Yamaha R1 Radiator Modifications

Curved motorcycle radiators like this Yamaha R1 model are very expensive to replicate. Manufacturing an aluminium motorcycle radiator from scratch is no mean feat in itself. I usually buy in the cores cut to size with end plates already welded.

The problem with this is that the actual core channels (the tubes where the water travels through to cool) are extremely thin material and takes considerable skill and specialist welding equipment to successfully complete.

When it comes to replicating curved motorcycle radiators in aluminium the failure rate of the cores can be over 30%. This is because the cores literally have to be curved by gentle rolling pressure in something like “pinch rolls”. This is typically a very “hit” and “miss” affair, often leading to core channels splitting during this shaping process.

2014-12-31 12.44.17 2014-12-31 12.44.32The customers requirements with this radiator was to chop off the end cans and manufacture custom made larger end cans with a specific pressed pattern to match up with the pattern machined into the ends of his fork clamps .

So first things first, take some measurements, look at how this rad has been made. Its a factory standard item and hence has cast aluminium end cans (in this case) which have been “clamped” and “sealed” in place. The idea with this is to fabricate new end cans from 1.5mm thick aluminium sheet, I will for ease, re-use the fittings, filler neck and pressure cap. I will have to spin up a couple of threaded bosses to weld to the cans to pick up the third fan mounting point. These will be machined from 16mm diameter HE30 bar on my lovely Myford ML10 lathe, with an M6 thread through.

2015-01-02 12.00.33Here we go then no turning back now, 4.5 inch grinder with a 0.8mm thick cutting disk. Cuts through most general metals like butter. Shown above then is the first end can sliced off, revealing the ends of the core channels and the end plate. Now what I have done is cut the end can off approximately 10mm from the sealed flange. I have made the new back plate with a slot cut out to neatly slip over the small “stub” left. This means I can weld around the inside of the can back plate to the stub – as shown below.

2015-01-03 14.01.57

So to weld all the way around the inside of this back plate to stub, I need to make sure that I cover the ends of the core channels as if my TIG welding arc “wanders” and catches the ends of these anywhere then potentially the material could be “blown away” by the wandering arc and leave me with a slight “pinhole” allowing water to leak out. To overcome this I covered the ends with a cut piece of steel.

2015-01-03 14.30.57-1

Sp once each end can back plate was welded on, I will need to level this as the welding will distort it. This is a relatively simple job with a Planishing Hammer and a block of aluminium.

The rest of the panels to make up the end cans were cut to size and pressed up as required. The main outside face that needs “press marks” in needs to be set up and pressed to make the pattern in required. The first thing I need to do is to make up a pressing tool to achieve the desired result in the aluminium panel.

The best way I have of doing this and controlling the amount of pressure on the actual press tool is to use my hydraulic press. So firstly I cut up some heavy duty plate that I have in the workshop.

2015-01-02 13.34.31Above shows the first “test”piece that I set up using 10mm plate and small “slices” of steel and 3mm round diameter bar. Trial and error with this using some scrap aluminium gave me the definitive design and dimensions I needed. I then proceeded to make the “actual press tool”. The result is shown below.

2015-01-03 11.31.402015-01-03 11.31.13

The above 2 images show the tooling laid out and also in the hydraulic press with the aluminium panel in position.

2015-01-03 15.39.05The image above shows the first end can panel work positioned and tacked up to the back panel. Its important for a good weld that all these panels are tacked up in what we call an “Outside Corner joint”. Shown below.

images

Image taken from – http://www.twi-global.com/technical-knowledge/job-knowledge/fillet-welded-joints-a-review-of-the-practicalities-066/

Using an AC TIG welding plant with a Gold Tipped Lanthianated tungsten I was able to weld the panels together. When aluminium welding thin material it can be very difficult for the inexperienced welder to be able to see when to add filler material. This can be because the oxide layer will float on top of the weld pool and sometimes masks the molten weld pool. Other issues with welding aluminium that can be a problem to inexperienced welders is the fact that aluminium dissipates heat  so quickly, this is different for mild steel for example in that the heat stays more local, whereas aluminium needs lots of heat input to achieve a weld pool as its a very good heat conductor. Even on thin material. Thin aluminium like this end can (1.5mm thick) leaves less window of opportunity between a weld pool being produced and the weld pool becoming so hot that it drops through and you end up with a molten blob on the bench.

