Harris mag3 homebrew project

[nlovien]

Hi - as per my intro I'd like to put my project build on here - and welcome all your good advise

bit of history leading up to the project: I have a magnum 3 in a sorry state after a big crash and a classic Peckitt and Mcnab Gurls blouse from the formula one TT series (late 70's early 80s) in need of restoring to original "as raced" condition. I enjoy the journey that a project build takes you on - as much as the finished project

Initial plans was to build two Gurls blouse CB900 based engines - one for each chassis - use the bent Harris as the mule / tinker and focus on the P&M as built restoration, however the cost of the Gurls blouse engine build is mounting - so I'm now looking at

1st repair and revise the Harris for the Gurls blouse - once sorted transfer the engine to the P&M and fit a GSX1100 or GSXR to the Harris = with an eye on an XR69 rep i.e. two of the great icon bikes from the TT series - that's now the present target

I've been learning techniques and making progress with the Harris frame so I'll be able to show this progress up to where I am at - then things will slow down as we move to real time

pics here are history of the both bikes - the Harris as I got it with a FI GPZ1100 - then adapted to GSXR; the P&M as when I got it (modified into a RS1000 endurance rep by the previous owner) and a picture of it when it was raced and what I intend restoring it to. It's got a great history - one of the two bikes P&M campaigned with a works RS1000 engine - a journalist for MC mechanics adapted it to the endurance rep. - some of you may recall an article on this bike back in feb 1985 in Mechanics "endurance specials on the road" featuring this bike and a tasty P&M Kawasaki KOOL rep - well this is how I acquired it

 

 

 

 

 

 

Edited March 9, 2017 by nlovien
title change

[vizman]

keep it OSS as per TFR

MORE of the suzuki, NONE of the other

Edited November 28, 2016 by vizman

[nlovien]

The Mag 3 suffered from a fair bit of impact damage - TBH could have been scrap so with nout to loose decided to see if I could learn to work with T45 chrome moly tubing and bronze weld. Decided on TIG bronze weld, so after a lot of practise with different techniques and filler rod spec - managed to work the T45 tube with a cheap e-bay special pipe bender and achieve reasonable welds using SIF albronze wire no. 32 - i'll go over what I learned doing this in more detail but suffice to say i'm reasonably happy with the end result - up to Harris std - nope - good enough to be safe ? - I'm confident - ok so I progressed with making a jig for the Harris to see how bent it was, surprised to find - even though there was significant tube damage - it appeared straight! - turned out it was the swing arm + bellcrank + rear suspension mounts that were bent - another "more detail" part that i'm presently attempting to sort - after repairing and modifying the main frame loops - I made up another jig to check for alignment and happy to say the head stock to swing arm relationship is straight and a fresh set of head stock bearings fitted i.e. no head stock deformation due to the work now done around it. I'm not finished the frame work yet - got the do the engine mounts and fit a low front cross beam - but i'm holding off until I get a GSX1100 engine to see if I can achieve a reasonable fit for both the Gurls blouse and the Suzuki engine around these parts ( the GSXR engine looks a better fit after modfying to fit the Gurls blouse engine than the original GPZ 1100 layout) so i already have a solution - but my heart is for an air cooled GSX

1st pic is the original Harris in the jig - you can see the main damage - I cut out all the original tubes except for the top rails and head stock and played around with various set ups until finalising on a layout that closely matches what Harris did in the 1st place - not because I specifically wanted it to be like it was - but simply because this was the best layout out - atleast to my way of seeing it. The last pic shows the jig for checking the post weld alignment - basically a lazer pen is attached to the tube extending out of the head stock - the two side rails hold the head stock accurately parallel to the swing arm pivots - the plate on the swing arm pivot is accurately perpendicular to the swing arm pivot, so! when you slide the lazer pen down the head stock tube - it projects onto the plate on the swing arm - if you get a parallel line from the lazer that matches a line scribed perpendicular to the swing arm pivot - on the plate, then its straight - it was !!! - not "almost - or close" it was like rail tracks - however it was offset by 2mm from what is the centreline between the swing arm frame rails - I can only surmise that this is how it was originally - no otherwise reason

[nlovien]

11 minutes ago, vizman said:

keep it OSS as per TFR

MORE of the suzuki, NONE of the other

will do

[TLRS]

Ingenuity with that laser pen.. cool stuff

[nlovien]

Bending the T45 Chrome moly tubing: 

armed with a 12T hydraulic press bender - about as cheap a tool as you can get for this job, my 1st attempt's  were crap but after a bit of trial and error I was able to achieve successful bends, not perfect - there is a bit of flattening but using the original framework as a comparison reference = well good enough. I can't say if this is the best way to do it but it's what worked for me

