[Diy_efi] some pictures...

[Ian Molton]

Mike wrote:
> At 09:11 PM 7/4/06, you wrote:
> 
>>I've heard a lot of what I think is probably snake-oil from turbo recon people who tend to claim the whole thing needs special balancing.
> 
> 
> It *is* correct, these things spin at rather high rates, 80,000 to 120,000 rpm
> and really do need to be balanced after re-assembly or they will howl and
> can oscillate and effectively explode at high revs but not necessarily straight
> away - it depends how long, what load etc.

Aha, you sound like someone who actually knows about this from experience :)

this is a _very_ tiny turbo, just a 3cm compressor wheel (and the 
turbine is smaller still!)

it does howl very strongly, and always has (the new one is a little 
quieter but not much). I know many do spin at 80,000+ RPM but this one 
doesnt *sound* like its anywhere close to that - the pitch of the thing 
is perhaps around middle-C to an octave higher or so. (of course, this 
could be a harmonic or something)

> Proper balancing will ensure it
> lasts the full service life, pot luck with no balancing and only assembly
> means you could be anywhere in terms of dynamics and that means it
> can fail at any time and probability suggests it will not reach a fraction
> of its service life if you expect it to provide any power and that means
> high revs...

how is the post assembly balancing done? One thing that puzzles me is 
that the compressor is onlyheld on by a tiny little nut. no spline or 
anything, so I'd have thought theres a chance it could slip slightly 
anyway, which would ruin any balancing done on it.

> Note: Even if you reassemble carefully you still must get it balanced.

Question there is 'how' ?

> Static balance might just be ok to around 5000rpms

engine RPMs or turbo? because if thats the case, Im never gonna get that 
high anyway :)