Rob T's 818R Build Thread

[Rob T]

Track Day was a Success! Weather was perfect...65F at start, 81 at finish, low humidity, some wind. Ran 4 sessions at CMP of 7-8 laps. All temperatures were good. No fuel system issues. No timing issues (must have been heat last time with Cobb pulling timing). Only issue was my run group was a bit crowded and I had a hard time getting clean laps. I had fun! More to follow.

[Rob T]

Here are some numbers from my track day. Video to follow when I have a chance to process it.

Best Lap: 1:54.070
Coolant: <190F
Oil: 195F Max
IAT: 111F Max on last session - No timing issues
Trans Temp: 195F Max on circulating fluid, 215F with IR on back top of trans
AWIC supply water Temp: 100F on last session.
Max Speed: 113mph
Max left G: 1.44
Max right G: 1.38
Max acceleration: 0.82g
Max braking: 0.96g
Calculated Torque: 345ft-lb
Calculated HP: 254
I used 11 gallons of 93 octane gas on 32 laps or 73 track miles - 6.6mpg

[Hobby Racer]

I used 11 gallons of 93 octane gas on 32 laps or 73 track miles - 6.6mpg

Wow, I only get about 3.5mpg at Watkins Glen. That's about 1 gallon per lap!

[Bob_n_Cincy]

Wow, I only get about 3.5mpg at Watkins Glen. That's about 1 gallon per lap!

Quit slowing down for the corners and your MPG will go up.

[Rob T]

Hobby: Those two extra cylinders don't move for free, you know....

[aquillen]

One thing I'm not going to complain about is getting 28 - 30 MPG this year in my 3.0 818, highway and often 70+ MPH for a big chunk of many fill-ups. I've done 4900 miles this year, don't tell Haggerty...

[jforand]

Great to hear you are up and running again Rob.

Wow, I only get about 3.5mpg at Watkins Glen. That's about 1 gallon per lap!

That is way lower than I would have expected.

My '16 Mustang GT with a big old 5.0L V8 weighs about 3850lbs with 435 stock NA HP. My record low track MPG is 5.6MPG. I would have thought at a car with half the weight and something close to the same or less horsepower would have done much better on fuel.

[Rob T]

Canceled my scheduled track day today. It's been raining for three days in my part of SC. Open top trailer, open top car and rain are not a recipe for big fun. Nothing scheduled at the moment. I hope CMP adds some days in December, January or February. It's disappointing, but all things happen for a reason. With my "extra" time, I can do an oil change on my daughter's car.

[Rob T]

BACK TO THE TRACK! I made it to the track (CMP) yesterday for my first session of the year. I spent Thursday doing and oil/filter and trans oil change as well as checking everything out. Interestingly, I did find that the left side axle nut had loosened about 15 degrees during my last track day, so I re-torqued that and made the "dimple" a little more robust. All three other nuts were fine.

I did three morning sessions before I packed it up and left - two reasons - my left front tire was going flat (not sure why yet) and a big storm was headed our way. At the end of the day, my Federal tires have 34 heat cycles on them and they are still doing great. I managed to log my second fastest lap time ever at CMP. It was between 45 and 50F which made the turbo happy, and everything stayed cool....Only about 10 cars in my run group, so lots of space on the track.

My second session was my best, with the fastest lap. Here are the stats from that lap....

Time: 1:53.332
Braking g (max) 1.06
Accel g (max) 0.75
Left G (max) 1.38
Right G (max) 1.28
Average spd 72.74mph
Max spd 116 mph
Calc TQ max 308 ft-lb
Calc HP max 256

I still have a lot to learn, and seat time is helping. I feel good about how the car is running, especially the dry sump with the additional 1/2 stage. I am working on braking later and better lines through the corners.

Since I started tracking the car in 2016, I have logged 20 track days and a total of 71 sessions of between 5-10 laps.

