S2's Build Thread

Scott

Member
Lifetime Supporter
Roger, those were a $4,999 Superlite option, but they’re no longer offered. I'm on my third set of pillow blocks, but I think that I have the front one sorted out. I'm waiting on some parts from Superlite to finish it up. The rear sway bar is more complicated and it doesn't make sense to do much with it until I figure out my tail and exhaust.

Superlite did a nice job on the sway bars. There isn't a lot of room for the front one. You're going to need a good fabricator to make them. In addition to welding you need a lathe to create an interior groove for the snap ring that holds the bearings in place.

Other than the actual sway bars, you can buy everything that you need from HRP World here. Swaybars & Components - Cockpit Adjustable

Superlite used 1-3/8" tube which is bigger than the pillow blocks, but HRP will custom machine a set for you. If you move forward I can share more about what I've learned. I haven't written it up yet as I've been chasing my tail a bit and I haven't solved it yet. I'm not going to use my Genesis 5-position dual sway bar controller, so PM me if you're interested.
 

Scott

Member
Lifetime Supporter
Pnut and I walked all of SEMA last year... which was a hell of a lot of walking. One of the primary things I was looking for was a rotary controller that I could integrate into my MoTeC ECU and PDMs. While I found several they were typically designed to work with a proprietary screen which meant that integration was a big question. I was also really disappointed that MoTeC didn’t have a booth.

However, this week I received some parts from my MoTeC supplier and this marketing flyer was in the box...

The ability to configure menus and sub menus via their PC-based application is really cool. They don’t have anything on their website yet and the spec sheets aren’t available yet, but I have one on order.
 

Scott

Member
Lifetime Supporter
I've been having some issues figuring out my Penske shocks and pnut visited me last week to help with the car and sort through the following issues:

  • I couldn't get the front ride height lower than 4.75".
  • The shocks are too long and required two people and a crowbar to install them. In addition, the wheels seriously hit the body when the suspension was at full droop (i.e., when the car is jacked or on the lift) which would eventually damage the body and splitter. Worst, it required me to separately jack the lower control arm to get the wheels off.
We solved all of those issues and more. Plus we found a way to mount the shock such that it can be easily removed without needing to even touch the upper control arm! We also decided to add a bump stop and a bump spring.

RIDE HEIGHT
Given the condition of the roads in in Boston, 4.5" seemed like a good target. Even if I adjusted to last thread on the shock body, I couldn't get any lower than 4.75" with the supplied 400 pound / inch springs and I want to run 600-700 pound / inch springs.

When the Penskes are fitted with hydraulic lift rams (to raise the nose for speed bumps, steep inclines, etc.), and zero-rate springs (to keep the springs aligned when at full droop) and short 4" main springs there aren't any threads left to lower the ride height. The zero-rate springs and associated spring divider take 0.8" when fully compressed. I might have been able to achieve the desired ride height without them, but it's a real pain in the ass to keep lifting and dropping the front end to get the shocks to seat properly without them. In addition, you have the potential for an unsafe situation if you unload the front end while driving and the shocks don't settle properly.

Unless I wanted to buy new shocks (very expensive) the only solution was to move the location of upper shock mounting point. My first attempt was to machine an aluminum spacer to relocate the existing steel bracket. Since this moves the bracket's top bolt directly over the monocoque's weld bead that bolt must be solely supported by the bracket. The bolt is 5/16" so I used some 3/4" stock.

The new milling machine and its digital read out (DRO) came in handy. As you can see below, we had a lot of man glitter to clean up. The spacer came out great, but it caused the hydraulic lift ram to hit the upper control arm at full droop – D'OH!






Given that the spacer wasn't going to work, we needed to prototype a new bracket that would be eventually manufactured the same way as the original one (laser cut, bent and cadmium plated 0.184" steel). In this case, we relocated the shock absorber bolt 1.4" higher vs. the supplied steel bracket. We also moved it 0.3" inches towards the wheel so that shock cap would clear the bracket's top mounting bolt.

