ATS spindle & C6 SKF Hub dilemma -- lower hub mounting location hardware clearance

[Blown353]

Might have painted myself into a corner and looking for some suggestions.



Decided to add a Teves/Continental MK60 ABS setup with the motorsports reflash to my car, so I needed to switch to late C6 hubs with the active speed sensors.

I picked up 4 SKF X-tracker ZR1 hubs along with ARP 12mm x 2.5" studs, not realizing the body of the new X-tracker hubs has a lot more material on it than the old C5 passive sensor hubs that came installed on my ATS spindles.

The new X-tracker hubs went onto my MillerBuilt rear floater setup with zero issues... but the front is a different story.

The bottom hole on the ATS spindle requires you to drill out the threads in the hub and use a thin locknut to install over the stud in the spindle; the problem is the much beefier X-tracker hub body doesn't leave any room for the nut required for the 6 o'clock position on the ATS spindle.

I think I may have a possible solution... but I wanted a sanity check before proceeding.

An M12 x 1.75 socket head cap screw has a head OD of 18mm and the head is short enough to clear the back of the wheel studs.

If I drill an M12 clearance hole in the 6 o'clock hub hole (as is normally required for an ATS spindle) and also bore an 18.5-19mm access hole through the hub flange itself between two wheel studs (like many hubs for other applications already have right out of the box so you can install the hub to spindle bolts) I could then use an 18.5mm endmill to spotface the SKF hub to clear the socket head cap screw, and bolt the 6 o'clock location on the hub to the ATS spindle with a socket head cap screw into the threads on the spindle. From what I measured, spotfacing the hub at 18.5mm for the bolt head would only remove some of the radius on the hub body near the mounting hole, and would either just barely clear or barely kiss the hub body OD above the radius where the hub seal and bearing outer race lies.

If I took this route I would have to remove the stud from the ATS spindle and thread the socket head bolt directly into the spindle. ATS probably used the stud at the 6 o'clock location for a good reason, but I'm not sure if that reason was for strength or simply to avoid having to drill an access hole in all of the wheel hub flanges for bolt access. My thought was as long as there is ample thread length in the spindle bolting into the spindle will be fine, or I could possibly install a time-sert into the spindle if I was paranoid and wanted the threads in the spindle to be stronger.

There were a couple of older threads on here stating the C6 standard hubs (non X-tracker) will work on the ATS spindles and the nut/stud arrangement will clear, and that is the easy way out... but since I already have the X-tracker hubs and can run a mill and have a possible solution in mind I figured I might was well use the stronger X-tracker hubs if I can.

A revised ATS spindle that uses unmodified hubs and uses 3x socket head bolts to attach the hubs to the spindle (as per the original Corvette installation) would be ideal, but I know that's not going to happen as it would require a substantial redesign and Speedtech has tons of money tied up in the forging dies and other tooling and such.

Looks like Church Boys racing has a billet spindle that takes unmodified C6/C7 hubs, but I'm not sure if their spindle is a tall spindle, how the steering arm geometry is, and they also say it narrows the front track width and I already have my Forgelines in hand, so that's a no-go.

Appreciate any input or sanity checks...

[TANKMASTERJ]

I see no reason why you couldn't do what you have described. I installed C6 hubs, the older ones in my ATS spindles. But I had the hubs redrilled for 5X120. It had press in studs with flat heads. I had the new holes tapped for 1/2x20. Not realizing the 12 point studs hit the bolts you are talking about clamping the hubs to the ats spindles. So I put sleeve retainer lock tire on the threads, then cut all the 12 point heads off. I've been running them for a few weeks now no issues, no noise, no vibration.

[protour_chevelle]

Only thing I dont like personally is the idea of threading into the aluminum on the spindle. It was engineered to have a stud live in there and have the torque applied to the stud thread, not the spindle threads. Your timesert idea makes me feel better.

Im also in this situation but haven't starting swapping over yet. Im looking at boring a hole in the wheel flange, retaining the stud in the ATS spindle and going to an ARP 12 point nut with the reduced socket head size. In a pinch at the track if I cooked a wheel bearing, I could clearance with a grinder on a new hub and use a box end wrench to tighten.

