Part 1
Part 2
Part 3

The upper control arm mounts, which also ended up including the front shock mounts were a real challenge.  I made some mounts, found out they were wrong, and had to scrap them.  I cut out the factory strut towers for clearance and found out that they might have been useful for mounting the shocks.  I bought shocks only to find that they were too long to fit in the front of the car.

The previous posts have shown the process of putting together the mounts.  Unfortunately, the shock mounts were made during the camera failure time, so I have no pictures.  They are a combination of simple box tubing and an off-the-shelf circle track mount that was at the wrong angle to work without modification.

The parts were then welded up and given a coat of appliance epoxy.  Here is the result.

I would like to add some bracing to support the shock mount since it is kind of floating in space, but that will be later on.  This should suffice for now.

That leaves on the strut rod mount left, and the front suspension is done.

Just a quick update today.  The camera is fixed and on its way back to me so I should be able to get back to adding content more regularly soon.

In the last entry on the upper control arm mounts, I talked about adding sleeves to the box tubing so that it wouldn’t be crushed were the mounting bolts over-tightened.  I found some cheap steel sleeves in one of the drawers at the local hardware store, and luckily they were just the right height.

They will go into the end of the box tubing and be epoxied in place.  I am going to use epoxy simply because it has a drying time that allows me to make sure that the sleeve is properly aligned with the hole in the bracket.  Welding offers no such luxury.

Up next, the work done to add the shock mount.  Unfortunately because of the camera issue, it will be only before and after photos, but the process is simple to explain, even if it wasn’t to accomplish.

In a past blog, I stated that I had made a mistake with the upper control arm mounts.  What I did was make the inner vertical dimension to the specifications of the outer dimension.  In other words, instead of having the upper control arms in their proper locations, they were just an inboard copy of the spindle mounts.  Oops.

It wasn’t the worst thing in the world, as it actually allowed me to make a shorter mount.  The original mounts were rather tall and likely to flex, so some good came out of the bad.

Beyond the issue of strength, because they would be shorter, I made the new mounts so that the bolts that attach it the car will go all the way through.  This eliminates the need to fiddle around getting a nut and a wrench in the box.  I found some sleeves that will add strength to the holes after I took these pictures, so look for details on these in an upcoming post.

That is the base.  Simple 1″ x 2″, 0.120″ wall tubing.  0.120″ wall is probably overkill, but easier to weld to the 3/8″ plate that the upper control arms bolt to.

The mount is an off-the-shelf item ordered from one of the many circle track supply houses.  I labeled each one so as to make sure I don’t end up with two of the same rather than one for each side.  I use a Sharpie silver marker and it works great on raw steel.

Also visible on the bottom of the mounting plate is a score line which marks the center point between the arm mounting holes.  Since the holes in the base have already been drilled, everything has to be lined up properly or the caster won’t be the same side-to-side.  Of course there will be other adjustments to make sure it is equal, but I’d rather get it as close to right as possible.

I beveled the bottom of the mount so I could make a better weld.  It took only a minute with the grinder and is just a good practice to follow when welding thick metal.  With the bevel finished, I mocked up the parts one last time.

Here you can see the score line on the base that matches to the mark on the plate.

Once bitten, twice shy.  I remeasured everything, held the parts together with welding magnets and actually put them in the car so I knew they would be right!  Next step was to tack weld the parts in place.

The line and NO WELD sign is there for when I add braces.  Anything above the line will interfere with the mounting of the arm.  I tacked both sides to try and minimize distortion.

I did not finish weld the mounts yet because I have to add the front shock mount.  I want to see if I can use the mounting base for the upper a-arm mounting plate for that purpose as well.  While it shouldn’t impact the way in which I weld these parts together, it is easier for me to do all the welding at once.

Once I know the shock mounting solution, I will finish the mounts and post about it.

While I do love using take-off circle track parts because of their incredible value, they do present some unique challenges.  When using OE parts, it is generally safe to assume that things will fit.  After all, they did when the car was delivered.  If you start mixing and matching OE components from different vehicles or model years, you can run into problems, but even then you have an OE setup that is a known working configuration and can be used as a target point.

No such target exists with race parts unless you buy complete assemblies, which are hard to find used and much more difficult to put together.  There is no factory service manual and exploded parts diagrams with part number call-outs for most of the race stuff.  Certainly not for add-on parts like brakes to hubs.

I found this out with brake rotor to spindle clearances.  The spindle uses a bolt-on steering arm assembly that also incorporates the lower ball joint taper.  There are two settings, high and low, and was previously covered in the front suspension design blogs, the lower hole is what I needed to use.  The problem was that the rotor to hat mounting bolts would hit the heads of the bolts that attach the steering arm.