2015-01-03 16.54.08

So above we can see the first end can zipped up. Now the “new” material was easy to weld up, the not so easy bit was to weld the end can to the end of the original flange left on the rad. Issues around this are caused through contamination from the sealer and material coating the sits between the 2 material thicknesses on the original end plate. When trying to weld the new end can onto this, due to the heat input and the nature of the aluminium weld pool, contamination (sealer / paint / coating etc) gets dragged through the weld pool and bubbles up on top of the weld pool. If this is not “burnt” out properly we end up with pin holes in the weld, called “porosity” causing leaks through the weld material – not what we want.

2015-01-18 13.06.06-1

Above you can see both end cans fully welded to the original core, with the pre-drilled holes ready for the pipe fittings to be fitted and welded in place. Why the cardboard? I needed to ensure that whilst cutting and sanding, laying the radiator over and on the bench and handling it that the cores were not damaged. Obviously as soon as it got hot enough the cardboard wanted to burn. During welding whilst it was being under the shielding gas it wouldn’t burn but, the minute the shielding gas stopped the cardboard would start to burn. This is because the shielding gas I was using was stops oxygen in the atmosphere from contaminating the weld pool. And hence also stops air getting to the red hot cardboard.

2015-01-18 14.39.11Pipe fittings welded in place complete with new mounting bosses to pick up on the third fan bracket.

2015-01-18 14.41.00Fans Mounted back on to check fit.

2015-01-18 14.41.24And after testing – pressure tested to 15 psi (just over 1 bar) and a bit of cleaning up, this was the end result. Hopefully it will look as good painted and on the bike?

Pictured below is the laser cut stone guard I drew up with the R1 logo in. This needed to be rolled to suit the radius of the core of the radiator and then bolted to 4 small welded bosses.

 

2015-01-17 12.01.47

 

 

We look forward to keeping you up to date.

For all your Custom Car & Motorcycle Parts, Manufacture and Welding / Repairs
Don`t forget to email or call us for all your custom made 1 off bespoke items.

Alloy welding / repairs, custom parts, Welding Instruction.

 

 

Custom Motorcycle Pannier / Carrier Rack

There are so many rack and carrier options and manufacturers on the market for most modern motorcycles its a bit of a  minefield. When it comes to older and less mainline bikes there seems to be less of an option.

With that in mind I have had a customer that wanted a bespoke rack for touring purposes. He had previously had a bit of a lash up and had made a clamped on section of an old rack with a new piece to sit and lock on a top box. This worked but was held together with bits of cut mild steel flat bar and hex head bolts. It did the job!

2014-11-09 09.20.02

Original Rack above

He wanted a better design and improved looking rack to fit in with his newly refurbished motorcycle so I took his home engineered item and grafted the 2 racks together removing the old clutter that was unwanted.

To do this and ensure the rack stayed square and level on its original mounting points several bracing bars were required to  be tacked on to support and hold the rack parts in position whilst the old unwanted parts were cut away and new supporting bars added in the relevant places.

original rack Braced up ready for cutting out unwanted parts and adding new support tube frame and blending in work. ready for dipping and re-chroming.

2014-11-09 10.28.15

Image showing rack with removed section laying in the back ground.

Once the old was removed there was lots of sanding and blending to ensure that the existing tubes were smooth and free from old weld, dimples and lumps, all of which would show up even more once chromed. Once this was complete I could concentrate on shaping, welding and blending in new material to support the top of the rack.

Bottom support tubes were bent from 16mm diameter 1.5mm wall seamless tube. The ends were then scalloped to fit the original tubes, positioned, tacked and welded.

richard rack 1 richards rack 2 richards rack 3 richards rack 4

Once complete the welds were blended in and “breather” holes were drilled at strategic points for the chroming process to be undertaken. Its important to remember that the material needs to be as smooth as possible to achieve a quality finish when chroming. Any little mark or pimple can potentially show up and lead to a poor quality finish.

Choose your chroming company with care and ask all the right questions. I will post another blog about this subject.

We look forward to keeping you up to date.

For all your Custom Car & Motorcycle Parts, Manufacture and Welding / Repairs
Don`t forget to email or call us for all your custom made 1 off bespoke items.

Alloy welding / repairs, custom parts, Welding Instruction.

Hayabusa Fuel Tank Chop

Modifications to the Suzuki Hayabusa fuel tank is required by our customer to enable him to get his head down lower underneath the “bubble” of the lowered screen. Thus reducing drag to enable him to go for a speed record on his Motorcycle.

standard tank

This task was particularly difficult as not only do I need to approach the task correctly I also need the vision to see what the tank needs to look like in the end.