[Blown353]

The stud is certainly the best approach, but I'd feel OK with a bolt in the 6 o'clock position as long as there was ample thread engagement. Threading directly into aluminum is OK occasionally (how else are you going to assemble an LS engine for example... lots of threads directly into aluminum) but for threads that see frequent tightening and loosening cycles a steel threaded insert like a helicoil, keensert, or timesert is certainly best. I would not want to constantly be threading that 6 o'clock bolt in and out of the spindle unless it had an steel thread insert, but luckily you're typically not changing hubs every week.

My hunch is ATS/Speedtech used the stud and thin nut arrangement in the 6 o'clock location so they would not have to machine a clearance hole in all their wheel hub flanges to drop a bolt through the wheel flange into the hub and spindle. Far easier and less effort for them to just drill out the threads in the 6 o'clock hub position with a drill press and use the stud and thin nut arrangement rather than fixture the hub and drill a clearance hole for a bolt in every wheel hub that passes through their shop.

After some more measuring today, I think I'm going to try the ARP 12 point nut first.

An M12 x 1.75mm socket head cap screw has a nominal OD of 18mm. The Holo-Krome grade 12.9 socket head cap screws I have measure 0.695"-0.698" at the head.

I took some measurements today on the hubs with gauge pins and also a CMM, and if I spotface the 6 o'clock mounting hole in the bearing hub right at 0.700" then the endmill will not contact the hub housing OD at all-- it will only spotface the bearing housing and relieve the radius at the bottom for the bolt head or nut to sit flat against. I really don't want to take any material off the OD of the hub body as based on a CAD cross section I've seen of the C6 X-tracker hubs the outer row of balls gets somewhat close to that area. I feel OK spotfacing the hub body mounting flanges for the bolt head as long as no material is removed from the hub body OD itself.

An ARP 12 point M12 x 1.75mm nut has a flange OD of 20.9mm, which is far too big and would require you cut into the body of the hub by 1.5mm / 0.060" when spotfacing-- too much material removal from the hub housing OD IMO.

My current plan is to take the ARP 12 point nut and reduce the OD to 0.695", same as it would be for the grade 12.9 socket head cap screws. I'll then bore a 0.700" diameter access hole in the wheel hub between two studs, and spotface the 6 o'clock location of the hub mounting flange at 0.700" diameter for the nut to sit against.

The ARP 12-point M12 x 1.75 nut uses a 14mm socket, and a super thin wall 12 point 14mm socket made for late model BMW spark plugs that a friend has measures 0.695" OD so it will fit through the access hole in the wheel flange... perfect!

This *should* work and allow me to keep the stud in the spindle, which I'd rather not disturb.

I modeled up a fixture plate for the mill today to hold the hub; it will bolt the bearing flange to a plate with M12 x 1.75 shoulder bolts using tight clearance holes in the plate to positively center the hub, use one locator pin sticking up from the plate which will engage through one wheel stud hole to clock the wheel hub and hold it centered between two studs locations over the 6 o'clock position on the bearing mounting flange that is going to be drilled and spotfaced, and has 2x M12 threaded locations for 2x M12 bolts to be installed through 2x wheel stud holes and thread down into the plate hold the wheel hub in place and keep it from rotating/vibrating while boring the clearance hole, and it has a large clearance hole milled in the center of the plate for the sensor to protrude through. The fixture plate should make it nice and easy to hold everything in place while machining and make indicating the proper location really easy-- just indicate that 6 o'clock hole location on the plate, zero the DRO on the mill, then bolt the hub onto the plate, make sure the mill is still at 0/0 on the DRO and go.

Hopefully I'll get a few minutes on the CNC mill tomorrow or Wednesday to make the plate, then I can modify the hubs later this week when the M12 shoulder bolts and the ARP M12 x 1.75 12 point nuts arrive.

Almost gave in and took the easy way out today and ordered 2x non X-tracker hubs and returned the 2 X-tracker hubs I still have in the box, but figured this was more fun with a bigger chance of making an expensive mistake. LOL

[protour_chevelle]

I understand what you are saying with threading into aluminum, I still feel the application is not well suited for a bolt being threaded in there. Theres a reason why GM put the threads on the wheel bearing and not the spindle, much like ATS/Speedtech putting the threads on a stud, rather than thread into. Would it be an issue? Doubtful but I wouldnt put my name to it.