[Hobby Racer]

So jealous! It snowed today and we got 8 inches

[Rob T]

Time to Upgrade my safety equipment: One of the things the new owners at CMP implemented was checking dates on safety equipment. My 6-pt harness and Hans device were expired. I replaced my harness with a rated Racequip 6-pt belt that I got from Naroescape Motorsports. I got my Hans recertified (new tether and sticker) at Discovery Parts in Dawson GA. You can just ship it to them and they turn it right around. Both vendors had exceptional customer service.

You'd think putting a new harness in would be "easy". Took about 3 hours by the time I removed some of the seat bolts and adjusted everything the way it needed to be.

[Rob T]

I was able to make it to CMP again yesterday. Weather was great, if a little windy. 50F for the first session, about 65 for the third. I drove 3 of 5 sessions. There was a 2 hour gap after the 3rd and before the 4th and I had a two hour drive to get home. A few more cars this time (about 20 in the run group), so multiple clean laps were harder to come by. Analysis to follow.

[Rob T]

Here are "stats" from round 2, my best round.

1:54.692 - a bit slower than last time, but track more crowded, so I didn't get as many clean laps. Also, 37 heat cycles on the tires, which are still holding up well and are good for practice.
9.8 gallons of fuel for 27 laps at 2.279 miles per lap or 61.53 track miles. 6.27mpg....
Braking g max: 1.06
Accel g max: 0.875
Left g max: 1.32
Right g max: 1.41
Calculated HP: 265
Calculated max TQ: 296
Max spd: 112mph

No issues with oil temp, coolant temp, AWIC temps. Trans temp got to 196F. Overall, a great day.

[J R Jones]

Rob, Rear axle nut torque?
jim

[Dave 53]

I noticed my right rear axle nut loosened up at the track several weeks ago (Big Willow Springs). A professional race mechanic friend that does Subaru rally cars said it's a common Subaru issue and the axle nut torque is "4 feet of wrench leverage and tight until it won't move". Tightened it up and hit the track the weekend before last (Thunderhill West). After Saturday, I noticed the nut loosened up just a bit, but I couldn't find the right socket in the pits. Just the right rear. After Sunday's sessions, I noticed it loosened up about 1/8 of a turn. Got home and it didn't loosen 1/8 of a turn. It loosened 1 1/8 turns! 6,000 road miles without a problem - this is a track car issue. Hammering the nut into the axle keyway isn't enough.

My friend suggested drilling and pinning the nut. I've done it on go karts and not sure I'm gonna do it on the 818. But of course, I will be bringing the socket to future track days and snugging up the nut with every tire pressure check.

If you track or autocross, I'd suggest taking the nice Factory Five axle nut covers off the wheel for regular axle nut inspections.

[J R Jones]

Dave,
My theory based on exhaustive discussion on the earlier failure is that the rotating tolerance in the drive splines allows micro-rotation of the shaft causing the nut to incrementally loosen.
The peen is not enough, perhaps safety wire would help. Since the right side is not loosening, MAYBE if the loosening side had left hand threads, the micro rotation would not be problematic.
jim

[Rob T]

From the previous session on 3/11, I noticed the left rear nut (the one that came off at VIR and caused all the excitement) had loosened just a bit against the peen mark. I snugged it back up before the track day on 3/25 and made the peen a bit more severe. I had torqued the nut to 160 ft-lb.and bent in the tab. No issues on the left front. All of the nuts are marked and I check them regularly.


I'll let Mitch Wright weigh in, but I believe he told me he used to put blue loctite on the nuts. This is a problem we need a solution for.

What does anybody think about tapping the nut at the collar and running a set screw into the keyway on the axle shaft?

[Hobby Racer]

What does anybody think about tapping the nut at the collar and running a set screw into the keyway on the axle shaft?

Ooh, that sounds like a good idea! I was going to go the locktite route but I like the set screw idea much better. The hardness of the shaft means the set screw should not damage the shaft at all.

The trick is you'll have the install the nut first. Mark where the set screw needs to go then remove and drill / tap it and reinstall. A bit tedious, but good piece of mind.