Sensible considerations need to be thought about before I put a cutting disc anywhere near the tank –

1 – Shape and form of the tank now

2 – Shape and form of the tank when modified

3 – Ways in which we can chop the tank and reduce the amount of welding during reassembly

4 – Heat input and distortion during reassembly

5 – Making sure that cutting and shut lines are on a natural edge, form or pressing mark where possible

6 – Re-instatement of drain tubes for filler neck

7 – Maintaining the shape and form of the tank bottom – ensuring refitting and alignment accuracy.

All of the above needs to be considered along with actual end product shape that the customer wants, how we undertake this tank modification may affect how much skimming with filler material and painting is required. Obviously the less the better as this will help both keep the tanks weight down and also the cost in terms of time and materials to “finish” the tank.

2014-12-06 14.33.42

This tank is a pretty big and bulky affair, I have the option of cutting away the internal baffle pot in the filler neck to reduce the height even further as the tank will only ever be filled with race fuel via a fuel can. The overall profile looking from the bottom of the tank needs to be maintained – hence as I weld the modified top I will need to strengthen the base with small box section to help prevent the shrinkage caused by welding the top panels from distorting the base panel. If I didn’t do this it may cause the mounting points to move and also if the base panel actually distorts it may then also conflict with the intake system on the underside.

2014-12-06 14.33.55

The image shown above and left gives an indication of the first “cut” line. How did I decide upon this?

The tank has “natural” form lines from original manufacture and it was easy to decide where I wanted to cut it. However, this also needed to tie up with the foresight of how the tank would go back together and if cut along these marked lines would I be able to fit the parts back together once a “slice” had been removed from the tank to get the top down lower.

With tank chops it is never easy to predict exactly how the job will go until I have actually removed some material. But, it is crucial that a good insight, skill and working knowledge be applied to the whole job planning before cutting anything commences.

Reasons for this is that if I cut wrong then I could be making myself a whole lot more work trying to put it back together later, typically this is true  if I then have to start adding in more welds or making infill panels to weld into gaps left.

Another thing to remember also is that an awful lot of tanks have at some point been repaired, as can be seen is the case with this one. Looks to me like a very unlucky rider may have had an accident and smashed his marbles into the back of the tank. Just hope they were steel ones! See the area of filler material added in the image below.
2014-12-06 14.44.42

2014-12-06 14.42.10

Before being able to completely remove the sliced of lid of the tank the drain tubes have to be drilled out (shown above). Then the top of the tank can be removed, revealing the inside condition of the tank. Which in this case is only surface corrosion as its obviously been sitting open for some time with no fuel in it. This one is in reasonable condition.

2014-12-06 14.45.52

Now a major part in modifying existing tanks like this is the fact that they are generally “Stamped” or ” Pressed”. Normally in 2 sections a top and a bottom. The pressing process normally starts with a single sheet of material which is inserted into a hydraulic press with matching top and bottom die tools. Once inserted the material is subjected to large hydraulic forces when the top and bottom die come together to “stamp” or “press” the panel into the desired shape. See example below –

stampingSourced from – www.learneasy.info

This process obviously puts stresses and strains in the formed material shape, not a problem for this tank that will be assembled on a production line where the two halves will be joined, typically by electric resistance welding process all the way around the bottom flange.

However, from my point of view as a sheet metal worker about to cut and modify the original tank, these manufacturing stresses and strains can cause me a headache when trying to reassemble the tank and then butt weld or lap weld the new joints. And the reason for this is that I have no idea how the material will react to heat input from welding until I actually start reassembly, tacking up and welding. The reason being for this is that the metal being no thicker than about 1mm will actually bend and distort the minute any welding heat is applied. This can cause the problems of buckled and mis-aligned joints. Increasing the chance that the finishing process will involve skimming the tank with a thin layer of filler material to flush and level / smooth out the metal ready for painting.

2014-12-06 17.00.57Initial cutting, showing the “lid” dropped back on the tank.

Where the diagonal line of masking tape is I have actually made another cut line to allow me to be able to shape the panel inwards towards the centre line of the tank.

2014-12-10 15.04.42

 This image now shows roughly how much “drop” has been achieved by slicing a triangle of material out of both sides of the tank. Approx 2.5 inches at the rear most position back to nothing at the front of the tank. See how much the top has dropped down. If I now remove the baffle tube from the filler neck I can get about another inch which will leave about 10mm clear between the underside of the tank pressing (which clears all the intake) and the underside of the top panel.

2014-12-06 14.47.012014-12-06 15.25.43Images showing the underside of the filler of which the small diameter baffle tube will be chopped off to gain a bit more drop on the top panel.

