Part I
Part II

So what was the solution to my physics dilemma?  No, I didn’t create anti-matter and therefore allow my rear axle and fuel tank to occupy the same space.  I bought a new tank.

It is basically an aluminum box with a couple of hose end bungs on the top of it, a pickup tube and a sending unit.  The large black circle is the filler inlet.  Unfortunately, the tank couldn’t be located in a way that allowed me to continue to use the factory fuel door and filler.  Hopefully, I can figure out a cost effective solution that doesn’t require me to open the rear hatch at every fill up.

Just like the mini fuel cell I wrote about in Part II, this unit is also made by RCI.  It is a 12 gallon tank.  It is not a fuel cell, because I couldn’t find one off-the-shelf that fit and custom cells cost a lot more than I can afford.  I bought this tank off eBay, and while I usually prefer to buy from established businesses when looking for new parts, in this case, I put in a bid for half the retail value, and won it as the only bidder.  I really didn’t think that was going to happen, but I was happy with the savings.

As you can see in the picture above, there are four tabs, one on each lower corner, that are used to mount it.  The problem I faced was that while the tank fit nicely between the rear frame rails, the tabs would not.  Thus, dropping the tank down onto a fabricated shelf was not an option.  It would have to go in from the bottom and bolt to a mount.

First step was getting rid of all the extraneous metal that was in the way.

We started with this:

Got cutting:

And had this to show for it:

We actually went on to cut out all the sheet metal in the floor, making the round hole square.  Much like the Shock Tower Removal Project, Kendt Eklund was a metal cutting machine.

It is at this point that I once again have to recognize the skill, talents and willingness to help of Vincent Keene.  He offered to come over and work on the Mongrel, as he has many times before, and I was pretty much able to give VK an overview of what I was thinking, and set him free.  He took care of all the details.  This is mostly his handy work, although there were other people who helped out too.

This is the driver side tank mount.  It is a simple box frame that is welded onto the factory frame rails.

This is the passenger side.  From this angle you can see the holes in the bottom.  Inside of each is welded a nut so that the bolt can put in more easily than with a nut and bolt combination.

Unfortunately, the final product is very difficult to photograph as it is tucked in the car very well.  Here are the best shots I have.

The one detail I should mention is that between the mounts and tank tabs where the bolts go through are some rubber insulator pads.  The same is true between the tank and the actual mounts.  This is to ensure that the inevitable vibration won’t result in steel rubbing aluminum.  We know who wins those battles.

So in the end, I went from having an old tank, to an old tank that looked like new, to a new tank, with new mounts that required removal of a lot of old metal.  Oh, and it required me to redo the fuel pump setup too.  That story will be in an upcoming entry.

In a previous entry I detailed the process of making my old gas tank look new.  I was pretty pleased with the results and didn’t give it a thought for some time.  Unfortunately, that didn’t last.

I start investigating if the SA22C was subject to fuel starvation issues when run on a road course.  The results were mixed.  Some said yes, some said no.  I decided that I could just make sure the tank was always close to full while on track, and deal with it if it ever became an issue.

Next I looked at the fuel pickup.  A 5/16″ fuel line was bigger than the inside diameter of the tube that led from in the tank.  5/16″ is smaller than AN6, which is approximately 3/8″ or 6/16″ as AN sizes are just sixteenths of an inch.  AN6 was the size I was told I needed to feed the engine.  I was told that 5/16″ fuel line choked flow at about 250 hp.  I concluded that since I was probably going to need two to three times the fuel flow of a rotary engine, that the OE tank could well cause a problem in this regard.  After all, Mazda had no need to consider fuel flow at the power level I am going to have.

Modifying a fuel tank is not something I felt comfortable doing.  However, I didn’t want to go to the trouble of buying a fuel cell or another tank.  So I look at other options and came up with what I felt like was a good one.

What I realized was that my fuel flow problem would only arise during full throttle.  While this is going to be a regular occurrence, I am not running something like the Silver State Classic where it is constant for long periods of time.  On a road course, even one that uses part of an oval, sooner or later a corner appears and the gas pedal is lifted.  So, I reasoned, all I had to do was make sure that I had sufficient fuel flow for the longest anticipated period of full throttle activity, and a margin for safety.

Never one to let an opportunity for improvement pass, I also thought that I could solve any potential fuel starvation problem due to slosh at the same time.  The two-for-one solution I came up with was a surge tank.

I have no pictures of what I did, since it was ultimately all removed, but here is a link to the site I used to plan what I did.  They are using a surge tank to later pressure, while I used it to alter volume, but the layout is the same.  I didn’t build my own tank, but the plumbing used the same components.  I used a one gallon fuel cell by RCI and ordered from Summit Racing.

I used a lower capacity to pump to feed from the OE tank to the mini fuel cell that acted as my surge tank, then a main pump to feed the engine.  The surge tank had to have a return to the main tank for idling.  However, since the surge tank was one gallon in capacity, I would have to be going full throttle for at least five miles in order empty it were the feed pump to fail completely.  I felt like that was a good compromise.

