Looks like 2011 ends in a bang.  Literally.

Yesterday morning the sun was out so I decided to take bowtie6 to work.  Unfortunately, I did not press the “Start” button long enough, the engine turned for just a split second and backfired when I let go of the button.  I’ve had this happen twice before with no serious consequences but this time, we had a problem: the “bang” cracked the intake.

There is nothing wrong with the design of the intake or with anything else.  This was my mistake 100%.  I just call this a risk of running non-factory parts.  The new intake is all aluminium; this is what the intake looks like:

We had to make this intake up because there is no room for the original plastic intake.  Actually there is, but it would have required the steering column to be relocated bigtime and it was just not worth the trouble.  The intake you see above is made in several pieces.  The flange that bolts to the head came from GM Performance Parts and is water jet cut aluminium.  This all has to be made in pieces and welded in place as such.  The four intake runners are aluminium tubing, cut and bent to fit the oval ports on the GMPP flange.  They were welded to the flange from the outside otherwise there would be a lot of machining to make the flange perfectly flat again.

The runners then were fitted to a flat piece of aluminium which made the intake side of the long plenum on the top.  There was a bead ran on the inside of that plate.  Then, the rest of the plenum was shaped and welded in place.  The seams were filed smooth and it all looks like one solid piece.  Finally a flange was made and welded where the throttle body gets bolted with four screws.

All good, except that the four runners ended up with a delicate bead around them, on the inside of the intake plenum.  So, when the backfired occurred it caused the seam on number 4 runner to assplode.  Take a look:

Obviously, it doesn’t take much for the thing to have a major vacuum leak and cause the engine to fail to run.  Needless to say, with this crack RPM’s went through the roof!  One interesting thing about all this is the ECM was smart enough to figure this out, and basically shut things down.

The outside of the runners needs to be welded again (duh!).  Had we not had the backfire, this would have not been a problem.  The seal has been flawless but unfortunately the thing just could not cope with the force of the assplosion.  As you can see, the intake has been removed and will be welded back today.  I hope to be back on the road hopefully tomorrow.

Here is a picture of engine-side of things:

Sorry for the picture being so dark, but you can see there is not much distance between the intake valves and the actual intake itself.  Not at all.  Also, the gasket seems out of place because it is not pushed all the way up against the head – in reality it fits perfectly with the intake opening.

No worries though – this should be a quick fix…  🙂


2 thoughts on “Backfire!

  1. Dick Olds

    Once again thanks for the blog. A question: As I recall reading, the LE5 ECM starter circuit remains engaged until the engine is running, regardless of how long the switch is activated. Did this feature get deleted during the ECM tuning, or was it in fact never there?

    1. bowtie6 Post author

      Very interesting question you ask.

      There is nothing in the ECM (that I can see) that has to do with the starter circuit. I did not disable anything regarding ignition.

      Having said that, what your are talking about perhaps comes into play with the Body Control Module (BCM) and perhaps the original ignition on the original steering column. On bowtie6, we are not using the BCM. We are also not using the original ignition nor the steering column. Instead, I am controlling ignition with an ignition switch, an ignition “START” button and relays.

      Bottom line, I just need to keep the start button engaged a little longer and make sure I have full ignition before letting go.


Leave a Reply