Virtual Chassis Construction

Over the winter months, I started teaching myself how to use Autodesk Fusion 360.  The main advantage of this software is that it does a free educational licence version - and being that my son is still at secondary school, I signed him up on my behalf!

I originally just wanted to be able to design a couple of brackets for the Cobra build - but things got a bit out of hand...

I had started to design a bracket to hold the brake pipes and clutch pipe on the chassis leg with some unions to allow the pipes up to the master cylinder to be added later once the body is on (there will be a separate post on this).  

I wanted to see how this bracket would sit on the chassis rail and to check the mounting position.  So I started modelling a portion of the front of the AK chassis...

Then I thought I could use this portion of modelled chassis to check the brake and clutch pipe routing - part of the tubular bracing on the AK Gen III chassis creates a very narrow path for the front brake and clutch lines.  This eventually turned into deciding to model all the brake and clutch pipe routings so I could decide where to put the pipe clips to ensure a maximum spacing of 300mm.

Well over many long winter evenings I managed to model up practically all of the AK chassis - and then spent a few more evenings remodelling some bits of it when it was apparent some of my measurements were not correct.

It's not fully complete - I still have to finish off the tubular bracing parts - but I've included a couple of pictures below.  I've added a sneak preview of part of the chassis with brake pipes and clips modelled - I'll cover this more fully in one of my posts on the fixing of the brake lines.  There's also a part-finished render from Fusion 360 of this section of the chassis (the processing power of my Mac isn't up to getting the full render done!) - let's see if it looks as good when I actually finish it!

Brake Lines Part 1

The routing of the brake lines is relatively straightforward.  The AK Build Manual shows the positioning of the pipe runs quite clearly and most other builders have had no problems with this layout and with fixing the pipes using P-clips or the plastic push-in type clips.

So clearly this is the way to go and it should all be simple, yes?

Well...

Clearly, for me, that would all be too simple...and I lay the blame squarely at the doors of the guys from Bad Obsession Motorsport (BOM).  I have been following their Project Binky build on YouTube for some time; there are several episodes charting their build of a Rally-spec Mini based on shoehorning the running gear from a Toyota Celica GT4 into the confines of the Mini.  These guys are evil geniuses when it comes to bespoke bracketry (of which their project required many) and when they imposed a "no P-clip" policy on the build and fabricated their own brake/fuel line clips, I started to get thinking.

The AK chassis is a thing of beauty.

 OK, maybe I'm getting carried away. but the chassis rails are all box section and fully sealed.  So why would I want to drill loads of holes into the chassis providing potential entry points for water and the risk of the dreaded metal rot?  I could inject Waxoyl into the box sections to mitigate this but surely it would be better to avoid the holes in the first place.

The BOM solution consisted of a steel plate, a couple of rivnuts and some 12mm thick nylon sheet; the resulting component can secure multiple pipes replacing several P-clips.  A CAD model of the basic concept is shown below.

Over-Engineered P-Clip Replacement Concept...

The steel plate, with the rivnuts welded in place, is welded to the chassis, the two halves of the nylon clamp are then placed over the rivnuts, bolted down and voila!  One pipe clamp with no drilling into the chassis rails.

After much mulling over this solution, I realised that I was probably being a bit anal about this.  On close inspection, there are already several holes drilled in the AK chassis for the body mounting bolts amongst other things.  I will also probably be unable to avoid drilling some holes for mounting the 3-way brake unions and eventually for mounting the fuel pump and filter.  Additionally welding the mounting tabs onto the chassis will involve removing some of the powder coat and having to try and recoat the patches, plus some welding into some tight spots between the chassis rails; which to be honest, sounds like a lot of faff (even for me).

(And actually by the time I have posted this I will have drilled a load of holes in the rear of the chassis for the fuel tank mounting...)

So maybe the answer is to bolt the clips to the chassis after all; in which case why not just use P-clips?  However, I still think the bespoke solution is quite neat and does have the advantage that I can use a single clip to hold multiple lines (including potentially the fuel lines) so that does reduce the number of holes that I will need to drill.

So with a slight modification, Version 2 of the P-Clip replacement concept was produced in CAD.

Version 2 - spot the difference!

The next step was to make a prototype based on the CAD concept.  The mounting plate was cut from a spare piece of 2mm sheet steel.  I drilled two 7mm diameter holes which were countersunk to accommodate two M5 rivnuts and allow the rivnuts to sit flush with the back of the plate.  A 5mm hole was drilled for the mounting bolt.