I drew it up, 3D printed it and installed. We determined that the shock would fit, but that the reservoir hoses would hit the upper control arm. So we bled the 150 psi nitrogen, drained the oil and disconnected the reservoirs (I didn't do this initially because it will cost $300 to have them re-filled and bled on the dyno at Penske).

We then wondered, "Are they strong enough to support the car so that we can adjust the ride height?" So I printed another bracket and gave it a whirl. They held up for several up/down cycles on the lift until I set the car down a little too hard and POW! they exploded and pieces were everywhere with one piece making it into the hallway.

Third time is the charm, right? Now that we knew the design parameters we made a more durable temporary set out of steel. We swapped the 400 pounds/inch springs with 700 pounds/inch springs, installed the shocks (still on the last adjustment thread) and lowered the car and it looked nice and low. We measured 3.75" inches. Despite increasing the spring rate by ~1.75x we were able to decrease the ride height by 1". We were able to then adjust the ride height to the desired 4.5" as which point we had 6 threads of adjustment left on the shock body. While the shocks didn't have nitrogen in them, the car will get heavier by the time that I'm finished which means that I'm in great shape with respect to ride height now.



I had previously called Penske because I anticipated that the reservoir hoses might interfere with the upper control arm. I learned that you can configure the shocks any way that you want. In particular, you can clock the collars (i.e., the orientation of the reservoir hoses) 360 degrees and you can chose from multiple NPT and banjo fittings. So, I ordered some parts from Penske to figure out optimal fitment. Superlite ships the shocks with a straight NPT fitting, so I ordered a NPT collar, a NPT hose and 45-degree and 90-degree NPT fittings. I also ordered a banjo collar and a banjo hose. Lastly I ordered a body and a body cap. The parts are beautiful, but expensive... that's $937.50 of parts! Fortunately, I can return them for no charge if they're in perfect condition.



After trying lots of permutations, the best solution was the banjo collar clocked 90 degrees so that it pointed directly towards the wheel. The banjo is then pointed towards the front of the car raised approximately 30 degrees from horizontal. This loops the hose up and over the upper control arm and sway bar as shown in the picture below. Pnut has decided that he wants to pursue a career as a hand model.



We decided that the best place to locate the reservoirs was on the wheel side of the aluminum panels that support the radiator. To accomplish this Penske will shorten the provided 20.25" hoses to 16". This will make it easy to adjust compression and nitrogen pressure. However, it will expose the reservoirs to road debris so I'll 3D print a protective bracket that contains some mesh to allow them to cool.

SHOCK LENGTH
When I took the front suspension apart I had a hard time getting the shocks out and I was unable to reinstall them. I called Superlite and spoke with Josh. He indicated that they were very difficult to install and that there were two approaches: (1) two guys and a crowbar or (2) compress the shock on the bench, use zip ties to keep it compressed, line it up and cut the ties (and I assume pray). His preferred option was the crow bar which is the approach that I used several times.

You really want another set of hands and no matter how careful you are you wind up scratching the really nice anodized finish on the shock and the aluminum on the control arm. You also put burs on the lower shock pin which requires you to sand/polish it so that it will easily slide through the mono ball. This is further complicated by the need to slip a high-misalignment washer and two grade 8 washers between the mono ball and the slot in the control arm. Once that's done, you need to insert something in one of the threaded holes to rotate the pin so that the socket head cap screws can be inserted. This isn't good for the threads. Beyond all of this steering tie rod ends up being the droop limiter which isn't good. Worst I needed to jack the lower control arm to get the tires off.

So I called Penske again. They're familiar with the SL-C and indicated that they were at a race when the Raver team approached them because they couldn't remove the front tires without separately jacking the lower control arm – the same issue that I was having. According to their notes, they determined that the shocks were 1.5" too long and the that there was a negative spring pre-load of 2.2" which was excessive.

Droop limiters in 1/8" increments


To fix this, they simply installed 1.5" of droop limiters. Penske stocks them in 1/8" increments and you can stack them to achieve the height that you need. You can also easily make your own on a lathe. As far as I can tell they're made out of Delrin.