[J R Jones]

From the previous session on 3/11, I noticed the left rear nut (the one that came off at VIR and caused all the excitement) had loosened just a bit against the peen mark. I snugged it back up before the track day on 3/25 and made the peen a bit more severe. I had torqued the nut to 160 ft-lb.and bent in the tab. No issues on the left front. All of the nuts are marked and I check them regularly.


I'll let Mitch Wright weigh in, but I believe he told me he used to put blue loctite on the nuts. This is a problem we need a solution for.

What does anybody think about tapping the nut at the collar and running a set screw into the keyway on the axle shaft?

Rob, "No issues on the left front." Good to observe that but the dynamics are not the same. Yes the brakes arrest the wheel from turning but the spindle shaft does not decelerate or accelerate. It is (rotationally) unloaded.

The rear hubs are similarly arrested by the brakes but the hub and axle is coupled to the engine and transmission.
Whenever the drive is engaged to the engine, and you are not feathering the throttle, there is load into the drive wheel hubs.
The load reverses:

* Under acceleration the axle to hub is trying to spin the tires. Hard upshift = big load spike to the splines.
* Under deceleration, the hub to axle spline is spinning the engine. Hard downshift = big load spike to the splines.

When the splines wear, the rotational clearance increases and the torque spikes increase.
Expecting the nut to constrain the drive torque under racing loads is a leap of faith.
Tight splines are more resistant to this problem, I do not know if Loctite on the splines would survive. It certainly would complicate service.
jim

[Dave 53]

Ooh, that sounds like a good idea! I was going to go the locktite route but I like the set screw idea much better. The hardness of the shaft means the set screw should not damage the shaft at all.

The trick is you'll have the install the nut first. Mark where the set screw needs to go then remove and drill / tap it and reinstall. A bit tedious, but good piece of mind.

The nut is recessed into the hub, so there doesn't seem to be access to a set screw tapped into the nut. Doesn't look like there is enough meat on the collar for a set screw.

The method of peening the collar into the keyway might make a difference. So that it isn't just a smooth dimple, but the collar actually gets cut in a way that wedges against the key way like a fish hook.

Maybe other than peening technique, I'm not seeing a solution that doesn't require pulling the axle off the car to drill a proper hole. Doesn't seem to be a street driving issue - only a track issue. At the track, I'm going to inspect after each session and give it a snugging up if there is any movement. Easy solution - just gotta bring the socket and know to do it.

Interesting that on my car, it's the right rear loosening and on Rob T's car it's the left rear.

[J R Jones]

Rob, I have always had a keen interest in fasteners, and it is a broad subject.
I referenced my FFR Assembly Manual and it does not specify torque beyond installing the axle nut finger tight during assembly WT? The WRX blog states 162 lb ft.

I have used jamb nuts but I never realized the science of using them. This technical paper/video describes the optimum way jamb nuts are applied:
https://www.boltscience.com/pages/twonuts.htm

Near the end they are described as perhaps the best method of retaining nuts.
The thin nut nut is installed first and torqued to 30-50% of total torque or 80 lb ft.
Then the wide nut is installed to total torque.

You should visualize that when the thin nut is installed and torqued, the force is loading the nut outward from the hub.
When the wide nut is (full) torqued against the thin nut, the thrust against each other pushing out on their respective threads..
The retention advantage is that total torque is achieved between the nuts and across 150% of one nut threads. It seems to me that even if the shaft to hub is rotating, the nut retention is between the nuts, not the hub surface. If the shaft rotates cyclically the nuts turn with it, but not loosen.

This all assumes there is enough axle shaft for two nuts. You may have to make thin nuts from thick nuts.