The clip body was made from two pieces 12mm thick Nylon-6 sheet cut 12mm wide.  The bottom piece had two 7mm diameter holes drilled to sit over the rivnuts and a 10mm counterbore 5mm deep to accommodate the head of the mounting bolt.  The top piece was drilled with two 5mm diameter holes with a 10mm counterbore 5mm deep to allow the heads of the fixing bolts to sit below the surface of the clip.

The final step was to hold the two pieces of the clip together and drill a 3/16 diameter hole and a 1/4 diameter hole (for brake and clutch pipe respectively) along the join line between the two halves.

All the components for the brake pipe clip...

M5 Rivnuts in place on mounting plate - large counterbore in bottom half of clip is for cap-head bolt

Bottom half of clip in place on mounting plate

Top half of clip in place - secured with M5 x 16 Cap-head bolts

Completed assembly - holes for 3/16 brake pipe and 1/4 clutch pipe

Rearview with M5 x 6 Cap-head bolt in place for securing to chassis

I'm pretty pleased with the prototype so now I just need to knock out a few more in various configurations depending on fixing location and pipes that need to be secured.

To be honest this is completely unnecessary and massively OTT compared to the perfectly adequate and usual methods of fixing brake pipes.  But it's another little way of putting my own stamp on this build.

Fuel Tank Mounting - Part 1

According to the AK Build Manual, the AK-supplied stainless steel fuel tank is fixed to the rear chassis subframe using M8 bolts; the fuel tank is hung under the subframe and bolted into the chassis through the mounting tabs welded onto the fuel tank.

Now petrol has a specific gravity of 0.720 (compared to 1.0 for water) and the AK fuel tank has a capacity of just over 65 litres.  The empty tank itself weighs approximately 15.2kg. So with when full, the tank and the fuel will weigh approximately 15.2 + 46.8 = 62 Kg.

There are 5 tabs on the AK fuel tank, each held up by an M8 bolt, so each bolt needs to take 12.4 kg.  The proof load of a single standard Grade 8.8 M8 bolt is 2120kg so there is a hefty factor of safety on the bolt even accounting for any dynamic factors.

However...

The rear box sections of the AK chassis that the tank is bolted to are only 2mm thick and an M8 bolt has a thread pitch of 1.25mm, so at best only two threads of the bolt are going to be engaged.  

When a tensile load is applied to a bolt it stretches slightly and the loading on each thread will be different; the first thread at the point of connection sees the highest load and decreases for each subsequent thread.  It is widely accepted that a minimum of six threads is required to establish a full strength connection (the first thread takes approximately a third of the load, the first three threads take approximately three-quarters of the load and the first six threads take essentially the whole load and beyond the first six threads the remaining threads are basically under no load at all).

There is some complicated theory behind all this...

No, I don't understand it either...
(Extract courtesy of Journal of Multidisciplinary Engineering Science and Technology (JMEST)

Basically, the outcome of all this waffle is that in order to ensure a full-strength bolted connection there needs to be a minimum bolt engagement of six threads and bolting the fuel tank to the AK chassis will only result in two threads being engaged (ignoring the fact that each bolt is actually only taking around 0.6% of its ultimate load capacity).  

My proposed solution to this is to create some bosses from 20mm diameter steel bar which will be tapped for a 20mm M8 bolt.  These will be installed into the rear chassis subframe around the fuel tank by drilling a 20mm diameter hole into the top of the rail, an 8.5mm hole in the bottom of the rail, inserting the bosses, which will be held in place by an 8mm bolt to maintain the position, while welding the boss into the subframe on the top surface.  See the diagram below

Section through rear subframe - fuel tank mounting boss

The fuel tank has sat in place on top of the rear subframe since the chassis arrived serving as a temporary shelf for various bits and pieces (I had put some duct tape over all the openings in the tank to prevent the entry of any detritus).

The first job was to remove all the crap off the top of the tank and even up its position within the rear subframe.

Fuel tank in position on top of the rear subframe

The rear subframe rails are 40mm wide steel box. I set some callipers to 20mm and scribed a line across each fuel tank mounting tab; even though the tab might not be sitting centrally on the subframe rail, this operation should ensure the resulting holes are all in the middle of the frame rails.

