
 Reproduction Parts
Reproduction Parts
  
    
      | 
      Macy's Garage, Ltd. 
      America's BEST 
      Triumph Shop! | 
  
 
 
Birth of a Reproduction Part
As compared to most vintage auto 
enthusiasts, we’ve got it pretty easy.  A very high percentage of the parts we 
need to restore our Triumphs and keep them on the road today are readily 
available.  With little more effort than a phone call (or internet order) and a 
credit card, UPS will deliver a box with shiny new parts to your doorstep in 
just a few days.  This was not always the case, and Triumph owners once had to 
hunt and dig for parts in much the same way that DeSoto collectors must today.
But did you ever stop and think about what 
it takes to bring just one Triumph part back into production?  Original tooling 
has been lost or destroyed years ago, and most outside suppliers to 
Standard-Triumph back in the day have ceased to exist.  To make any obsolete 
part available again is a major project requiring lots of time and money.  This 
is the story of what I experienced to produce new cubby box doors and hinges for 
the TR2-3B models.
First, there must be a demand for any part 
to be reproduced.  It must be something that either wears out, or deteriorates 
with age.  In the case of the cubby box door and hinge assembly, both are known 
to happen.  One of my own cars arrived here without an opening cubby box door.  
An upholstered board was bolted in place to cover the hole, but no hinge or 
aluminum door was there.  Reproductions were not available, and my search for a 
used one turned up many with broken check rod tabs, rounded out lock holes, and 
rusted hinges that wouldn’t move.  Watching eBay, I found that average used ones 
were selling in the $40-$50 range, which started me thinking about reproducing 
them myself.
At first glance, these door and hinge 
assemblies didn’t appear to be anything special.  The flat aluminum door was 
probably stamped originally, but the laser and water jet cutting technologies we 
have today would make this an easy part to have made.  The aluminum rivets that 
attach the door to the hinge had a striking resemblance to rivets I’d seen on 
airplanes, so I didn’t anticipate too many problems sourcing these either. 