The loading, stress and fastener count/diameters is a contributing factor when using fasteners in aluminum engines such as motor mounts or cylinder heads. Also why in performance application you see the standardized move to stud hardware over bolts, regardless of base material.

At any rate, Do you still have a copy of the CAD cross section of the SKF wheel bearing you mentioned above? I would like to see the thin areas you alluded to.

I think you have a solid path forward now.

[Blown353]

I 100% agree that using a stud in the aluminum threads and not cycling a fastener in and out of aluminum threads is the best approach for long term durability, which is why if I ended up putting a socket head cap screw directly into the spindle from the bearing side I’d install a timesert into the spindle so all the rotating thread contact would be steel to steel. Even if it isn’t 100% necessary, it would give me a warm and fuzzy feeling that it isn’t going to fail in the future.

The CAD cross section I was referring to was taken from a video SKF made years ago when they first released the X-tracker ZR1 hub bearings. The video is on SKFs youtube channel. There’s a portion of the video where they show a very good cross section of the C6 X-tracker hub model, and if you pause the video at that point and take a screenshot you can then snap your own scaled dimensions to it based off actual dimensions from a hub you have in hand and get a good feel for where the bearing races lie and the wall thicknesses of the hub body. The cross section view in the SKF video was good enough to show that you probably don’t want to have the endmill carving a semicircle out of the hub body OD for hardware clearance. If you’re an SAE member, you may be able to get the actual hub model from GM and get exact numbers.


Had another thought, but now that I think about it more it probably won’t pan out…


It may be possible to avoid boring an access hole through the wheel hub for spotfacing the hub flange and also for hardware access if you set up the bearing assembly vertically in the mill and come in from the side with an endmill to do the spotfacing and radius relief for the 12 point nut, and then use a 12 point flare wrench or crowfoot to tighten the nut. However, there’s a few problems with that idea…


1. The flare wrench would only allow a very limited amount of rotation each time before contacting the hub body while you're tightening it, so it would be difficult to get a proper torque reading on it since you're going to have very limited wrench throw and you'll be constantly repositioning the flare wrench.


2. There’s only 0.650” clearance between the mounting flange on the hub and the back of the wheel flange. The ARP M12 x 1.75 12 point nut is 0.504” tall. That only leaves 0.146” of room to slip the 12 point flare wrench in over the nut and under the wheel flange, so you would have to severely shave the height of the flare wrench down, weakening the flare wrench.


3. I'd be worried about putting 90ish ft-lb of torque on a severely shaved down 12 point flare wrench; seems like that's asking for it to spread open and slip. Speedtech doesn't publish the torque specs for the ATS spindle to hub hardware since they want you to send in the spindles for hub replacement, but GM factory C5/C6/C7 hub torque spec is 96 ft-lb. Mind you, that's with fresh GM bolts that have a dry threadlocking compound pre-applied to the threads so the running torque will be a little higher than a clean bolt. Since Speedtech uses liquid loctite on the threads of the 2x bolts and 1x nut I would reduce the GM torque spec about 10-15% as the liquid loctite will act as a bit of thread lubricant.


Because of the possible issues above I’m going to proceed with the access hole through the wheel flange and 12 point nut approach.



Next question is if I only bore one access hole or 5 in the wheel hub flange. 5 holes Would satisfy my OCD for symmetry and maintain balance on the hub itself, but I also know that a single hole in the wheel hub that close to the axis or rotation will only cause a slight imbalance, and that imbalance from the single hole in the flange is going to be negligible compared to any slight imbalance in the rotor and wheel/tire which are much further from the axis of rotation. There have been millions of rear axles and hub bearing assemblies made over the years with only a single hardware access hole drilled in the wheel flange…

[protour_chevelle]

Id stick to your current plan. Crows feet suck, It would also likely have to be an extended foot to fit. Getting proper torque value with a crows foot is very particular. One of my tasks at work is bolting up in oil/gas plants. We avoid crows feet at all costs.

For balance on the wheel hub after a hole is knocked in it... I get ya there with the OCD. I feel putting 5 holes in is reducing too much meat. Personal opinion. I dont think you will notice a vibration but now that you mention it, its going to weigh in your head the first time you feel a vibration and you'll overlook the u joint that wore prematurely lol. you could drill a hole relatively close opposite and that would cancel a ton of it out.

edit: heres the youtube link of the cutaway. Thanks for that! https://youtu.be/T2d8_G0zwAQ?t=25

[Blown353]

I agree, crowfoots are a last resort, especially when applying proper torque is important.