Scribing centreline of rear subframe rails on mounting tabs

I scribed the centreline of the mounting tab in a similar way and centre punched each of the tabs at the intersection point.  I then drilled a 4mm diameter pilot hole through each of the mounting tabs and through the top of the rear subframe rails. Stainless steel can work harden if it gets too hot while drilling, which can severely reduce the working life of drill bits, so I used plenty of cutting fluid to try and keep the drill bit cool.

Drilling pilot holes...

After removing the fuel tank, the next task was to attempt to drill a 4mm pilot hole in the bottom of the subframe rails perpendicular to the holes in the top of the rail.  I made a simple jig using an offcut of 50 by 25 rectangular steel section.  Using a pillar drill I drilled a 4mm hole through both sides of the steel section from the top; so in theory, these two holes should be exactly aligned and perpendicular to the face of the section.

Using the 4mm drill bit as a dowel I aligned this jig above the pilot holes in the top of the subframe rail and clamped the section in place.  I then drilled through the jig, the pilot hole and through the bottom of the rail.  Hopefully, all these holes will be fully aligned and the bottom hole will be directly under the top hole.

"Verticality" jig from off-cut of 50x25RHS

Next step was to open out the top holes to 20mm diameter using a step drill.  Step drills actually result in a proper "round' hole whereas normal drill bits produce a slightly triangular hole. I did check the length of the bit to make sure that I didn't inadvertently open out the bottom hole to too big a diameter; as if I had planned it perfectly, as the top hole opened out to 20mm, the 4mm step in the drill bit would be sat in the bottom hole!

2-20mm Step Drill Bit

Drilling out top holes - almost there...

...and done!

Unsurprisingly drilling out a 20mm hole creates a large amount of swarf and metal chips.  I used a magnetic probe to try and fish as much as I could out of the rails and then deburred the holes on the inside and out.  

20mm hole in the top of the rail with 4mm pilot (hopefully) dead centre below in the bottom of the rail.

The final step of this operation was to open the bottom pilot holes out to 8.5mm.

To make the mounting bosses I used a section of 20mm steel bar.  In theory, each boss needed to be 38mm long to sit flush with the top of the chassis rail.  However, I noticed that on one of the rails, the welded seam of the box section was on the bottom face. So to be sure I measured each boss individually by inserting the bar stock into each hole until it bottomed out and scribing the bar level with the top of the chassis rail.  I then cut all of the blanks for the mounting bosses (numbering them so I could tell which hole they were for).

Cutting 20mm Bar Stock using a Bandsaw

Five blanks cut to length and numbered

Now each blank needed to be drilled and tapped to take an M8 x 20 bolt.  I marked the centre point of each blank by setting my vernier callipers to 10mm and scribing three lines at roughly 1/3 intervals around the bar.  The intersection point of these three lines should be bang on in the centre of the bar.  This point was then centre-punched.

Centre point scribed and punched

I then set the blank into a drill press vice, checking to ensure it was perfectly square to the drill bit.  With a 4mm diameter bit, I then drilled a 25mm deep pilot hole into the end of each blank in turn.  Again I made sure to use plenty of cutting fluid and removed drill bit and cleaned out swarf every 4-5mm of advance.

The holes will be tapped for an M8 x 1.25 thread so the holes need to be opened up to 6.3mm diameter before tapping.  Again lots of cutting fluid and cleaning out every 4-5mm of advance.  As a final touch, I countersank each of the holes.

Opening Pilot Holes out to 6.3mm Diameter

Ready for Tapping

I then tapped the holes using an M8 x 1.25 tap.  As with the drilling I used plenty of cutting fluid and the recommended technique of 2 turns forward and 1 back to snap off the 'chip' caused by the tapping operation.  I also removed the tap occasionally to clean out the flutes.

Tapping in progress!

Final bosses - all tapped and ready to go!

I did a trial fit of the bosses into the subframe rails.  I was quite pleased to see that they all fitted flush with the top of the frame rail and that the hole in the bottom of the rail matched with the hole in the boss so that I could install an M8 bolt!

All fits perfectly!

I need to weld the bosses into the subframe rails to complete this part of the build.  I will address that operation at some point in the future. 

This is all probably massively over-engineered but I can sleep easy that no matter how much fuel is sloshing around the tank during hard driving conditions, the fuel tank is going to remain securely attached to the car!

Rear Axle Strip Down - Part 9 - Hub Cleaning

In amongst getting various parts powder-coated, I have been progressing with cleaning up the rear hubs.

These are an aluminium casting and, after many years of service on the donor vehicle, were suffering from some heavy oxidation, particularly around the area where the steel brake back-plate sits, as well as having a good coating of general road grime.