1st - junk the roller supports that come with these benders and replace with saddle supports using some tube with an ID slightly larger than the OD of the pipe you want to bend for the saddles - doing this prevents the brooze mark's that the point loading of the roller supports creat

2nd - turn down a broom handle to make tube plugs and fill the pipe with fine builders sand - take your time doing this - tapping the pipe and packing the sand as tight as you can get it - then hammer in the 2nd end plug ( best over fill with sand - try fitting this plug - remove a bit of sand - repeat - you don't want to under fill

3rd - the trick is to make a bend former fit really tight to the tube OD. I 1st made  inserts from various bits of steel tube fitted inside the selected former to bring the ID of the former close to matching the OD of the tube - easy done - cut pipe in half and use another bit of scrap that fits this ID to enable forming this insert into former using the bender. For the last insert, to match the tube OD - I used alloy tube, bit of fudge factor here, the ID of the alloy tube slightly smaller than the OD of the T45 i.e. enough to make a really tight fit and use the alloy's ability to flow into shape - don't initially form the alloy into the former, just lay it across the former with the T45 inside it and use the bender to 1st squeeze the T45 into this alloy sleeve, then you start bending both the T45 and this alloy sleeve together - I think whats happening is a) you get a tight fit and b) your spreading the bending force versus a point loading at the peak of the former

4th - work the pipe in small increments - take up the spring 1st, then just a few mm of bending - release and repeat - think a few hrs of this per bend to give you an idea of the small steps

lubricate around all pressure points on the T45 to ease removal ( you'll have to chisel the alloy sleeve off)

I couldn't get to a full 90deg bend doing this but not far from it and more than enough for the purpose of the frame - in the pics the final alloy sleeve is after bending the T45 - imagine it was was initially straight with the T45 laid across it

OK so yi don't need an expensive pipe former - great if you do but it can be done with an e-bay special

 

[nlovien]

making the jig

for me, there is only one key relationship to get right - get the steering head perpendicular to the swing arm axis - within limits all other aspects can be otherwise accommodated for ( and used ref: the existing frame geometry)

1st - set the support for the steering head - I used an existing yoke stem accurately fitted to an extension tube ( 100% parallel to the stem) and weld this to your base frame - vertical angle relative to the length axis is not important but make sure it is in the centre of your base frame and 100% 90deg perpendicular on the width axis

2nd I used an old set of bearings to lock the frame head stock into this stem (big note! - learned from not doing this!! - make sure you thoroughly degrease these bearings, if not - any welding work around the headstock will get contaminated by the grease via the heat / capillary action) - using your headstock bearing set nuts - clamp this down tight

3rd - now you'll find the position of the swing arm pivots - make the x 2 supports for this identical - i.e. don't match it to the frame - measure to only one side and make both sides from this - in this way you'll find out if the frame is twisted (i.e. it will not fit!!) -- initially when I did this I had to work the swing arm pivot in through the supports and frame, seemed like there was a bit of misalignment (twist)- about 1/8th" - ok there was enough flex to get it in but you can see - if i'd made it to fit the frame I would have been an 1/8th out

4th - then I started cutting out the damaged frame tube's ( RH side had the crush damage as can be seen - LH side looked like there was a slight buckle on the lower rail - quit a bit of spring when the tube separated i.e. there was a lot of tension in it - kept cutting away until I was left with just the two top rails connected to the headstock

5th - after cutting out - I removed the frame from the jig - and re installed it - hayho - the 1/8th error had gone - spindle fitted much easier now - ok confidence is good, no signs of need to initially straighten the frame 

if you look at the previous pics - you can see I chose to cut the main frame rails just fwd of the swing arm vertical part - my reckoning being, this is a relatively low stress / flex point being supported by the rear engine mounts on one side and fwd lower engine mounts on the other - I did bush this joint for added support - there's x 2 welds here - 1st was the bush to the tube ID (V notch) - then the top tube butt weld

[nlovien]

managed to adapt a swing arm to replace the bent Harris - keeps the same wheelbase. Made up a bellcrank to also replace the Harris item - also bent but i've done this to"fit" versus "tune" and it needs further tweaking to get the swing arm rake right. My thoughts were - make what fits as long as it isn't to far from original - get bum on seat feed back and use the Tony Foal suspension program to tune it right - However! project direction is starting to evolve in a new direction

 turns out there is someone reasonably close to where I stay that has a trident frame XR69 - I have a GSX1100 engine ( and a GSXR1100L) - you can see an obvious direction - the Suzuki floater suspension would be pritty easy to adapt to the mag3 to, i've changed too much of the frame already to call it a mag3 anymore so why not go the whole hog

so i'll stop this project here and when I firm up on the above - i'll start again on the project page