Another option would be a 12 point 14mm torque adapter-- easy to do the math and maintain proper torque. However, you would still have to shave it down to ~0.150 thick to get it between the top of the nut and the back of the wheel flange, and you'd also have to thin it way down to clear the hub OD... sounds like that double-thinned torque adapter is probably going to break when you try and apply 90 ft-lb.

Access hole and thin wall socket will be the way forward. I'll post pics once the job is done, hopefully this week. ARP 12 point nuts are here, shoulder bolts for the machining fixture will be here tomorrow.

[Blown353]

Got the fixture made and the hubs modded today.

My OCD got the better of me and I drilled 2x 0.550" diameter holes on a 4.75" bolt circle (stud circle) opposite of the 0.710" hardware installation / socket access hole on the 116mm bolt circle (hub to spindle hardware bolt circle) for balance, that keeps the hubs perfectly balanced about the axis of rotation. Had to do some digging in the books as I didn't remember the equation off the top of my head to calculate the imbalance forces resulting from only drilling the single 0.710" hole in the wheel flange, but a single 0.710" diameter slug of steel removed from the hub at 58mm / 2.284" distance from the axis of rotation creates a 2 lbf imbalance at 60mph, 8 lbf at 120mph, and 18 lbf @ 180mph (that v^2 gets you at higher speeds...) decided I didn't like those numbers, hence the 2x balance holes.

Anyways, here's the hub mods. 0.710" access hole in wheel flange, 0.695" spotface on hub surface where the 12 point nut will sit, 2x 0.550" balance holes in the wheel flange opposite the 0.710" access hole. The endmill *barely* kissed the hub housing OD by about 0.005-0.007" above the radius for the mounting flange when doing the 0.695" spotface. I was hoping to not hit the hub body OD at all, but that would have meant reducing the OD of the 12 point nut too much for my tastes. I'm not too worried about kissing 0.007" off the hub body with the endmill, if it fails from that there wasn't enough safety margin designed into the hub to begin with, LOL. All the holes and the spotface were interpolated with the CNC on the mill.

Hub sitting on the machining fixture. Start with 3x 12mm x 1.75 shoulder bolts holding hub to plate to accurately and repeatably locate it, 1x 1/2-13 bolt at 12 o'clock with the top 0.700" turned down to 0.486" to slip into a stud hole and clock the wheel flange properly, 2x M12 x 1.75 bolts through 2x wheel stud holes at the front to clamp the wheel flange to the plate to keep it from chattering while cutting.


Interpolating the 3x holes in the wheel flange and spotfacing & boring the 6 o'clock hub hole. Ended up adding a kant-twist clamp at the back for additional clamping on the wheel hub. If I was doing more of these, I would modify the locator pin/bolt at the 12 o'clock position and cut M12 x 1.75 threads sticking up above the 0.486" stud hole locating diameter and use an M12 nut and washer to clamp the hub down at the 12 o'clock position, that way I could probably omit the clamp.


Done-- 3x holes in flange, 1x 0.710" socket/hardware access hole, 2x 0.550" balance holes


Showing side of hub, kissed the housing above the mounting ear radius by about 0.005"-0.007"


Gauge pin slipped in showing access and clearance for 12 point nut and 14mm thinwall socket



I still need to go back and deburr/chamfer the 3x holes in the wheel flange and then use a cratex wheel to smooth and radius the sharp edges on the hub body mounting ear where the spotface kissed the radius, figured I don't want to leave any sharp edges there to act as possible stress risers.

Ran out of time before I could reduce the OD of the 2x ARP 12 point nuts to 0.690" to fit on the spotface and also ream out the 10x stud holes to 0.495" for installing the ARP 12mm studs (holes in the flanges from SKF are at 0.488", ARP recommends a 0.493-0.503" diameter in a steel hub for the press fit for the 12mm studs. They'd probably press in OK, but that's a really heavy press at 0.488")

One thing to note, if you are using hubs with the speed sensor you need to pay very, very close attention to which hole you drill the threads out on and spotface on the hub body as there is a right and left side orientation-- the sensor wire must exit pointed to the connector support bracket. If you consider the spindle stud hole to be 6 o'clock, 1 hub needs the speed sensor wires oriented to about 10 o'clock and the other hub needs the wires oriented to about 2 o'clock. If you drill and spotface the wrong hole in the hub, the speed sensor connector will not reach the support bracket.