Raver's 1.5" seems consistent with the new bracket. Recall that I moved the mounting point 1.4" up and 0.3" outward. I am now able to get the tires off without jacking the lower control arm. They rub a little bit, so I'm considering having 1/8" droop limiters installed. While I don't know the suspension's geometry, the outward movement mitigates the upward displacement somewhat. That said, our measurements are in the ball park.

I also asked Allan to measure the length of the QA1s at full droop and he got 14". I then measured mine. It was a little difficult to get an accurate measurement because the mono balls swivel and the shock body makes it hard to get close to the mono balls. So, I 3D printed a couple of tools to get a more accurate measurement as shown below. We measured 15.2", a 1.2" difference from the QA1s.





I then spoke with Allan regarding Preston's car which also has Penske shocks. He had the same issue and his solution was to cut the side profile of the leading edge of the front wheel arch. This is a fair amount of work and it's not something that he had needed to do to cars with QA1 shocks. Will spoke with Ed whose wheels also hit, but this is mitigated by his custom sway bars. So, four of four of the SL-Cs with Penske shocks that I know of have the same problem.

My conclusion is that the Penske shocks are approximately 1.2" to 1.5" too long depending on what wheels, etc you're running. The good news is that this can be easily fixed by installing droop limiters which, to my understanding, can be done in the field without the need to drain the oil.

MAX COMPRESSION AND BUMP SPRINGS
The next step was to figure out what would happen at max compression. Given that I moved the shocks up ~1.4" I assumed that the wheel would hit the body well before the shock bottomed out. We removed the spring, slid the shaft position o-ring to the top of shaft and jacked the lower control arm until there was a small gap between the top of the tire and the body. We then let the control arm down and measured the distance that the shaft position o-ring had traveled (these Penske guys think of everything).

We then determined that after the wheel lifted the body up the suspension's travel would be eventually impeded by the steering column tie rod which isn't good. This can be simply solved by using a bump stop. While a bump stop will protect the body and the steering tie rod, the 700 pound / inch spring rate will suddenly go exponential which will upset the driver if not the car. A better approach is to use a of bump stop and one or more bump springs. A bump spring acts like a really stiff main spring that's mounted on the shaft like a bump stop. This provides a more progressive and manageable experience before max compression is reached.

Some race setups...


There's a great article here on bump springs. Apparently people get paid big bucks to optimize bump springs as shown in the picture above. For my purposes, I'm going to use a bump stop and pick one bump spring that's a good bit higher than 700 pounds / inch.

REMOVING THE SHOCKS
Once the shock is unbolted you need to do the following to remove it:

  • remove one of the upper control arm bolts
  • remove one of the bolts holding the bracket for the above
  • loosen the other bolt holding the bracket for the above
  • rotate the bracket so that it no longer captures the control arm's heim joint
  • pull the control arm up so that the shock can be removed
  • You need to redo all of the above to reinstall the shock, but the real pain in holding a high-misalignment washer and two grade 8 washers on both sides of the upper control arm's heim joint when sliding the bolt through.

I no longer need to do any of the above and I no longer need a crow bar to jamb things in. Changing the shocks or springs is now a pleasure!

SUMMARY
It was really great to have pnut help out. I don't know how many times we had the shocks in and out, but we're about efficient as a F1 team. I also can't say enough about the support I got from Penske.

The Penske shocks as delivered by Superlite are approximately 1.2" to 1.5" too long which makes it hard to install the shock and requires you jack the lower control arm to remove the wheel. This can be easily and inexpensively fixed by installing droop limiters.

If you have Penskes and use a hydraulic lift and zero rate springs (IMO both are must haves for a street/track car), you won't be able to get the ride height low enough. This can be fixed via a custom bracket. In addition to fixing the ride height issue, it mitigates and potentially removes the need for droop limiters. Furthermore, it means that you can remove/reinstall the shock without touching the upper control arm. However, you must clock and potentially replace your shock collars. If this is the route that you want to go, make sure that the shocks are configured the way you want them before you order or you'll going to be dropping ~$700 before you even get to the custom bracket.

