[Alex W]

I just finished the first 3D printed prototype on my new knuckle design for the front.

Bolt on bearing (of course), plus adjustable Ackerman and a little bit of weight reduction (about 2lb per corner).

 

[DefSport]

Oh yea… looks very nice!

Did you benchmark stiffness vs OE spindle? I’ve been shocked of how good readily available photo scanning to 3D modeling has gotten in the past 4-5 years if you don’t have a good OE spindle model.

 

[Alex W]

Oh yea… looks very nice!

Did you benchmark stiffness vs OE spindle? I’ve been shocked of how good readily available photo scanning to 3D modeling has gotten in the past 4-5 years if you don’t have a good OE spindle model.

I have not, but the stress looks good (for fatigue) at the estimated loading for 1.75g, so there should be pretty good stiffness there. I have been tempted by some of the cheaper 3D scanners recently, might be something I need to look into further...

 

[1950884]

Looks terrific!

 

[Zigspeed10]

These look great! What hub is that from? Does the brake rotor need to be positioned in between the bearings? That might not even matter but I assume it would cause some weird loading on the bearings if not, only ask as those hubs look like they have a lot of offset.

Great to see these being made, count me in for a set.

 

[DefSport]

I have not, but the stress looks good (for fatigue) at the estimated loading for 1.75g, so there should be pretty good stiffness there. I have been tempted by some of the cheaper 3D scanners recently, might be something I need to look into further...

Check this out, seemed very easy and I checked the model and it was within .010” to my measurements on another upright after rescaling based on a known dimensions.

Z32 Rear Upright 3D Scan - Nissan Road Racing Forums

 

[Alex W]

These look great! What hub is that from? Does the brake rotor need to be positioned in between the bearings? That might not even matter but I assume it would cause some weird loading on the bearings if not, only ask as those hubs look like they have a lot of offset.

Great to see these being made, count me in for a set.

2002-2011 Camry rear. They do have a lot of offset, and that's the one thing I don't really like about them, but there actually weren't that many choices of Toyota hubs of the proper year range to still have passive ABS sensors that will work with the stock (or in my case Spyder) ABS system. I limited myself to looking at Toyota stuff so as to not require custom brakes, not to mention questionable ABS compatibility if pulling from other brands, and I also wanted a relatively small footprint on the knuckle side so that I didn't have to build in TOO much geometry correction (and thereby limit control arm options to those that use the stock ball joint). The bearing diameter (and seal diameter, which is important since that's where the stock hubs break) is about 13% bigger than stock (49mm vs 43), which should provide a pretty significant increase in strength on the hub since the area moment of inertia of a shaft involves the diameter^4, so a little extra diameter gives you a lot of extra strength.

 

[Alex W]

On the subject of hub strength... I made a quick model of a stock MR2 hub (using measurements from a hub that is, surprise surprise, broken), and ran a simple static stress analysis on it using the loading I estimate for 1.4g lateral. The result is a peak stress of 128ksi, at the exact location where the stock hubs fail.

I modeled the Camry hub using what I could get for measurements without destroying one, plus my best estimates based on NSK's spec sheet, and with the same loading and constraints got a peak stress of only 55ksi. Which isn't to say they will last forever, or that there couldn't be some other weak point that I haven't considered, but they certainly should be a big improvement over the stock ones.

Broken front hub for reference:

 

[DefSport]

Yikes, I didn't realize the radius on the stock front hubs was that bad. That's like max Kt, minimum life design for the mass. lol

 

[Alex W]

Yikes, I didn't realize the radius on the stock front hubs was that bad. That's like max Kt, minimum life design for the mass. lol

Yeah, they are really bad. The 93+ version is slightly better, in that it has that thickened area around the center, but still no significant radius between the flange and the stub.