I believe that the reason this happens is because I am using the 8 on 7″ bolt pattern rotor.  This is the most common, and therefore cheapest one on the market.  The 8 on 7″ pattern is used by Wide-5 hubs (about which I have written before) that don’t use a hat, but have an integral rotor mounting system.  More importantly, it is also the bolt pattern used by my rear hubs which also have an integral mounting system, so in order to be have just one rotor configuration that fits all four corners, I would need to stay with this design.

When I got the spindles, they came with some bolts that had been cut in a way I didn’t like.  So I ordered some replacements.

Little did I know at the time that I would find out why these bolts had been trimmed in this way.

The new bolt, front, is from ARP and made especially for this purpose.  A ¾” fine thread bolt could probably hold up a bridge, and when it comes from ARP, it is likely going to be able to do anything you ask of it without complaint.  The head of the ARP bolt is smaller than the originals, but not quite enough.  I still had to trim them.

Here you can see the problem.  The rotor mounting bolts would touch the spindle bolt head were it not ground down.  I forgot to take pictures before the bolts were ground down; probably because I was so frustrated at the time.  Even with the heads trimmed, you can likely see where the clearance problem would have occurred.  I once wondered why there would be both 8 on 7″ and 8 on 7.625″ mounting patterns for rotors.  Answer above.

So, I took my nice, new bolt and did to them exactly what the previous owner of the spindles had done to his steering arm mounting bolts: I ground the heads down.

As you might imagine, these ARP bolts aren’t the easiest things in the world to cut.  Each pair of bolts consumed a 4″ cutoff wheel.  I wouldn’t even think of using a saw.

All’s well that ends well.  There are no clearance problem now and I have new bolts holding things together.  I am sure the previous owner of the spindles would chuckle were he to read this as he likely went through the same thing and for the same reasons.  Wish I’d have known.

In my recent entry, Action – Front Suspension – Upper Mounts (Part I), I stated:

First order of business is to make room for the new upper arms, a process detailed in a previous entry.  Once you have the space, it is time to figure out what to put in it.

Unfortunately, I am guilty of a carriage-equine sequencing error.  Before you think about the upper control arm mounts, you have to know which arms you are going to use.  While that is part of the planning phase rather than the action phase, I realized that I may have glossed over the details of the upper control arms and in so doing, left out information vital to understanding the process of making the mounts.  Since I made an entry for the planning involved in the lower control arms, I am doing one for the uppers as well.

Since there is no factory upper control arm, there is a greater degree of flexibility in choosing parts compared to the lowers.  I have written previously about the design aspects of a custom SLA suspension, and while the computer generated design might be the best one on paper, if the pivot points are trying to share space with the cylinder heads or the parts the program says are needed don’t exist or don’t fit your budget, changes must be made.

For me, the dimensions of the lower arms was dictated by the design of the Mazda front subframe.  The upper arms are proportional to the lowers, as determined by the program.  Since the lowers were short, and the uppers are always shorter than the lowers, I had not problems with the upper control arm mounts needing to be somewhere they could not.

Picking the front control arms was actually very easy.  I had heard nightmare stories from muscle car guys who were swapping the stock arms in their rides over to tubular arms and paying over $200 for each one.  My arms cost $8 each, ball joint included, and I will put their quality up against any on the market.

What are these miracle arms?  Take-off NASCAR parts, of course!  I bought them from Raceproven Parts and they are absolutely the bargain of the century.  At the time of writing, there are 163 arms listed for sale, $6 or less for most without a ball joint, $8 for ones with a ball joint.  The ball joints included are top notch items.  Smooth and free moving.  I bought two pair; one set approximately 7/16″ shorter than the other, so that I can play with roll centers and camber gain curves should I ever feel the urge.

As you can see, the plane of the upper ball joint mount is not the same as the arms.  This is because the upper control arm slopes downward from the ball joint to the chassis mount.  By incorporating a 10° angle into the mount, the ball joint is closer to being in the center of its travel when at ride height.  This means maximum articulation before binding.

Also not the grease fittings for the ball joint and the inner pivots.

Above are visible the mounting holes used to attach the upper control arm to the chassis.  The distance between them is 6″; a standard distance that allows ordering many different off-the-shelf mounts from countless suppliers.  In the arm above, they are 1/2″ holes, but arms are also available with slots.  I suggest getting slotted arms if possible.

The slots allow for caster adjustment with the upper control arm.  Caster will also be adjustable through the strut rod, but it cannot hurt to have more options.  The arms are sold with a confusing listing of offset which is the distance that the center of the ball joint is from a line drawn perpendicular to mounting bar and through the center-point between the two holes.  The confusing part is that what is positive on the driver side becomes negative on the passenger side, and for sale listing follow no pattern I could discern regard which was which.