Key measurements:

  • Lift puck: 3.12" (confirm with mic)
  • Zero rate spring and spring divider: 0.8"
  • Penske Full Droop: 15.2"
  • Penske Max compression: 11.05"
  • QA1 Full droop: 14"

Key changes:
  • New bracket moved upper shock bolt 1.4" up and 0.3" towards the wheel
  • NPT collars replaced with banjo collars and clocked so that they point directly towards the wheel
  • Reservoir hoses shortened to 16"
  • Spring rate increased to 700 pounds / inch
  • Added a 1/8" droop limiter (pending)
  • Bump stops and bump springs installed (pending)

The results were:

  • Ride height is correct with room to adjust either way
  • Wheels easy to remove (no need to jack lower control arm)
  • Shock easy to remove and reinstall (no crowbar and no need to touch upper control arm out of way)
  • Max compression is properly managed

Next steps are to ship the shocks back to Penske and to have the final version of the bracket made.
 
Very nice mod to move the upper shock mounting point.

As an FYI - I run the QA1 front shocks with the hydraulic lift and my front wheels are super close to touching the forward edge of the wheel arch, ditto my rears (actually think my rears did make hard contact).

I think the SLC just has too much wheel arch wrap.

I'm not sure what kind of recontouring Allan did on Preston's car but I did a recontour of my contact points as well, using Howard's cut-a-pie method. It's a relatively simple mod - minor glass work and it maintains a fairly OEM look by slightly straightening out the curve.

I think it's a mandatory mod on SLCs to be frank. Every time I tried removing my wheels with the bodywork on I either needed a hydraulic jack to lift the suspension or play "Operation" with my 50# wheels and hope I don't scrape up the wheel or my calipers while inching the wheel off the studs.

Here's a pic of my front wheel well, with the forward edge shifted forward 0.5":

 

Scott

Member
Lifetime Supporter
Cam, the modification that Allan did to Preston's car was similar to what you did. However, he has installed the hydraulic lift on pretty much every car with QA1s and none of them needed that modification. Preston's needed it because the Penskes are too long. Perhaps your body is slid back a little bit. In any event, you got it to work so it doesn't matter.

Frank, my zero-rate spring and divider were a little taller that I expected when I compressed them in a vice. However, the difference between your stack and my stack wouldn't likely have fixed the ride height issue. Keep in mind the the shock is on an angle so a quarter of an inch on the shock body equates less in ride height.
 

Scott

Member
Lifetime Supporter
The fuel-level sender is configured to integrate seamlessly with the Koso gauge. Specifically, it applies signal damping to prevent the reading from bouncing when the fuel sloshes in the tank and it maps the unique shape of the SL-Cs fuel tank to eight discrete restive/ohm-based values. The Koso then maps those eight values to the ten display bars on the gauge.

The MoTeC system can provide a more accurate and granular solution so I contacted from Centroid Products to purchase a sender with an analogue 0-5V output an no tank shape compensation. They informed me that they would re-calibrate the sender at no cost, a nice surprise.

To calibrate the system, I will put some fuel in the tank and run the low pressure pump until it stops pumping being careful to not damage the pump by running dry for very long. I will take a voltage reading which will be the "empty" setting. I will then add a gallon of fuel and measure the voltage. I will repeat this until the voltage doesn't change at which point the tank is "full" as far as the sender is concerned. Of course, more fuel could be added at the very top of the tank and the fuel filler, but that's not a concern. Given that the tank holds 19.2 gallons, I will have 20 data points with all but one being approximately 5% apart. I could also go with half gallon increments, but I doubt that I'll have that much patience.

This data will then be entered into a custom MoTeC map.
 

Scott

Member
Lifetime Supporter
I upgraded the hydraulic fluid reservoir from 3/8" push-on barbs to -6 AN fittings and hose. I spent a fair amount of time looking for reservoir with AN fittings to no avail. So I decided to modify the supplied one. I purchased a -6 AN Fuel Cell Bulkhead Adapter Fitting from Vibrant Performance which came with a hex nut and two PFTE washers; not bad for $9.99. I removed the barb with a Dremel cut-off wheel, filed the bottom flat and then carefully enlarged the hole. The nut was too big to fit so I had to grind it round and press fit it.