Like I said, I am somewhat guessing at the radius on the new one, based on the drawing on page 39 of this NSK document: http://www.nsk.com/common/data/ctrgPdf/e4201b.pdf

The dimensions for the NSK bearing reference # 49BWKH17 from the table below exactly match the dimensions for this Camry bearing (even though the part number is slightly different), and based on the fact that this document exists, and that the NSK bearings I bought have a Toyota part number stamped on the ABS sensor, I suspect NSK may be the OEM supplier for these.

 

[arthursc2]

Wonderful!
When you have these ready, I would love to get a package deal with rear/front and coil overs from you. Bring the old girl into the modern era for sure

 

[reboosted]

Silly question time: Since you have a printed piece, have you looked into metal printing these? Apparently this makes manufacturing of Titanium parts more affordable 😈. You might be able to eliminate the multi piece design by not having to deal with machining or material stock restrictions. Unless your planning on using interchangeable parts to change geometry for specific applications, like road race vs. drag race and so on. The Best Metal 3D Printing Materials for Additive Manufacturing (markforged.com)

 

[Alex W]

Silly question time: Since you have a printed piece, have you looked into metal printing these? Apparently this makes manufacturing of Titanium parts more affordable 😈. You might be able to eliminate the multi piece design by not having to deal with machining or material stock restrictions. Unless your planning on using interchangeable parts to change geometry for specific applications, like road race vs. drag race and so on. The Best Metal 3D Printing Materials for Additive Manufacturing (markforged.com)

The multi piece design is how I get adjustable ackerman (move the toe arm to a different position / orientation). And the bolt on strut tab allows for (maybe someday) a double a-arm conversion on the front like I did on the rear. So there are benefits to the modular design outside of just reducing material size.

Even 3D printed, I can only imagine how expensive these would be in titanium!

 

[DefSport]

3D printed metal is not cost effective versus relatively simple subtractive machining on 6061 aluminum. Yes, the cost per lb or kg of powder is fairly low, but it has its own set of design challenges and the machine print time plus cleaning is very pricey. Additive manufacturing by definition contains many small flaws at best, so you have to account for that with lower stress states to get the same damage tolerance life.

Besides, 3D printed aluminum and titanium are kinda garbage for mechanical properties. It works much much better for very tough materials like Inconels or overaged PH stainless steels, even then, the subtractively manufactured parts still tend to come out lighter for the same life span.

(the above is from direct experience with it in some of the highest performing/largest metallic additively manufactured parts currently made)

 

[SixAppeal]

2002-2011 Camry rear. They do have a lot of offset, and that's the one thing I don't really like about them, but there actually weren't that many choices of Toyota hubs of the proper year range to still have passive ABS sensors that will work with the stock (or in my case Spyder) ABS system. I limited myself to looking at Toyota stuff so as to not require custom brakes, not to mention questionable ABS compatibility if pulling from other brands, and I also wanted a relatively small footprint on the knuckle side so that I didn't have to build in TOO much geometry correction (and thereby limit control arm options to those that use the stock ball joint). The bearing diameter (and seal diameter, which is important since that's where the stock hubs break) is about 13% bigger than stock (49mm vs 43), which should provide a pretty significant increase in strength on the hub since the area moment of inertia of a shaft involves the diameter^4, so a little extra diameter gives you a lot of extra strength.

How much further out do these hubs move the front wheels?

 

[Alex W]

How much further out do these hubs move the front wheels?

None. I was able to recess the hub into the machined part so the final geometry is the same. So the larger offset isn't an issue other than that it complicates the design a little. Maybe also increases stress on the bearing housing portion of the hub, but I'm less concerned with that than with the hub itself.

 

[Alex W]

Finally made time to get the prototypes test fit on the car. Found a couple of small changes I need to make, but nothing too major.

 

[Eloquaint]

I love the bolt in hub design, it's such a huge time saver. No need for a press and 100 different sized pucks and sleeves to get the stuff out. Not to mention the puckering terror of getting the bearing to move out of the knuckle, usually with some precarious stack of spacers and shims in the press.

 

[DefSport]

Nice - when is the front double a-arm setup coming?

 

[cbrkp]

Nice - when is the front double a-arm setup coming?

+1 to this