I'll post pics of the hub on the spindle after the ARP nut ODs are turned down and I test fit everything. I still need to make new upper ball joint inserts for my spindles and fix those, as last year I accidentally installed the wrong taper upper balljoints into them and boogered up the taper in the spindle inserts. By the time I noticed things didn't feel right it was too late... oops. (Note to self, always test fit the ball joint in the spindle by hand with a sharpie or dykem to check the taper fit BEFORE snugging the nut up with a ratchet.) Luckily I have early ATS spindles with the replaceable ball joint inserts, so I can machine new inserts and then shrink fit them into the spindle with liquid nitrogen (they're about a 0.0035" shrink fit.) I was hoping Speedtech still had some old inserts still hanging around the shop, but its been too long since they stopped making the spindles with the inserts and they didn't have any, so I have to make my own. I already have 7075 barstock and a grooving tool for the retaining ring groove, but had to spend a decent amount on a good spiral flute taper reamer-- way cheaper than buying new spindles though. Once the new upper inserts are made and installed I can mock the front end back up, do the mods on the frame for the coilover upper mount, and order coilovers.

Got a beat to hell Mk60 ABS unit off ebay a few days ago so I can figure out where to mount it and fab a mounting bracket and figure out brake line routing. Decided I didn't want to be doing any fab work around the new Mk60 unit with the motorsports reprogramming and risk getting debris in it or accidentally damaging it. The beat to hell used Mk60 off Ebay was less than $35 shipped. If anyone is considering doing the Mk60 ABS swap, Doug at RHT Motorsport Solutions is a really good dude to work with.

[chevelletiger]

"Speedtech doesn't publish the torque specs for the ATS spindle to hub hardware since they want you to send in the spindles for hub replacement"

Well I wish I knew this before buying my spindles.
THEY SHOULD let the buyer know so they can replace themselves

[Blown353]

"Speedtech doesn't publish the torque specs for the ATS spindle to hub hardware since they want you to send in the spindles for hub replacement"

Well I wish I knew thus before buying my spindles.
THEY SHOULD let the buyer know so they can replace themselves

Speedtech is probably worried about someone breaking the loctite on the stud loose when removing the nut while changing the hub. This risks having the stud back out of the spindle and having less engagement into the spindle body and also risks snagging the wheel studs on the back side of the hub if you reinstall it with the stud protruding too much. Or the stud could end up threading in deeper while reinstalling the nut and end up set too deep and then there isn't adequate thread engagement into the nut.

It's not a difficult job, but you do have to pay attention. As I mentioned above, GM spec is 96 ft-lbs for C6 & C7 hubs and spindles, but that is with factory bolts that have preapplied "gummy" threadlocker on the threads which will increase the running torque. With liquid high temp / high strength loctite on both bolts and the nut (which is how Speedtech assembles them) the loctite will lubricate the threads a bit and I'd probably back off that 96 ft-lb about 10-15%.

Also, based on some recent pictures of newer AFX spindles I found it looks like they are now shipping them with the C7 X-tracker hubs. Not only does Speedtech say they now use the C7 hubs, but the most recent pictures on their website show the larger X-tracker hub installed on the spindle and you can also see where Speedtech has machined a hole into the face of the wheel hub for hardware installation and socket access, much like I did.

If someone has a newer production AFX spindle that shipped with a C7 X-tracker hub installed, I'd love to see pictures of the hardware they used to attach the hub to the spindle at the stud location and also what machining they did on the wheel hub, specifically if they reduced the OD of the nut and how much they clearanced the hub body for the nut.

[chevelletiger]

^^I send some pictures of my ats spindles I got them in the beginning of the year.

[badbu68]

Speedtech doesn't publish the torque specs for the ATS spindle to hub hardware since they want you to send in the spindles for hub replacement, but GM factory C5/C6/C7 hub torque spec is 96 ft-lb.

Sounds more like a liability issue than they want your money.