 

[Gammaboy]

11 hours ago, nlovien said:

mag3

I have to say - I'm not sure that's a Mag 3 frame - looks more like one of the Harris TTF1 frames

[nlovien]

thats a fair comment given the similarity and the changes I've done to it - TBH its now neither - more "in the style of", however the only diffrence ref: repair work is - I used a tube to brace from the rail that triangulates from the down tubes to the top rail; cross brace to the head stock (Harris used  folded sheet for this) and the angle of the two downtubes is swept back more ( Harris for the GPZ fit was more fwd / vertical - you can see the hassle this caused when fitting the GSXR motor ref: 1 and 4 down pipe routing) - becomes a mag 3 when you see it with its clothes on 

Edited December 2, 2016 by nlovien

[nlovien]

not made any progress due to my daily workhorse chucking its chain so been sorting this out - I did confirm the GSX1100 i've bagged has less than 30K on it 

so I thought i'd voice what I learned using TIG to bronze weld the chassis I'm a hobby welder so please appreciate this is not the words of a professional

Initial plan was to do the tube work - tack it up and give it to a professional - in due course this may have been the right thing to do  - hayho - but I started and didn't stop

1st why braze versus weld ? - you can TIG T45 easy peasy with std carbon steel wire and you don't need to post heat treat for the given tube wall thickness - if I was building from a start I would have done this - ok so I wanted to use the same method as used by the original frame builder - but use TIG versus gas - there are good reasons to bronze - I found out a few during the build - no distortion of frame or headstock around the bearing area - no excess heat damage to existing braze joints - no penetration through what was left of the Harris tubes. When you look at the properties of the braze looks like its great for providing a bit of give without fracturing and its not as weak as expected as long as you select the right wire - your using a relatively low heat and not melting parent metal so much less risk of heat affected zone stress failures

2nd why TIG versus gas - cause I can make a reasonable effort with a TIG unit - never done gas - but more because I just wanted to try and learn it 

3rd - the wire - SIF albronze 32 - its the recommended stuff-  check the properties of it, way above your typical silicon bronze braze wire and there's the twist, it doesn't behave like a typical braze wire and it costs a fookin fortune ( £90/kg) - before committing I did do a bunch of destruction test comparisons - using various other braze rods it was always the braze that would fail - using the 32 wire, I could flatten the chrome moly tube and not shear the braze - ok its good

4th - initial practise - used std silicon wire ( for TIG - not gas - don't use gas wire) - quickly got to producing nice fillet brazes, even getting the "stack of pennies effect" lovely - don't spend too much time doing this, the 32 wire doesn't flow like the cheap stuff - you got to learn with the 32, I also found that, as much as i've enjoyed working with an old BOC air cooled DC TIG welder, just not the right kit for doing this - looking for an excuse to get an AC/DC unit anyway to learn alloy - so bagged a half decent chinese hobby unit giving me auto start versus scratch ( you don't want to be scratch starting on a high chrome alloy = stress failure points) - I got a peddle and more knobs to play with than a graphic equalizer

5th - the joints - this is key, a nice tight 360deg fit for the tube interfaces - I used an 0.008" feeler as my gauge - achieving this is by way the most time consuming job when all yi got is a die grinder - you don't have templates and your doing one off's - but with practise using crap steel tube 1st you do get a "feel" for making the profiles - lots of practise and a fair bit of scrap though

6th - using DC versus AC ? both work both have pro's  / cons  - DC is clean, neat but I was struggling to get nice fat fillets  - fine with silicon but with the 32 it just doesn't wet out and flow the same - it does lay up fine but single pass only, trying to overlay multiple runs the heat just gets absorbed  - crank the heat up and your now into the risk of causing parent metal flow = fail - note I was using 40 / 50 amps to make a nice single pass - i.e. very little risk of parent metal flow - OK try AC,  better for producing fatter fillets - the AC is great as a heat source - this is more like alloy welding and you get to go play with all them knobs on the welder to find the right settings - the +/- bias is the most important one - i'll not  pass on my settings as its specific to each welding unit - but do play because when you find the sweet spot - it makes a big difference - with AC I was initially using 100 amps but I ended up nearer 130 amps as I gained confidence and use of the pedal - you don't have a peddle ? go get one - brilliant makes life so much easier - get the pulse rhythm going  with it and yir sorted - downside to AC? contamination and erosion of the electrode - I tried x 3 different types - ended up with white or red but they all quickly mess up -I was using a nice fat electrode with a ball end - still messed up - ok so you get used to stopping / cleaning / re grinding, but where you can - go back to DC for progress

