They gained 22WHP on one of our NA HO S62 builds on back to back dyno runs, same car, same engine, same dyno, same day, 15 minutes apart. Why? Because we went to the trouble of making a 180 degree bellmouth. Remember this for later because it informs the design of our E9X plenum.

The carbon in our E9X plenum is made as a single piece. It is our opinion that this design is stronger that two-piece plenums that have been bonded together, for many reasons, but one of them is the much larger likelihood of assembling error with gluing two pieces together. Too much glue, squeezing it together to tightly, wrong cure temp, oily hands on the area where the glue should adhere--all provide a high possibility of a mistake that is just not present with single-piece carbon fiber. That is our design choice, we stand by it, if you disagree, that's your right, but we think you are wrong. And the response we have received to our design from multiple carbon fiber fabricators whose opinion we respect corroborates our conclusion.

The stack design: The S65 is intended to be a much higher revving motor than the S62, breathing well in the 7k to 8.6K rpm range. For that application, shorter stacks make sense and typically perform very well. But the length and twist angle of the stacks is immaterial next to the shape of the bellmouth. From the experience of others to our own testing--which it took to convince Chrisyphyus (lol) that I was right--a flat opening to the stack is disastrous for air flow, whereas 180 degree bellmouth is pure magic. The E9X plenum stacks have a beautiful contoured 180degree plus bellmouth. That combined with the increase in area just above them as a result of our removal of the side "ribs" in the plenum WILL increase performance over stock AND over competitors that do not run the requisite bellmouth. No, we haven't compared them side by side, in part because those competitors are friends and colleagues with whom we do business in other areas. So instead of throwing shade on a particular plenum, we are just touting what we reasonably believe to be the superior design of our own plenum.

As for the Lexan window: I don't like it either candidly, but LOTS of customers wanted it, both here and in our S62 plenum that we just launched tonight. I refused to do a Lexan window with the S62 because it would just offend my sensibilities, and Abdul kindly deferred to me there, so I deferred to the wishes of Abdul and many customers here. That said, we ARE making a one-piece version WITHOUT the Lexan for those who want it.

For the record, we had nothing to do with the prior EAE plenum which was admittedly a POS. As we would have expected of Abdul, he refunded all of those purchases, and we are now offering our plenum to those customers at a discount. Draw negative conclusions about the quality of our product from that unfortunate scenario (much more for Abdul than anyone) if you want, but they are unwarranted.

Finally, and maybe this is just me showing my age (I'm 58) and the fact that I've practiced law for 32 years, but I would greatly appreciate it if we can tone down the rhetoric (on both sides) and just have a friendly informed debate or expression of differing opinions. Intelligent mature people should be able to do that. Thanks.

--Peter

[M3SQRD]

I’ll own my mistake. I mis-carried my decimal place.

Series design question. Depending on which version of Lexan you’re using, it could have a CTE as high as ~40 ppm/C, with the adhesive you’re using, are you concerned about the combined thermal deformation/cycling + vibroacoustic response of the top CF-Lexan cover introducing a peel loading in the adhesive? This would be a concern if the Lexan cover has much higher bending and axial, and to a lesser degree OOP and IP shear, stiffness relative to the CF.

[Chrisyphus]

Originally Posted by M3SQRD

I’ll own my mistake. I mis-carried my decimal place.

Series design question. Depending on which version of Lexan you’re using, it could have a CTE as high as ~40 ppm/C, with the adhesive you’re using, are you concerned about the combined thermal deformation/cycling + vibroacoustic response of the top CF-Lexan cover introducing a peel loading in the adhesive? This would be a concern if the Lexan cover has a much higher bending, and to a lesser degree shear, stiffness relative to the CF.

The short answer is no, it should not be an issue.


The longer answer is that the CTE is 37ųm which is a lot. We took this into consideration and worked with it.

Lexan section is just under 15” in length and 10 in width. If it were 2 or 3 times this size, it would be a much larger factor to consider, but at this size and with the temperature fluctuations it should see (not far outside of ambient) it’s very manageable.

[M3SQRD]

Originally Posted by Chrisyphus

The short answer is no, it should not be an issue.


