Heat shields, Reuse or No reuse; that is the Question!

[Mark(The Miser)UK]

Here is almost all I know about heatshields. 
I'm an advocate for multiuse, but others are not.
Users can determine their own safety factor, knowing what I've discovered.
Here in the UK they can run to nearly $10 each if you go to the wrong stealership. I hear that they are becoming a traded commodity in the USA recently. Money is important to me, but not the sole criteria.
Knowing how to reform a shield can become important if you are experimenting, and changing those injectors regularly, several times a day in some instances.
These shields are a form of mild steel, and are designed to deform under injector tightening to form a seal between injector and shield, and also between shield and the head. The head side of the seal is a little ledge which actually becomes furrowed by the concentric ridges that are on each shield.These ridges help to seal, the first pic actually shows a shield with high points on the edge which has allowed carbon to fill the gap.
This is not neccessarily a bad thing as the bare metal perimeter bands show this shield is sealing.



Each shield is pretty well different on the surface due to the machining, which means each change of shield makes new  cuts into the soft ledge, and wears the head out a little more.


It is for the above reason, that I reuse the same shield in the same injector seat. It is my conclusion that reusing the same shield reduces this inherent wear.
I can't remember how many years I reused a single set in my car. If I say 5 years, then I won't be far off. I do remember though that I reused the set of 4 for 14 times before one failed. It failed catastrophically on it's 15th reformation. The other three survived, so I don't know how much life they had left in them. Here is it's pic, alongside one of it's surviving sisters  and another one of their underneath:EDIT At this point, I changed all shields for peace of mind, and ease of accounting ]

For a short time I went into panic mode when the shield failed, and felt like those 'Doom and Gloomer's'  "What if the bits had fallen into the engine"
However, studying the dead shield made me realise that the fracture was always going to happen during the 'resetting'  process, because the bearing I use pushes the inner area and tries to rip the hole bigger. If it doesn't happen, then the injector actually recompresses the inner wall together in a non fracturing direction. There only needs to be a resetting of a little over 0.5mm or 20 thou, and I use my thumb nail to check the gap between the shield and the injector by resting it on an injector see pic

I have found the 'magic of the 'Mole' grips is that they can deliver a metered pressure that can be repeated. I know previous advocates have hammered, punched, or used a drill press, but this method is convenient and accurate. As grips are not graduated, initially, under distorting and checking is the way to go.

To the pessimists who aren't members of the extreme, imminent damage clan, but don't believe that the shields can seal correctly, I say that of the 56 resettings on these 4 shields, only about 4 or less times has the shield needed a little extra tweeking to stop a leak. That equates to 1 shield in 15. I am sure that that is a better record than some single use advocates. Part of the reason is that the final tightening that distorts the shield actually creates a polished wring fit contact between shield and injector

Sometime ago I posted a little puzzle for fun when these shields had a mere 8 times reuse. They were placed alongside 4 single use and asked people to identify which are which. Only one dared humiiation, here it is again
 
I hope that this little essay explains the rationale of my position, and justifies it Especially as I have saved 10 x the purchase price of my car by doing this

Mark
EDIT:
 Re the ball bearing, I've found that about 3/8th to1/2 inch 'ball park' works well, because you don't want the clamping force to be acting too much radially from the centre of the hole, but more nearer perpendicular to the hole's circumference.

Before finding a spacer, I used an M12 19mm AF nut quite successfully...