7th weld finish ? I didn't master the stack of pennies effect whilst doing tube ends - keeping the wire dipping into the pool whilst keeping the tungsten at the right angle whilst walking around the tube - in tight space where the wire will melt to the parent if it gets too close before it gets to the pool etc.. etc.. - excuses from a layman versus a professional - but thats cosmetic, so I aimed to overbraze then cut back the surface to form the finished shape - doing this I was able to remove any contamination which tends to come to the surface and check for porosity / braze bridges - porosity wasn't an issue - I did create a few bridges at the tighter point of V parts of some joints - ok grind out and re lay the braze - point learned - start at the most difficult tightest bit and work away from it, don't leave it as the last bit to do working towards it

8Th checking - I did make up quite a few test pieces - and a few bodge joints that I removed - when removing the bodges I gradually grinded back the work to observe how the braze was relative to the T45 joint - I can confirm that it may not look pretty on the outside but you are getting a full root weld 360deg and again - no signs of porosity - even nicer is no signs of any parent metal flow into the weld - no parent metal shrinkage = no contamination of the chrome moly = which you do not want, as per striking an arc - this will just lead to high stress failure points. You do get surface etching when you use AC but no where near what you would call shrinkage. What I didn't expect was not much sign of any flow beyond the joint - i.e. into the inside of the tube - like what you would get with a solder, forsure it had wetted to the tube inside diameter - but not beyond it - I was using a 2nd argon supply into the tube inside - maybe this played a part - anyway the inside of the joint was nice and clean

summary ? - learn and use 32 wire it doesn't behave like std bronze wire - use DC and AC where applicable - both work and both have pros / cons - you really should have an auto start TIG welder to avoid scratch starting and a peddle makes things sooo much easier - practise!

to any professional welders I bow to you - to the guys who make the works of art that comes out of the likes of Harris / Spondon / P&M etc... I have an even greater appreciation of your skills

[busa ash]

looking good too me , keep up the progress .

[Pheasantplucker]

I must try bronze with the tig myself, keep up the good work.

[nlovien]

mind changed - again, really need to stick to a plan -- the GSX1100ES that I acquired for the engine, its just too complete and good, ok its a previous owner back n black all over spray tan but with less than 30K - a bit of TLC  it will make a good bike - shame to rob the engine from it, sooo the 1100L motor goes in - given whilst repairing the frame I worked it around another jap engine ( of old school) - the GSXR1100 motor is floating around inside the frame - not a problem to fit just the usual juggle to find the best compromise - want the crank as fwd as possible versus the sprocket as near to the swing arm pivot versus getting the no 1 and 4 ex pipe either around or inside the frame front down tubes versus getting the rocker cover off in situ versus where do I fit the frame lower cross tube whilst enabling ex 2 & 3 down tube clearance - and  the winner is ???? - No 1&4 ex downpipes going inside the frame tubes  - compromise is I'm moving the engine back from where I'd like it about 10mm - not a big deal, and i'll make a curved frame cross member that will also be the engine front mount location, this will give clearance for pipes 2&3

 

[nlovien]

OK, not making great progress but have finalised on engine position - basically as far fwd as possible whilst being able to get the rocker cover off - will need to make an ex pipe to work around the front down tubes - hoped to not do this - hayho, sprocket to swing arm pivot is the compromise - its too far fwd, - lesser of two evils as they say - keeping an eye on the rear swinger to sprocket relationship to maintain squat versus anti squat when under load - this may turn out to be a fingers crossed versus well engineered solution

Think I've got a neat idea to manage fine alignment in the horizontal - to fine tune the chain run ref: tight clearance to frame inside rail. I'm going to space out all the mounts so there is aprox 8mm clearance ( 4mm each side) - then i'm going to ream out the engine mount holes in the crankcase so I can insert steel stepped bushings, I can then play with the side to side of the engine to get the sprocket run spot on by adjusting the flange face thickness of these stepped bushings  - added bonus is I will also get a real tight fit between bolt OD and bush ID - what you think ? see any downside to doing this

cheers

[Gammaboy]

Scrap my comment about being a TTF1 frame, for some reason Google was showing me Mag 2s and Mag 4s when I googled Mag 3 - just googled Mag 3 again and this baby came up:

http://sinlesscycles.com/index.php/past-sins/detail/1985-harris-magnum-3

Never realised the Mag 3 was so divorced from the Mag2-4-5 evolutionary path.