Note that the hex nut has already been ground round and that only one of the PTFE washers was used.



Tight fit



Final result
 

Scott

Member
Lifetime Supporter
Spring Soccer in Boston...

So it's the first week of spring and here are a few photos from my son's soccer game... yeah, all that white stuff is snow. You'll note that the boys in blue have hats and leg coverings while my son's team, the boys in black, have a hardcore coach who insisted on bare legs and heads. Fortunately it was OK for the parents to remain bundled up.

In any event, it should be apparent why I'm building a SL-C rather than a Le Mans.



 

Scott

Member
Lifetime Supporter
I upgraded to Tilton master cylinders and reservoirs which have -4 AN fittings. I didn't want to use stainless PFTE because it's heavy and installation goes pretty much like this; I get stabbed, I bleed, and then I swear like a sailor.

I found some really nice aramid-braided PFTE hose from Goodridge which can be bought from Pegasus Auto Racing. It's not cheap, but it has lots of benefits; it's:
  • Extremely easy to install; no special tools, no bleeding, and no profanity
  • Lightweight; about 43% lighter than stainless
  • Very flexible; 2-3 times smaller bend radius than standard hose
  • Bullet proof; the aramid sheathing is ballistic-rated for body armor, I have no plan to test that!
  • High-pressure rating; the -4 AN hose is rated to 1,320 psi
In the picture below the foremost object is the PFTE liner with the sheathing removed. This exposes the convoluted outside diameter which is what provides the super-flexible bend radius. Note that the interior diameter is completely smooth. The middle objects are the hose ends. The silver part simply slides over the hose and the black part threads into the hose. The top object shows how flexible the hose is.




The installation instructions suggest that you wrap the point to be cut with low-stick painter's tape. A Dremel cut-off wheel makes a clean cut, but I found that removing the tape caused fraying which was a hassle to stuff into the silver collar. I began using a couple wraps of Teflon tape before the painter's tape which significantly reduced the amount of fraying. I latter figured out that I could just wrap the hose exactly one time with a 1/4" wide piece of painter's tape. After removing any small frays with a high-quality micro shear (I like Xuron), I was able to simply twist the silver collar on without removing the 1/8" of tape (the 1/4" tape was cut in the middle). After putting a couple drops of light oil on the threads, you twist the end on until it's tight and then you spin the silver collar a couple of times... that's it.


To keep the hoses in place I designed and 3D printed a custom bracket.








 

Dan Carter

Member
GT40s Supporter
Impressive Scott. Could you please stop with all the great ideas AFTER i have complete that phase...LOL. I fully understand the bleeding option as I used the steel braid and actually invented a few new swear words not heard by man before.


Great work, wish I had thought of it before.
 
Scott.... Hoosier Performance is no longer in business. You got your front hubs from them just in time. They closed their doors in January of this year.
 

Scott

Member
Lifetime Supporter
Ron and Dan, thanks for the encouragement.


Ken, that's too bad that they went under. I'm glad that I got a set because the machined pieces are really nice and they seem indestructible. Good thing that they're completely rebuildable with third-party bearings!
 

Scott

Member
Lifetime Supporter
I need a custom pulley machined that will mount to the front of my ATI damper to run my alternator and compressor. It needs to have 6 grooves, three mounting holes and a concentric locating ring. The one shown below is pretty close. My understanding is that the mounting holes are the same as a big-block Chevy and that the locating ring is similar. I have Googled "custom billet pulley" etc. and what I've found are lots of companies that have custom productized pulleys, but none that specialize in one-off pulleys. I assume that there's a better starting point than going to a CNC-shop and having them figure out the groove specs from scratch. Anyone have any ideas?
 

Attachments

Scott

Member
Lifetime Supporter
Ken,


Thanks for the pointer. That form is for a custom damper. All I need is a custom pulley to bolt to the front of my existing damper and they don't make those.
 
Top