The longer answer is that the CTE is 37ųm which is a lot. We took this into consideration and worked with it.

Lexan section is just under 15” in length and 10 in width. If it were 2 or 3 times this size, it would be a much larger factor to consider, but at this size and with the temperature fluctuations it should see (not far outside of ambient) it’s very manageable.

I’m not suggesting your design has an issue(s).

What’s the CTE, shear strength, modulus and peel strength of the methyl methacrylate at RT and elevated temp? There are too many versions of methyl methacrylate adhesive to randomly pick one plus very few mechanical properties are listed (at least in the material databases I have access to).

Engine bay temps not far from ambient temp? My engine bay, and plenum, are quite hot after a 30 min track session. Have you made predictions of the relative OOP deformation and rotations of the Lexan-CF at the free edge of the Lexan? They’ll be an OOP loading, in addition to the IP loading from thermal deformation, on the adhesive and, generally speaking, peel strength is low for structural adhesives.

[Chrisyphus]

Originally Posted by M3SQRD

I’m not suggesting your design has an issue(s).

Yeah no I get that, we are getting into more productive discussions now, which I appreciate.

What’s the CTE, shear strength, modulus and peel strength of the methyl methacrylate at RT and elevated temp? There are too many versions of methyl methacrylate adhesive to randomly pick one plus very few mechanical properties are listed (at least in the material databases I have access to).

The CTE of the MMA in this application is all but entirely negligible at 10-14 um per *C per meter on the scale of a bead that is only several mm across, we are nearing expansion in the nanometer range

The shear strength of the MMA we used is a bit over 23MPa (~3500PSI) and the standard temperature range (in which there are no major deifferences in performance) is -55c to 120c. per the TDS supplied, when bonding lexan to carbon it is expected that the substrate will fail before the MMA will give up in shear. We would not randomly pick anything. You are absolutely right that structural adhesives are typically far weaker in peel strength, that is true of all adhesives to my knowledge.

Engine bay temps not far from ambient temp? My engine bay, and plenum, are quite hot after a 30 min track session.

I did not say engine bay temps. I said temperature fluctuations- in reference to the temperature fluctuations of the material itself. The engine bay obviously gets quite warm, but the inside of the plenum should be taking near ambient temperature air from the intake which will cool the Lexan to an extent, but even if it sees a 200+*f temperature fluctuation, we're looking at a total expansion of ~.6mm along the length of the lexan which will result in a sub 1mm OOA deformation and a tensile load on the adhesive well within 23MPa in any direction and a very negligible lifting force.

Have you made predictions of the relative OOP deformation and rotations of the Lexan-CF at the free edge of the Lexan? They’ll be an OOP loading, in addition to the IP loading from thermal deformation, on the adhesive and, generally speaking, peel strength is low for structural adhesives.
see above

Thanks, Good questions!

[M3SQRD]

Chrisyphus

Engine vs. plenum temperature - good point about it not being hot while on the track when it’s subjected to vibroacoustic response/loading. Heat soak occurs when the plenum is subjected to conductive and radiative heat transfer mechanisms. Max IATs of the e9x M3 is the thermal environment that would occur at the same time as the mechanical loading environment.

Peel strength - true, all adhesives have poor peel strength so poor word choice However, I work with thermal engineers that want to use thermally-isolating adhesives that have abysmal peel strength!

Thickness ratio of Lexan vs. CF fabric/layup - how thick is the Lexan? Have you estimated the OOP displacement and IP rotations of the Lexan at the bonded I/Fs with the CF? The relative bending stiffness of the Lexan and CF will produce a curvature/slope discontinuity and, thus, generate a peak peel loading in the adhesive along the free edge of the Lexan - higher Laxan-to-CF bending stiffness ratio = higher peel load. As you know, workmanship issues during the prepping of bonding surfaces are critical for developing peel strength. One time we had shear clips between composite square tube sections at double lap shear joints that had used standard surface prep and bonding procedures yet the clips were popping off with a < 10 lbf transverse shear load (luckily a sleeve on a clean room jacket snagged a clip; design load requirement was >=100 lbf). This wasn’t a peel load issue, it was a smooth cohesive failure of the entire adhesive bonding surface. Well, after many stressful days and nights, we found out it was the surface prep agent that had contaminated ALL shear clip bonding surfaces with a silicone oil. Contamination had occurred at the manufacture’s facility. All shear clips had to be removed and all bonding surfaces had to thoroughly cleaned and prepped…if your design has similar bending stiffness between the Lexan and CF then please ignore my comments. If there’s a high ratio of the Lexan-to-CF then this is something that should be addressed in the design. Combined mechanical-thermal loading environments aren’t a major concern.