[nlovien]

nice picture, thanks not so many out there to find. I suspect the mag3 must have hurt Harris financially  - they put a lot into the build of this - it's more of a complete bike package around the FI GPZ and a few GPZ900's, more like a Bimota but they didn't sell anywhere near enough of them to recoup the investment - can't make my mind up to include the alloy side plates in this present build - I can achieve the assembly without it, the gpz900 didn't have them - but they do stand out as a signature for this frame  - presently in the +/- bin

[Gixer1460]

11 hours ago, Gammaboy said:

Scrap my comment about being a TTF1 frame, for some reason Google was showing me Mag 2s and Mag 4s when I googled Mag 3 - just googled Mag 3 again and this baby came up:

http://sinlesscycles.com/index.php/past-sins/detail/1985-harris-magnum-3

Never realised the Mag 3 was so divorced from the Mag2-4-5 evolutionary path.

Seeing those pictures - timewarp! - the footrest plates do nothing for the looks - yes they serve multiple purposes but must have taken days of milling (pre CNC!) and cost a fortune! A M8 had a 3 with a 1000RX motor and candy cherry red paint - how I lusted after that bike but it went to another M8 and became a dragbike ,,,,,,,,, can't ever remember it having those plates though! I'll find some piccies later.

[vizman]

no crying when non suzuki pics get deleted

[Dezza]

5 hours ago, nlovien said:

nice picture, thanks not so many out there to find. I suspect the mag3 must have hurt Harris financially  - they put a lot into the build of this - it's more of a complete bike package around the FI GPZ and a few GPZ900's, more like a Bimota but they didn't sell anywhere near enough of them to recoup the investment - can't make my mind up to include the alloy side plates in this present build - I can achieve the assembly without it, the gpz900 didn't have them - but they do stand out as a signature for this frame  - presently in the +/- bin

Include the plates, or slightly modified versions of them as they are integral to the Mag 3's identity. Same goes for the Mag 2 but that's another story/argument.

[Dezza]

1 hour ago, vizman said:

no crying when non suzuki pics get deleted

Often counter productive when Suzuki fans can get ideas from those building bikes of other makes, e.g. AC Sanctuary Z1rs

[Captain Chaos]

the internet is not limited to OSS you know....

[nlovien]

got an opportunity here to ponder over, via the alloy side plates looks like I could bring the swing arm pivot fwd to better match the GSXR motor - too big a step change at this moment but once I get some miles under the present set up - could be a stage 2

cheers

[nlovien]

right i've been spending too much time juggling with engine position - got of the fence and fixed on a location - yi got to ride it to find out - mentioned in a previous post my hassle has been - got too much room to play with, so too many options

Started by making x 5 sets for the front lower mount  - the 5 plates varied in height and fore / aft mount position so with this + the ability to pivot the engine on this mount I've got a compromise that ticks most of the box's

1) cam cover can come off

2) ex pipe can clear the frame downtubes (just!) whilst I can also make the 1st turn without too long a run from the head

3) carb fits without hitting the back of the top rails

4) engine weight is as fwd as practical can be + the crank height is in the right ball park

5) The sprocket axis to swing arm pivot is looking like the anti squat should be tunable in the right way 

The negative is the sprocket it too far from the swing arm pivot - for me == the least negative consequence

the pics show how much adjustment just by pivoting on a set of lower mounts is available - its the fwd pitch i've gone for, benefit for it is - it brings the header nearer the frame whilst it raises the sprocket to improve potential anti squat - also pitches the weight of the head fwd - actually achieved more carb clearance aswell - added icing

knocked up a set of alloy solid bush's for the front then bolt up alloy plate to the cross brace - can now go align the rear mounts - nice to be moving fwd again 

 

[nlovien]

Well decision made ref: plates, tried a range of options for the top rear engine mount but the one that ticks the right boxes is to use the alloy plates. OK thats the engine mounts tacked up - x 3 I can adjust via removable plates - only the bottom rear is a direct fit so pretty easy to change the engine position if its not right - spent a lot of time drilling and bushing the holes in the engine and where the plates flange up - a lot of hassle because you then need to match drill everything - simple trick to achieving this is the make a centre point insert at each fit so when you offer up the matching bit - a wee tap on the insert finds the true matching centre for drilling, should be worth the effort, the overall fit is nice and tight - right - next is to fit the rear seat rails