What does Sgt Gator do to retain axle nuts in endurance races?
jim

[carnutdave]

I was on a Formula SAE team in college and we used locking spanner nuts to retain our front hubs in the hub carrier. They used a Nylon locking element and didn't need to be peened. We also used smaller Nylock nuts on the axles. Never had any issues with them loosening to my knowledge. I did a little digging and it seems the WRX axle nut thread is M22-1.5 (check out Dorman PN 615-160). McMaster Carr carries Nylock spanner nuts in that size, PN 3552N29 and may be worth a try. May have to make a tool that adapts the spanner flats to a torque wrench though.

[J R Jones]

Rob, Sometimes no reply is a reply.
We can agree track sessions is the most expensive way to test your prospective solutions. High risk too. Here is another suggestion.
Shop test your axle nut retention configurations and pick a winner Use "breakaway torque".
Assemble the axle nut to torque and remove it with the torque wrench. What is the breakaway torque?
Reinstall the nut and peen the interference tab. What is the breakaway torque?
Loctite the threads, I suggest red. What is the breakaway torque?
Install Nylock nut. What is the breakaway torque?
Install jamb nuts. What is the breakaway torque? This is more difficult as an effective test is breakaway torque applied to the innermost nut.
jim

[Rob T]

I like the idea. I have been tied up a bit lately and did not reply quickly. I have also been thinking about the idea. In my heart, I really like the positive stop idea, like the "fish hook" that was described above. I'm thinking: torque the nut. Mark it, remove it, then at the mark, use my dremel cutoff wheel to cut the outside peen edge of the nut parallel to the axle, reinstall the nut, then really bend the one end at the cut down into the keyway on the axle. This would provide a positive stop and not just a peened edge to be "rolled over". It would be relative simple to bend it back up when then nut needed to be removed.

[J R Jones]

Rob, cutting the lock flange would IMO weaken the material cross section.
Brainstorming: You could drill and tap the hub next to the torqued nut flat and install a socket head cap screw or a roll pin. It would likely be the perfect location once. Might work in subsequent installations, might need a second hole.
Kind of an act of desperation.
jim

[jforand]

I would be VERY CAREFUL with Red Locktite on the axle nuts. I have this issue on my Mustang and finally got annoyed enough to red locktite them. The next time I had to pull the nut off I tore up the driveshaft threads. This was of course the day before a track event and I spent all day running all over Atlanta trying to get the dye, a socket to fit it, and a new nut. Of course they formed the perfect triangle around the entire city.

Red locktite really is considered permanent unless you heat it and I think it is somewhere near 500 deg F when it liquifies. Not a huge fan of pouring all that heat to them. I use a lot of blue locktite, mark the alignment and keep and eye on them.

[J R Jones]

Rob, Red Loctite has been good to me, or perhaps I am lucky. I have even used green retaining compound in desperate times.

Safety wire is a traditional solution easier than some of our other discussion. It would be interesting to accumulate race time on a safety wired set-up, then, assuming it is intact, remove the safety wire and measure the breakaway torque. That would tell you if the stack-up under the nut is diminished, causing the nut to loose clamp load, (and loosen).
jim

[Rob T]

Lots of really good input here and a lot to consider. I always have the 32mm socket with me so checking it after every session is easy. The last three sessions, especially after I made the divot bigger, there has been no movement past the keyway.

[Dave 53]

Seems a castle nut is the answer.

On my go kart, I've drilled lots of nuts and bolts. We wired or pined any bolt that got hot and used Nylock nuts on bolts that stayed cool. Never used loctite. Castle nuts are preferred in locations that the nut has to routinely come on and off. On a go kart, the front axles because we would routinely make track width adjustments. A castle nut would also solve our 818 on a track issue.

To drill a proper hole in the axle, I'd put the axle in a drill press because the hole has to be pretty precise. The hole has to be at, for example, 12:00 and 6:00. If the hole is misaligned, if it's at 12:00 and 6:30, the pin wont got through. I've never had much luck freehand drilling a bolt with enough precision to use a castle nut.

I'm not gonna pull my axle just to drill a castle nut hole, but next time it's off for some other reason... Until then, routine checks between track sessions.