[Chrisyphus]

Originally Posted by M3SQRD

Chrisyphus Engine vs. plenum temperature - good point about it not being hot while on the track when it’s subjected to vibroacoustic response/loading. Heat soak occurs when the plenum is subjected to conductive and radiative heat transfer mechanisms. Max IATs of the e9x M3 is the thermal environment that would occur at the same time as the mechanical loading environment.

Yeah, things are very often more complicated than they seem at first glance. There’s a million variables when it comes to temperature, load conditions, and use cases. Thankfully (not really) I suffer from debilitating self doubt which causes me to cover down on as many situations and conditions as I can dream up & then stress Peter and Abdul to no end with my various contingencies for potential problems that typically never come to fruition.

Peel strength - true, all adhesives have poor peel strength so poor word choice

That comment wasn’t directed at you. You’ve demonstrated significant technical literacy and so I expect you to know that. On the other hand, this is what I do for a living and I am here to sell parts. I have to think about people who do not come from a technical background and do not want those people to infer from your comments that our adhesive is particularly susceptible to peeling, so I need to clarify.

However, I work with thermal engineers that want to use thermally-isolating adhesives that have abysmal peel strength!

Gotta love design constraints from other engineers. Especially when their focus is on an entirely different field. They tend to be overly specific with conflicting constraints. Unless they have significant practical experience.

Thickness ratio of Lexan vs. CF fabric/layup - how thick is the Lexan? Have you estimated the OOP displacement and IP rotations of the Lexan at the bonded I/Fs with the CF?

Yeah, as with so many other things, it’s a lot more complicated than simply relative thickness. With the lexan it’s more straightforward, but we aren’t dealing with a flat or even simple curved part. It’s thermoformed with a multiple compound curves molded into it. This was done primarily for aesthetics, of course, but it has the added benefit of predetermining the direction of out of plane distortion as a result of expansion which makes everything else far more predictable. The carbon, as I’m sure we’re all well aware, is FAR far more difficult (read: nearly impossible without destructive testing) to predict and a thickness measurement is all but useless on its own. The local geometry is every bit as important in determining local stiffness. For transparency’s sake though, our lexan is (and I hate to switch to standard now, but it’s what these measurements came in) 0.125” (1/8th) and the carbon in the bond area is 0.065 (on average) but as I mentioned earlier, that’s only half the story.

The relative bending stiffness of the Lexan and CF will produce a curvature/slope discontinuity and, thus, generate a peak peel loading in the adhesive along the free edge of the Lexan - higher Laxan-to-CF bending stiffness ratio = higher peel load.

Absolutely right. First of all there’s the magnitude of the distortion to consider- .024” across the 15” length of the lexan which nets a potential peak OOA distortion of 0.9mm in the Z+ orientation normal to the plenum body. Within 1 inch of the inner edge of the bond, the distortion is sub 0.1mm this paired with the similar local stiffness of the lexan and the carbon results in a peel force of fuck-all (metric fuck all, of course.)

As you know, workmanship issues during the prepping of bonding surfaces are critical for developing peel strength. One time we had shear clips between composite square tube sections at double lap shear joints that had used standard surface prep and bonding procedures yet the clips were popping off with a < 10 lbf transverse shear load (luckily a sleeve on a clean room jacket snagged a clip; design load requirement was >=100 lbf). This wasn’t a peel load issue, it was a smooth cohesive failure of the entire adhesive bonding surface. Well, after many stressful days and nights, we found out it was the surface prep agent that had contaminated ALL shear clip bonding surfaces with a silicone oil. Contamination had occurred at the manufacture’s facility. All shear clips had to be removed and all bonding surfaces had to thoroughly cleaned and prepped…