Which leads to what is perhaps the most difficult part about using NASCAR takeoff upper control arms – finding a perfect pair.  When only turning left, suspension geometry for each side of the car can be different.  This means that arms are rarely made in equal pairs and finding two that match can be difficult.  This is another reason to search out slotted mounting holes.  Two arms of the same length with slightly different caster offsets can be used with no ill effects I can see.

The arms come unpainted, but I am thinking of shooting them with a coat or two of Rustoleum Appliance Epoxy to keep them looking fresh.  At $8 each, the paint might end up adding a significant percent increase in cost!

Going from the factory strut front suspension to the superior SLA appears to be one of the most compelling parts of the V8Mongrel project.  Lots of Google searches regarding this topic bring people here, and as a result, I am going to try and write as much detail about the process as I can.

Please read the Front Suspension archive to make sure you are up to speed with what has been going on up to this point.  I am going to try to keep these next posts on how to do this, not why.

If you are crazy enough to take a strut car an SLA design, recognize that you are likely going to have to make some changes that can never be undone.  First order of business is to make room for the new upper arms, a process detailed in a previous entry.  Once you have the space, it is time to figure out what to put in it.

In the case of the SA22C chassis, Mazda provides a little help.  There are two long bolts that run down through the front frame rails and hold in the front subframe.  They are spot welded in at the top, but are not too difficult to remove.  The result is a hole sized from an M10, which in the USA, means that you need to run a 15/32″ drill bit through it a couple of times in order to get a readily available 7/16″ bolt to fit.

The bolts that go through the holes provide an easy way to attach an upper mount.  The reason I chose this method is twofold.  First, Mazda has sleeves in these holes so the frame rail will not collapse if the bolts are overtightened.  Second, the metal that makes up the frame rails is not particularly thick, making it more difficult to weld to and perhaps subject to tear out under load.

Making the upper mounts is quite easy if you know where to look.  They are readily available from suppliers such as A&A Manufacturing, and designed to be adapted to many different cars.  A 1979 RX-7?  No, probably not what they  had in mind, but the part is still easily configured to a variety of applications.

That is Vincent Keene welding up the A&A mount to a plate through which the bolts will go.  After the plate, I needed to raise the mount in order to get the desired geometry.  Here is some foreshadowing – get this right the first time.  I used some 2″ x 3″ rectangular tubing for strength.  Here is the finished mount and after I added some bracing and paint.

You can see how the bolt holes are not at the same height.  This is for anti-dive under braking.  By angling the upper control arm from horizontal, with the rear lower than the front, the braking force is used to lift the front of the car, decreasing the amount the nose drops during heavy deceleration.  Here are some more pictures, this time with the mounts in the car.

Looking carefully, you can see that the designed allowed for the control arm to be attached both inboard and outboard of the mounting plate.  Adjustability and flexibility are paramount in this type of project in my opinion.

Once the assembly was complete, mounting it to the car meant running the bolts up from the bottom, through the subframe and frame rails, then fishing a nut into the square tubing so that it can be tightened down.  Not particularly elegant, but it worked.

Up next, what I did wrong, and how I am going to fix it.

In order to make room for the planned short and long arm (SLA) front suspension, certain vestiges of the strut configuration had to be eliminated. Number one on the list was the strut tower itself.

The removal of the strut tower, much like the removal of many components, was not a decision easily made. Keeping the strut tower would allow me to have a hedge against the SLA project failing to work as I intended. But in spite of my attempts to somehow keep the tower by bolting the mount for the inner control arm to it, the compromises were too great. The strut tower had to go!

The first step was to mark the metal.

I planned to cut the metal back to the level of the inner fender. This isn’t necessary for A-arm clearance and might be overkill for shock and spring clearance too. However, in the interest of keeping me in the garage and out of the hospital, I decided that trimming the edge back away from the suspension components that I would be spending a lot of time working with was a solid plan.

Then I would follow the seam down to the bottom and around the other side. I wanted to bow out near the bottom to allow for the arm to pass through more easily.

Unfortunately this has left me needing to figure out how to cover the holes that will result from removing the strut towers. I haven’t quite figured that out yet, but perhaps a combination of fiberglass and sheet metal. Held in with some quarter turn fasteners for easy service.

Kendt Eklund volunteered quickly to do the destruction work. I was a little concerned about his eagerness to destroy stuff, but I found out later that it wasn’t because he was using it as training to be a mass murderer, but to up his macho image with the ladies. Thanks for the help Kendt.

Here is how the strut towers were removed.

First Kendt drilled out the spot welds. Those that were accessible from inside the wheel well were drilled that way because there was less interference.

In those areas where the spot welds were hard to get to or where the cut line needed to follow a path different from the welds, Kendt employed the reciprocating saw.