Unfortunately, all the time, money and planning went down the toilet when the old problem of two things trying to occupy the same space reared its ugly head.

Note that the gas tank has been removed and the axle doesn’t even clear the spare tire well which sits above it.  All that work for nothing.

Below I have included some of the other pictures I took when assessing the clearance problems.  The show some of the surge tank setup.  If you have any questions, post them in the comments and I’ll be happy to answer them.

This is a story of how even simple things often aren’t.

Getting fuel to an engine really shouldn’t be that hard.  You need:

• a container for the fuel
• a pump
• and lines to take the fuel from the tank to the engine, via the pump

Simple, right?  And of all the things on that list, the tank should be the easiest.  The V8Mongrel came with a tank, and while I have previously documented the interesting pump setup, changing pumps really isn’t so complex to worry me.

In light of this, one of the first things I did when picking up the car was restore the stock tank.  This ended up being a complete waste of time, but I will detail the process in case anyone else is thinking of doing the same thing.

Here is how the tank looked when I got it.

Pretty unimpressive.  Yet, it didn’t leak, it wasn’t rusted, and I knew it fit.  Certainly better than the fuel pumps.  So step one was to clean it up.

Note the tape over the holes.  This is important.  Getting dirt in the tank is a lot easier than getting it out.  The cleaning was accomplished through a combination of engine cleaner (Foamy Engine Brite being my favorite), a putty knife and a wire wheel on the end of an electric drill.  Nothing fancy.

Cleaning the inside was more of a challenge.

While hard to see, hopefully you notice the discoloration visible on the left side of the hole.  That is varnish and it occurs when gasoline sits too long.  In fairness, it looks like this tank was very well drained by a previous owner because the varnish was only on the lowest spots.

Based on internet research, I used a gallon of mineral spirits poured into the tank to break up the varnish.  I would pick up the tank and shake it around on occasion too.  How much shaking it accomplished, I cannot say, but it made me feel useful.

I also cannot say if the mineral spirit treatment did a lot.  The stuff came out a sickly yellowish brown, so I know it cleaned up something, yet the inside really didn’t look much different.  In fact, at this point I am not even sure if that photo above is a before picture or one taken after the mineral spirit soak-n-shake routine.  Either way, I felt like I had a good clean gas tank, at this point, at least on the inside.

Up next was fixing the outside.  I used a rag (it is blue, so hard to see in the photo) soaked in more mineral spirits to clean off any residual dirt or cleaner.

The primer was nothing fancy.  Just some automotive spot primer.  I don’t know if I should have used an etching primer or something else, so I just used what I had.  It only took a couple of fog coats to cover up what was below.  The top coat was even easier.

For this, I used some chassis black I had picked up from parts unknown.  It said it was designed for use under a car.  While this is the top of the tank, it sounded good to me.  Once the paint was dry, it was on to the bottom.

On the bottom of the tank, I didn’t want to tank out the wire wheel because it had a nice textured rubber coating on it.  I figured that the coating would be hard to duplicate, so I should stick with cleaning processes that didn’t result in any coating removal.

The first step involved Simple Green and lots of water.  Step two involved the parts brush seen in the photos and then a pot scrubbing brush I liberated from the kitchen sink.

I felt that it looked pretty good, so I then gave it a nice top-coat with the same chassis black.

I was pretty happy with the results, so I moved on to the straps.

The same wire wheel cleaning procedure was followed by some primer.

I decide to add a little flare at this point.  I really liked the look and am sad that the car will never get to showcase this work.  Here is the ‘final’ product.

Unfortunately, as I have indicated, this was going to end up being all for nothing.  More important parts were going to need that space.  Next time, I will detail where things started to change and why.

This is a picture of the current setup on the car and it is clearly not stock. There appears to be two pumps that are supplied by one feed line with a simple T fitting splitting the flow to both pumps. After that, the flow is merged via another T fitting and then hooks to the stock line that runs under the car. All in all, a rather puzzling setup.

Speaking with Bruno, the car was at one time equipped with a 13B rotary in place of the original 12A. Unless it was an extremely hot 13B, the general wisdom surrounding the stock pump is that it will supply 300 hp and therefore should have been more than adequate. The stock pump is for a carb, so if the 13B was injected, it would most likely require a different pump. However, it would also need new lines for the extra pressure, and those attached to this pump don’t appear to be anything other than stock. I will report more in the filters and lines section.

Running pumps in this fashion, in parallel, is not something I have ever seen before. I do not know what advantage it has over running them end to end, in series. I think Corky Bell in Maximum Boost and Supercharged, talks about it in some depth so I will have to read those books.

Since the car has no battery or even semblance of a working electrical system, it is hard to determine what is going on. It could be that the pumps can be selected via a switch; essentially a redundancy for reliability. Since my goals put me above the recommended power level for the stock RX-7 pump, this is in some ways little more than a curiosity as I have already bought a used Holley Black fuel pump in a package deal with my carb.

Pic courtesy of Holley.com – mine is dirty as it was bought used.