Years ago I dropped a tool onto a part we were polishing before shipping, which on its own was annoying and meant more polishing, at least, but to my horror I saw that a small amount of clear had lifted.
So I did what any reasonable person would do and started tapping it all over with a small ball peen and saw that there was poor adhesion all over the part. So then I did the next most reasonable thing and took my hammer to every part on my shelf ~20k worth of parts and thankfully isolated the issue to the most recent batch of parts. Finally isolated the issue to my former apprentice deciding that it was no longer necessary to wipe the surface prep away after scrubbing it on. I had to explain to him that without wiping away the still-wet product in 3 or more stages, you simply spread a perfect film of whatever oils, release agents, and other contaminations the part has picked up through the demolding, trimming, and sanding processes all over the part. Morale was low while we stripped all of them back to substrate. My point being: I feel your pain lol.

if your design has similar bending stiffness between the Lexan and CF then please ignore my comments. If there’s a high ratio of the Lexan-to-CF then this is something that should be addressed in the design. Combined mechanical-thermal loading environments aren’t a major concern.

If nothing else this has been a good opportunity to demonstrate to anyone on the fence about us that at the very least, I have done the work and that we are not, in fact, a basement operation just throwing stuff together.

[M3SQRD]

Chrisyphus

Thanks for the quick and detailed response.

Relative bending stiffness isn’t an issue even with the Laxan 2:1 thickness ratio. Modulus of Lexan is 0.35 msi vs. the CF being in the 10-20 msi range so flexural midulus, D (similar to EI in a beam but for a unit width), is proportional to E*t^3 so the lexan thickness ratio makes it 8x stiffer but the modulus ratio advantage is for the CF with a value of > 10:1 so lack of local flexibility will minimize any curvature discontinuity and, therefore, reduce any adhesive peel loading.

No doubt. It’s obvious you’re not throwing things together in a dark corner of the basement

[e92dud]

What’s the lead time on getting one of these plenums if ordered right now? There’s no picture of the non glass top on the website, has one been produced?

[Adam M]

Have to say as a visiting Brit on this forum, I was really put off by the attacking tone some posters had towards Abdul and Chris.

I’m really glad the thread has ended up productive.

I’ve ordered one of these plenums which I am looking forward to installing in the uk.

I just wanted to point out that on the advice of the guys I have gone with the Lexan window version.

I’m not keen on the appearance but don’t tend to drive with my hood open, I just wanted to point out that my main criteria in induction roar volume and I have been informed that the lexan version is the louder of the two choices. By how much and for what reason I do not know.

So long as it’s the loudest choice, I couldn’t care less if it looks better placed on a Japanese car.

[herrubermensch]

Originally Posted by e92dud

What’s the lead time on getting one of these plenums if ordered right now? There’s no picture of the non glass top on the website, has one been produced?

We finished the production mold for the top part of that mold today (it uses the same bottom portion of the mold as the windowed version) and should pull the first production version of that part later this week and will post some pics. We don't like to show prototypes because prototypes change. We like to show you what you will actually receive. We are generally running about four weeks from order to shipping, but each plenum is made by hand in the order in which it was received, so exact receipt time is hard to predict.

Happy to answer any production-related or other questions you might have, as is Chris. Chris can be reached at Chris@parteeracing.com, and I can be reached at Peters@parteeracing.com.

--Peter

[e92dud]

Originally Posted by herrubermensch

We finished the production mold for the top part of that mold today (it uses the same bottom portion of the mold as the windowed version) and should pull the first production version of that part later this week and will post some pics. We don't like to show prototypes because prototypes change. We like to show you what you will actually receive. We are generally running about four weeks from order to shipping, but each plenum is made by hand in the order in which it was received, so exact receipt time is hard to predict.

Happy to answer any production-related or other questions you might have, as is Chris. Chris can be reached at Chris@parteeracing.com, and I can be reached at Peters@parteeracing.com.

—Peter

Thanks, will be waiting patiently.

[reidetailing]

Originally Posted by Abdul

I received your preorder for the valve covers. I also can’t see a reason why it won’t be. But let me check with the engineering team and get back with you.

Hey Abdul , following up with you on the Nuke rails and your valve covers fitment.