2014-12-10 15.05.01Here showing the tank with the lid just sat back on, now you can see I have some panel beating and reshaping of the side scallops to blend in with the contour of the top panel. The shaped metal under the masking tape and towards the back of the tank will require shrinking to get it to “pull” in towards the centre line of the tank as well. This will be achieved by manually using a block or “dolly” and a planishing hammer. Good old bit of elbow grease! What I have to do is hammer this down to the block held behind the panel and effectively “shrink” the material to get it to pull in. There is an art to this as if inexperienced many people will “thrash hell” out of the metal and effectively end up “stretching it” which is what we dont want. Careful and gentle hammering will improve the chance of the metal molecules overlapping and “shrinking” in. If I cant get enough by using dolly and planishing hammer, some heat will need to be applied.

Heating mild steel to a cherry red and then allowing to cool down will “shrink” it. Theory behind this is the fact that when steel becomes hot all the  molecules become exited and bounce off each other, hence expanding; upon cooling the molecules calm down and pull in tighter together and effectively shrink the material. Rule of thumb is that steel always contracts that bit more than it expands – hence achieving the desired shrinkage.

2014-12-10 15.15.15Planishing the material to shrink it

And below heating the material to shrink the edge in a bit.

2014-12-21 16.44.49

2014-12-10 15.31.16Just a comparison above of the original and chopped down tank.

2014-12-14 12.30.38So once I was happy with the general shape of the panels I proceeded to tack the top panel up to the bottom section. Care had to be taken to make sure that the butt weld joint preparation was set flush with minimal gap between the two edges. This is a critical stage as the more time and effort spent at the set up stage the better and easier the welding stage will be.

Its also important that as I want the tank to be even side for side, some basic measurements along the weld joint were taken before tacking up the joint to ensure some symmetry once finished.

2014-12-14 12.30.542014-12-14 12.31.10

Once the tank was at the above stage the hard work really begins. I needed to effectively “blend” and “fill” the gaping open areas of the tank. Now to start with some simple paper and card templates were made up. This shape was then transferred onto metal and the piece cut out. The actual metal pattern will need lots of  shaping, trimming and trial fits to get it to sit and look right. Again the tools of the trade for this are the trusty Planishing hammer and dollies, pinch rolls, bossing mallet and sand bag.

2014-12-14 13.31.07You will notice that I have used lined paper for a basic template and also numbered up the lines to give me a better degree of accuracy and symmetry.

2014-12-14 13.36.13Template laid onto steel ready to be cut out. Note the extra 10mm line around the outside of the template. This I have added to allow for a “joggle”, this will slip under the metal at the front of the tank and help maintain some shape and act as a guide when inserting the infill panel. Although I have one template this will be cut into 3 panels to allow for easier shaping and fitting to the tank.

2014-12-21 10.55.32Pinch rolls are used to give some required form to the infill panel. As you can see the 2 black lines represent cut lines where I will chop the panel into 3 sections for ease of fitting.

2014-12-21 10.58.52Shown above the infill panel with some rough shape. I will now chop this into 3 parts and start further shaping and trial fitting starting with the top panel.

2014-12-21 11.43.07Once shaped the top panel will be joggled around 3 edges ready for fitting. Additional shaping and forming will be required after as the joggling process will pull some of the original shape out.

Once happy with the shape I clamp the infill panel in place and tack it in a couple of places and recheck fit to make sure that I am happy with the fit up.

2014-12-21 16.09.142014-12-21 16.18.06

Top panel in and work starts on shaping, forming and trial fit up of each side infill panel.

2014-12-21 16.50.462014-12-21 16.51.10

Once infill parts are shaped and trial fitted (this process may take half a dozen attempts to get good fit) then I can tack up.

2014-12-21 16.48.402014-12-28 13.02.33And here you can see the top and right infill tacked up in place, just the left hand side to complete.

2014-12-28 13.02.39

So all tacked up and ready for welding, what process and why?

I will be MIG (Metal Inert Gas) welding the joints on this tank. Reasoning behind this is the fact that the tank is made from thin – about 1mm thick mild steel. If I use MIG MIG(Image taken from – www.mechanicalengineeringblog.com

I can maintain minimal heat input and adding filler material so that I dont blow through the thin material. If I was to try try and use the TIG (Tunsten Inert Gas) welding process,TIG

(Image taken from – en.wikipedia.org)

this is much slower and puts more heat into the job and I would not be able to add filler wire quick enough to fill the joint before the metal was melted away. I could also try and use the MMA (Manual Metal Arc) welding processMMA(Image taken from – www.steelconstruction.info

but this would be too hot and melt away the tank metal before any filler material could be deposited. I could gas weld the tank if I had the correct oxy-acetylene equipment and this would indeed be possible as like MIG you can control the heat input easily (if experienced in gas welding).