The first step was to give the hubs a good soaking in a bucket of boiling water and some detergent - I used a couple of dishwasher tablets.  Vigorous brushing with a stiff nylon brush shifted some of the grime.  

One grimy rear hub ready for a bath...

...after a long hot soak and a bit of brushing

Note it is not recommended to use steel or brass wire brushes on aluminium as particles of the wire can embed themselves into the soft surface of the aluminium which can lead to future rust staining in the case of the steel brushes and severe corrosion in the case of the brass brushes especially if exposed to road salt.

After initial cleaning - note heavy oxidation around brake back-plate mounting flange

Next on the list of beauty treatments for the hubs was a session with the sandblaster.  Actually quite a lot of sessions.  Over many weekends...

The sandblasting gave reasonable but variable results.  Some areas cleaned up quite nicely - generally the flat areas.  The blasting removed the white aluminium oxide corrosion around the back-plate flange.  However, under the oxidisation the surface was quite pitted and, despite trying different media, the sandblasting did not remove all the of the oxidisation.  The resulting finish still had quite a lot of black oxidised aluminium in amongst the shinier areas.  This was the same case for some of the edges and recesses within the casting.

Hub after sandblasting - some areas of ingrained oxidisation remain...

The rear side of the hub - it was tricky to get sandblaster nozzle
 into the bottoms of all the recesses

Oxidisation still present in areas of pitted aluminium

The next treatment involved a liberal dousing with Frost Ali-Clean and a vigorous brushing with an old toothbrush.

Time for another scrubbing...

I have to say I was quite disappointed with the Ali-Clean.  It is supposed to be used on unpolished aluminium castings to clean and etch the surface.  Having applied the product as per the instructions and rinsed off with water, I'm not sure I noticed any difference in the finish.  The aluminium surface was no brighter and the areas of deep black oxidisation still remained.

This left me resorting to the electric drill and a brass wire brush (totally ignoring my own recommendation above....).

Recommendations?....more like guidelines...

This treatment did improve the finish slightly - but still did not remove all the areas of black ingrained oxidisation.  I even tried a further treatment of Ali-Clean (and I was still unimpressed...).

A before and after comparison - front side....

...and rear side.

When you look at the before and after photos, the cleaned hub does look significantly better than the 'before' version.  But I thought there was still room for a lot of improvement.  

Given that to get to this stage had probably taken me about 5 or 6 weekends and I had only 'cleaned' one hub, I came to the conclusion I needed a Plan B.

Plan B involved a trip to see the professionals, in this case, a local sandblaster in Braughing, RFI Screening.  They offered to vapour blast both hubs for the princely sum of £20 each.  In hindsight, I should have just taken this approach in the first place.  The results from RFI are amazing and I could have saved myself a lot of effort.

Hubs - straight back from the sandblasters...

...vapour blasting leaves a clean satin finish...

...even recesses on the rear are nice and clean.

Differential - Part 4 - A strange sense of deja-vu...

In Part 3 of the rear axle assembly I had got as far as installing the rear wishbone tie and the differential tie bars.  I then undid most of that work and, at the back end of last summer, I stripped and re-powder-coated the wishbone tie.  So, one of the first jobs back in the garage for 2020 was to actually reinstall the wishbone tie.  The assembly was cunningly straightforward - almost as if I had done it before...!

Generous reapplication of copper grease on rear diff shafts...

Mounting Bracket re-installed...

Inboard Compliance Buffers back in place...

Wishbone tie re-installed and generously filled with copper grease...

Outboard Compliance Buffers also generously coated in copper grease...

With the outboard compliance buffers installed but not fully tightened I was able to jiggle (technical term) the ends of the differential tie bars into the wishbone tie and tap the bolts into place.

Ends of Diff Tie Bars lubricated with - you guessed it - copper grease...!

Reassembly complete!

The sense of satisfaction at getting this back together was slightly overshadowed by the realisation that I am now back to the same stage that I was last July!   But still, at least I have now done something useful on the car in 2020, so its progress of sorts.

I am still not happy with the length of the thread of the 7/16" bolts that hold the differential tie bars in place.  However, despite many hours of internet searching, it would seem that all UNC partially threaded bolts come with a standard length of the threaded portion (at 1.125 inches).  The only choice I seem to have to make these more aesthetically pleasing is going to be to cut the end of the threaded portion down a bit. 

Another job for the future...

An offensive length of protruding thread...