As you can see, there was little danger of the blade hitting anything if it wasn’t pushed all the way through.

One all the welds were drilled out and the saw used to everywhere else, Kendt simply hammered the piece away from the car in a nice solid chunk like the one shown in my previous entry.

The process was done on both sides and now I am the proud owner of two big gapping holes!

Beside the safety aspect mentioned earlier, I left the heavy metal lip on the top of the strut tower for two reasons. First, it might serve as a good place to mount something one day. Second, it might play some structural roll, although I doubt it is relevant without the strut acting upon the area.

Now with space available, it was time to work on the part to occupy them.

Yesterday, (May 3, 2008) I was fortunate enough to have a number of friends come to my house and help with the Mongrel. Many things were accomplished, and I will take the time to detail them all later. For now, two pictures are sufficient to show that the SLA conversion has reached an important milestone – the point of no return.

The RX-7 uses a conventional, production style front sway bar that connections to the chassis next to the tension rod mounts and then to the lower control arm. There is absolutely nothing wrong with this setup, and ones that are similar in execution are used on production based race cars the world over.

This is the stock bar. My son is helping me measure it.

However, since I had determined that the factory front suspension design would be completely eliminated, I felt like designing around an OE sway bar might become more of a liability than it was worth. So, like the SLA project, I cracked open the circle track parts catalogs and started cruising eBay for bargains.

Before I go on to the planning of the front sway bar project, I will share the dimensions of the stock unit for those that need them. Also, the replacement will need to be approximately the same since I am retaining the front strut rod (tension rod) style setup and (preliminarily) plan to locate the new sway bar in the same general area and fashion.

Inner width:

Outer width: (picture of what this is measuring is the one with my son in it, above)

Arm Length:

Arm Offset:

Next up, the plan for the replacement.

Before continuing on the details of my own front suspension project, I’d like to share a couple of links that are likely to be helpful to most people undertaking the same sort of project as me, yet they may not be easily found due to their location on a non-racing site.

“Chopsaw Notching, Tube Strength & Link Construction”
“A Joint for Any Occasion”

It is my opinion that these superbly written articles are worth reading multiple times regardless of the fact that they are tailored to the builders of off-road vehicles. Proper construction techniques and part selection are the same whether the vehicle will be bumping a rounded, one inch curb at triple digit speeds or traversing a jagged, twenty-four inch rock at walking speeds. Thanks to the author, BillaVista, for making the effort to publish such detailed information.

While Dan Durusky was in my garage, with the help of the diagrams he had sketched, we began to manipulate the variables in the suspension. What made the process difficult was the number of variables that we had to consider.

1. The height (distance from the ground) of the chassis mount for the lower control arm.
2. The height of the chassis mount for the upper control arm.
3. The depth (distance from the centerline of the car) of the chassis mount for the upper control arm.
4. Two different lengths of control arms.
5. ±¾” length for the lower control arm.
6. Top and bottom mounting holes for the lower balljoint taper on the spindle.

I am sure that someone could do the math to figure out the number of permutations available just considering ¼” increments, but suffice it to say, we could have been plugging in scenarios for days. In order to forestall this possibility, Dan and I fixed certain variables on the list . In the same order as above:

1. Low control arm mount height set at 8 inches, making the lower arm flat.
2. Upper control arm mount height left variable.
3. Upper control arm mount depth set by the angle of the upper control arm, baseline 10°, as that is what is built into the arms I purchased.
4. Use the longer upper arm and change to the shorter upper arm as a single variable change.
5. Fix lower control arm length at 13 inches.
6. Set lower balljoint to spindle mount on the lower hole.

With the variables thus limited, we were able to define our goals and observe patterns in the changes we would make.

Goal A – Roll Center – A roll center height between 2 and 4 inches from the ground.
Goal B – Camber Gain – Approximately 1° of camber gain per inch of travel.
Goal C – Swing Arm – Maximize front view virtual swing arm length.

As we manipulated the data, we found that the lower the upper mount, A was raised, B was increased and C was shortened. Predictably, raising the mount reversed the trends. Raising the inner mounts of the lower control arm (originally not a variable we wanted to manipulate) provided the same results and lower the upper mounts.

Here is a screen shot of Performance Trends software of the first setup.

As you can see, we got a roll center height of 2.4 (A) and 1.08° of camber gain (B). The front view virtual swing arm length (C) is something I still need to evaluate. For anyone who has the Performance Trends software, here is the file of this setup. You will need to right click and then ’save as’. If the file type saves as .htm, don’t worry, the Roll Calculator software still opens it without problem.

Dan is still plugging away at the different variables to see if he can come up with anything better. I am measuring the inner fenders since they will need to be removed and making sure I have all the parts that I need. We aren’t done, but we are much closer than we were a week ago.