MIG welding is my preferred method as this gives me control of heat input and control of the speed of filler wire input, it also allows me to “fill” some gaps and many other “parameters” that I can manage to help complete the job successfully.

Using the MIG welding process to complete this task still needs a large degree of skill and understanding of the actual process and welding techniques to be used. For this job I use a technique called the `Pull` Technique, in which you actually pull the torch from left to right (if your right handed – vice versa for left handers); I also pulse the weld -this means to weld in short bursts and by allowing less than a second between pulses for the steel to cool down I can still ensure that each “burst” melts into each other; sealing the tank . The longer you leave it the more chance you’ll have of cold joints and pinholes between the welds. By utilizing this method we minimise heat input and have the best chance of not blowing through the thin sheet metal.

stitch-weld This image (taken from http://www.mig-welding.co.uk/thin/stitch-weld.jpg) clearly shows the effect that you should end up with. Now not only do I use this process I also fill the tank with Argon gas, this is an inert gas and shields the underside of the weld from atmospheric attack. This should leave a nice clean penetration bead on the inside of the tank.stitch-undersideUnderside of weld should look something like this. (Image taken from http://www.mig-welding.co.uk/thin/stitch-underside.jpg).

So theory into practice and the following images are what I ended up with.

2014-12-28 16.15.36 2014-12-28 16.15.46 2014-12-28 16.16.44

Even with all the effort and knowledge to complete this job, there are still the unknowns of how the stamped material will react to the heat input from welding. There will still be buckling, bowing and distortion to some extent. This often ends up with a butt weld that has been set up flush, distorting on one side and either a “lump” or a “dip” one side of the welded joint. This can be seen slightly on the last image shown above just under the end of the Suzuki Logo. As I cant get into the tank now to be able to dolly this out, a thin skim of filler will be required.

Main tank done and now I have to complete the drain pipe welding. Shown below.

2014-12-30 14.32.25 2014-12-30 14.39.20 2014-12-30 14.51.36 2014-12-30 14.52.01

So once everything was welded up, pressure testing the tank to check for leaks is critical to ensure a fully sealed tank.

I had to make up some flanges and seals to pressure test this tank as I didn’t have the cap or pump assembly to fit and bolt in place to seal it.

2014-12-30 14.56.55 2014-12-30 16.29.59

So from the template I made up a steel base plate and silicon rubber seal. I could then insert my pressure and blow off valve assembly into it. The filler flange was sealed in a similar way with a pre-made plate and seal arrangement of which I have loads of differing sizes that I have made up for previously tested tanks. If I didn’t have the blow off valve in place I could easily blow the tank up into a balloon shaped item if excess air pressure is used. I usually only use a 1 bar blow off valve (about 14 PSI), this is quite adequate. Once pressured up I spray all the welded joints with a mixture of fairy liquid and water. If there are any pin holes this shows up as bubbles on the outside, which I can then spot up.

Sometimes the “spotting up” is the tricky bit, as if there are pin holes or mini fractures anywhere in the metal or weld areas once I apply heat and filler material to it to seal it by welding, the metal expands and contracts again. Sometimes this makes the problem worse and I end up “chasing” a pin hole or hairline fracture along the edge of the welded joint.

The finished item ready for the paint shop.

2014-12-30 18.15.13 2014-12-30 18.15.26 2014-12-30 18.16.36 2014-12-30 18.16.44 2014-12-30 18.17.18Another job done!

We look forward to keeping you up to date.

For all your Custom Car & Motorcycle Parts, Manufacture and Welding / Repairs
Don`t forget to email or call us for all your custom made 1 off bespoke items.

Alloy welding / repairs, custom parts, Welding Instruction.

2014 in review

The WordPress.com stats helper monkeys prepared a 2014 annual report for this blog.

Here’s an excerpt:

The concert hall at the Sydney Opera House holds 2,700 people. This blog was viewed about 20,000 times in 2014. If it were a concert at Sydney Opera House, it would take about 7 sold-out performances for that many people to see it.

Click here to see the complete report.

Awesome Audi Build

 

This is worth a look

http://www.youtube.com/watch?v=pcUK_pLXWOY

 

Anyone interested in building cars will appreciate it.

Enjoy

We look forward to keeping you up to date.

For all your Custom Car & Motorcycle Parts, Manufacture and Welding / Repairs
Don`t forget to email or call us for all your custom made 1 off bespoke items.

Alloy welding / repairs, custom parts, Welding Instruction.