Detailed drawings are necessary 
to "reverse engineer" 
reproduction parts.  
The hinge was going to be the hardest 
component. Originals were chromed steel (which rusted), but I planned to use a 
non-rusting stainless steel hinge that was highly polished to resemble chrome.  
I assumed that I could buy long lengths of stainless piano hinge, cut them to 
length, cut one side a little narrower than the other, add a bend, drill a few 
holes, have them polished, and ‘presto chango’, new TR2/3 hinges.  And this is 
the point where the wheels came off of the wagon!
The internet is a wonderful research tool, 
and by searching for ‘piano hinge’, I was able to learn enough hinge terminology 
to become dangerous.  I learned that the hinge I needed was a 3/32” diameter pin 
with 1” knuckles (the parts which wrap around the pin), and unequal leaf lengths 
(where the hinge attaches to the door and the dash).  No luck finding pre-made 
hinges in a useable size, so I went searching for a custom hinge manufacturer to 
make up a few hundred of these. 
On more than 30 different occasions I was 
told that this particular hinge could not be made.  The “industry standard” for 
a 3/32 pin diameter is a ½” knuckle, which would mean twice as many segments on 
each hinge.  I could get hinges with 1” knuckles, but the pin diameter would 
have to be ¼”, which is a huge increase.  Other than those two choices, the 
required tooling did not exist.  Since the knuckles show when the door is 
closed, both options were unacceptable to me.
Original door (top) with 
rusted hinge, corroded aluminum, and worn lock hole.
New reproduction door and hinge assembly (bottom).
Not to be discouraged for too long, I 
surmised that these hinges were originally from England, so perhaps they were 
operating under a different “industry standard”.  Using a number of U.K. search 
engines, I found several hinge companies in Great Britain, and even a couple who 
said they could produce the needed type!  I quickly shipped samples across the 
pond and smugly waited for the estimates to come in.  WOW!  Was I ever in for a 
shock.  All of the cost estimates from the U.K. were well above what anyone 
would think to be reasonable, and did not include the polishing which was to be 
done here.  The retail cost of a door and hinge was going to be around the 
century mark, and once again, an unacceptable option.  I’d been chasing this for 
several months by now, and I hated to give up and admit defeat.  I just had to 
find another way!
Next stop was to try the orient.  Many of 
our reproduction parts come from that part of the globe, so why should this 
simple little hinge be any different?  Sure enough, I found several places 
who could make this configuration with no problem.  The only question that 
remained was “how many thousand do you want?”  I was beginning to see why this 
part was not available from any of the usual sources.
And then it happened.  Out of the blue and 
by shear luck, I received an e-mail from one of the U.S. companies who’d said 
“NO” originally.  They were kind enough to suggest that I call another company 
(no web site or e-mail) and speak to “Mr. D”.  In what I believed would be my 
last futile attempt to have these made, I picked up the phone.  Yes they had the 
tooling, send the sample and they would work up a quote on a few hundred 
hinges.  I was surprised again when I received their quote a couple of weeks 
later.  They were going to be affordable!  For the first time in about 6 months, 
I was starting to believe that these parts would become available again!
Orders were placed for the aluminum doors 
and hinges, and the doors were the first to arrive.  I wasn’t completely 
satisfied with the finish on the bare aluminum, so off to the metal polisher 
they went.  When I got them back, I still had to countersink the holes for all 
of the tiny wood screws used to hold the facings in place.  Ten weeks after 
placing the order, I finally received the hinges.  They were perfect copies of 
the originals, but still had to be polished to look like chrome.  Another 2 
weeks passed before I got the shiny new hinges back from the polisher, and final 
assembly could begin at last. 
 Using 
aircraft cleco fasteners to hold the hinge and door together in the correct 
orientation, and a special aircraft rivet squeezer tool, I was finally able to 
join doors and hinges to create perfect copies of parts which have probably 
never been available since the original Triumph supplies were exhausted.  Ten 
months had elapsed since I first got the wild idea to make these parts, and 
required a huge investment in both time and effort (not to mention $).  When I 
compare my old nasty original samples to the bright, shiny and new 
reproductions, there’s no doubt in my mind that it was all worth it.  Would I do 
it again?  Just keep watching the Parts page of this web site to find out what's 
next!!!
Using 
aircraft cleco fasteners to hold the hinge and door together in the correct 
orientation, and a special aircraft rivet squeezer tool, I was finally able to 
join doors and hinges to create perfect copies of parts which have probably 
never been available since the original Triumph supplies were exhausted.  Ten 
months had elapsed since I first got the wild idea to make these parts, and 
required a huge investment in both time and effort (not to mention $).  When I 
compare my old nasty original samples to the bright, shiny and new 
reproductions, there’s no doubt in my mind that it was all worth it.  Would I do 
it again?  Just keep watching the Parts page of this web site to find out what's 
next!!!
Update 10/07/2008:
We've just completed the reproduction of TR3 
(smallmouth) Grille Reveal Moulding Joint Covers, p/n 604340 (Moss 870-035).  
The process this time took over 10 months, and 3 outside companies were involved 
in making these "simple" little parts available once again.
The first to get involved was a company that 
makes stainless steel trim for 1950's and 1960's American cars.  I 
persuaded them to make the male and female dies needed to stamp the small joint 
covers, and then to hammer out a sizeable batch of these small shiny parts for 
eventual use on our Triumphs.  From there, I took the newly stamped joint 
covers and several boxes of stainless steel screws to a precision laser welding 
company, where the screws were firmly attached  to the back side to provide 
the attachment "studs".  From there, they traveled to the metal polisher to 
have their outer surface buffed to a mirror like shine.  Special care had 
to be taken at this point to keep the edges of one part from scratching the 
polished surface of another, and when I picked them up from polishing the studs 
were punched through a couple of large pieces of styrofoam to keep all of them 
apart.  Once I arrived back at Macy's Garage with this new treasure of 
Triumph parts, I had to attach the nut and washer so that they wouldn't roll 
around in the bag and scratch the front side, and individually bag them for 
further protection.  After 10 months of persistence and frustration, I can 
now look at these shiny new parts (which I've never had for either of my cars) 
and safely say that it was worth every bit of the effort!