[Sgt.Gator]

Up to now we haven't done anything outside the Subaru FSM to secure our axle nuts on both the 818 and the STI. The simplest additional would be to use Loctite. I no longer use Red now that they have Orange. At PRI 2019 the Permatex engineer said the Orange is as strong as Red but can be broken free with normal hand tools, no heat required.
For it to work the axle obviously can't get too hot. I'll put our IR thermometer on them at our next track day to see how hot the nuts are getting. I'm guessing that if the wheel bearing is failing that will heat things up. Maybe a good indicator? Just guessing here.

[Rob T]

Thanks Gator. That will be good to know. I know how hot my rotors are after about 3 minutes. Just shy of 300F with an IR. No real science, but a data point. No way the axle or bearings would be that hot strictly from braking, and for me, I am only driving up to 25 minute sessions.

[Sgt.Gator]

It may be worth noting that the axle nuts are one time use nuts. You're supposed to throw the used ones away and use a new one anytime you remove and install them. I keep a couple of old ones around for the Cam gear weld on nut trick.

[J R Jones]

Sgt Gator, That is a terrific fix but also an excellent repair video. That guy knows his craft and how to demonstrate in less that 180 seconds. Great communicator.
BTW I pulled the oil pump (in-car) on a Wisconsin H6 with 180K miles. I had to weld-extract eight (of 64) M6 cam cover bolts.
jim

[jforand]

Great video on the cam sprocket bolts. I have personally had to do it three times. Worst designed bolt ever! The allen head is simply not capable of dealing with the breaking torque generated by that large flange. I would love a Hex or Torx alternative.

[Sgt.Gator]

Great video on the cam sprocket bolts. I have personally had to do it three times. Worst designed bolt ever! The allen head is simply not capable of dealing with the breaking torque generated by that large flange. I would love a Hex or Torx alternative.

Here you go, TIC FU Cam Bolts. They are sold by many of the tuners, Outfront, FastWRX, IAG, Flatirons...but may not be in stock. Call before you order them.

https://www.outfrontmotorsports.com/...it-single-avcs

https://www.flatironstuning.com/tic-fu-cam-bolt-kits

[jforand]

Ah, super awesome. So I think I remember torquing to a specific value and then it was another 45 degrees. This is the classic instruction on a torque to yield bolt. 52 bucks a pop gets steep, but then again I think the Subi ones are 8-10 each as well. Maybe they aren’t torque to yield and then a one time purchase ends the saga. I’ll have to research

Thanks.

[Rob T]

Track day today at CMP. 5 cars in my run group. Perfect weather. It doesn’t get any better. Did three sessions of about 7 laps each. Car ran great. No issues with oil temp or loose axle nuts. More to follow.

[Rob T]

It took a me a while to get my data analyzed from the 4/22 track day. Life has been busy and I will not likely have another track day until August at the earliest. Overall, a good day with three sessions and a total of 25 laps. I used 10.2 gallons of gas. After today, I have 40 heat cycles on my Federal Tires. Tread still looks pretty good and my times have not dropped too dramatically.

Best Lap: 1:54.198
Best Theoretical Lap: 1:53.087
Braking g's max: 1.11
Accelerating g's max: 0.84
Right turn g's max: 1.38
Left turn g's max: 1.37
Horsepower max: 259
Torque max: 304
Max speed: 111.61
Highest average lap speed: 71.99

Oil time, Water temp, pressures all good.

[Sgt.Gator]

Those are impressive Lat G's for old heat cycled Federals! You're doing great. Maybe time to get some A7s and head for Road Atlanta!

[Rob T]

Keep in mind that those are "max" numbers per session. When I look at the friction circle and the "cloud" of data points, there are a lot of values in the 1.25g range, which is still really awesome. The filtered data has me at a solid 1g+ on the longish sweepers. And that's me driving at 7/10.

The cool thing about these tires is that they have been extremely consistent for me and, of course, a great value.