The first welds that broke were tig welded on there, probably with a 309. If it's pushing the pulley off maybe stiffening the pulley is the answer...

No I don't want to take it to a shop, what's the fun in that!?

 

Absolutely no fun at all! Just throwing it out since it could be cheaper than buying everything you need to DIY.

I can see where the pulley is flexing because the hub is so offset as KB mentioned, putting a ring on the back of it to add stiffness should help with the flexing. Maybe do it in 2 halves so you don't have to take the pulley off?

How about that and adding a bolt to it like the crank bolt to hold it on? I wonder if the shaft is through or case hardened. Atomic, did it feel any different once you had drilled through the outer layer with the carbide bit? Did it go any easier?

 

Yea I thought about doing a ring as well. If I were to do it in one piece the pump would have to be removed to press the pulley back on....hmm, really just need to figure out the failure mechanism...

I got everything working as it sits and reconnected the heater hoses so the truck is officially back to being a complete street truck with A/C AND heat!

It did get a lot easier once I got it started, the bit seemed to be working much better after the first little bit, so I guess that points to it being case hardened.

 

i have 309 and a tig. its a weird offer considering our geographic disparity, but ill weld it if you cover transportation.
i did this

 

Thanks for the offer, but I am going to try to figure out something other than welding and hoping for the best.

I thought about it and it seems the most likely failure mechanism is low cycle fatigue due to alternating stress. In my mind I picture everything spinning along and then at a shift, momentarily the belt tension is in reverse (blower is driving crank and everything else). This is because the blower has significant inertia and does not instantly change speed same thing with the waterpump moving large amounts of water at high speed. The HD tensioner and 8rib belt exaggerates this issue by not allowing the belt to slip making the cycle amplitudes higher. I measured the HD tensioner to be about 33% stiffer than the normal one. So basically after enough of these cycles the welds eventually crack and on the pulley shafts spinning on the shaft and works its way off.

So if this is the failure mechanism then pinning it should be a much better fix than welding. I used a 3/16" drill and it measures 7/32" so if I can find a pin that size I think I can press it in and maybe use some of that sleeve retaining compound so it doesnt back out. Or a tiny tack weld to hold the pin in wouldnt be hard.

On another note, I got a new phone (had a galaxy s3 for 3 years now). I got the galaxy s6 and the camera is freaking awesome and is the main reason I wanted it. I think its now officially better than the point and shoot camera I usually use.

This is what happened when I tried to tap the hole for a bolt. The pulley has threads but the shaft didnt have a scratch on it



I drilled it deeper to about 9/16". The pulley snout is only about 3/4" thick.

 

So a question from a pd blower newb. Is the water pump issue one that plagues LSX only, or also mod motor Fords and hemi Dodge guys or is it merely a symptom of a very aggressive build that requires an extremely stiff tensioner? I don't remember reading about 03-04 Cobra guys having wp issues? Wouldn't the electric pump from a rigidity standpoint be superior as it's just a freewheel pulley?

 

Yea im sure the electric ones would just fine because there isnt any work being done by the pulley. Part of the problem I think is the amount of work being done by the LS waterpump and then suddenly reversing belt tension then reversing again. Just a freewheeling problem wouldnt have that problem. I think this is part of the answer of why the power steering pumps dont seem to have problems, there isnt near as much work being done by the pump compared to the waterpump.

I dont know about ford and dodge guys...

 

It's all because of the offset mounting of the water pump hub. Tremendous side loading when you put a lot of load on the belt system. The center line of the load is like and inch or so behind the mounting hub and it doesn't like that!! lol The pumping action of the pump is actually very little load in the system. It's a very efficient pump to begin with and the flimsy pulley doesn't help either! lol

 

Eh, I'm not entirely convinced that's why it fails even though its true about the loading being offset. Do you have a video of the pulley under high load? That might be interesting to watch. The pulley feels like 1/4" thick steel, its pretty sturdy.

 

If you can get me the pulley dimensions and offset, I can run a quick FE analysis with an arbitrary load so we can see the deflections that occur.

If the waterpump doesn't have much of a load it, does it care where the tension is coming from? Also, does it matter which of the other pulleys is driving the system? Wouldn't the water pump pulley see the same thing (belt spinning in one direction) regardless if whether the crank or supercharger are driving it? Did you mean when the crank is slowing and supercharger is still trying to continue (inertia)? If so, wouldn't there be a spike in temsion there not necessarily a reversing? The tension along the belt should be pretty uniform on either side of the pulley minus the resistance the water pump provides right? If my thinking isnt completely off, the deflection (especially with the tension spike when during a shift) is enough for pulley to walk slightly with each revolution until it can fall off or be pushed off in the deflected state. The belt does have a bending moment perpendicular to the shaft axis, but I would think it needs to walk some before being pushed off.