Long response 
No plasma, I've got a gas rig, MIG and stick welders along with drill press, grinders, cutting wheels etc. The bracket pieces were cut with a cutting wheel, ground smooth and sanded. Slot was done with drill press, 5 holes connected with jig saw then smoothed with a small sanding roller.
I tack welded the pieces on the engine to ensure fit, then stick welded it all, smoothed the welds, primed and painted. I have a perverse enjoyment making stuff like this.
I've got engineering/construction ingrained from my career, so I apply the mindset where reasonable. Even a little pulley has factors to consider, so here's what I thought about....a peek into my brain....
I wanted something effective, relatively simple (considering the engine has no real provision for a left side idler), and low cost. From a load standpoint, the idler has to resist twisting and load pull from the balancer with consideration of clockwise rotation. The balancer is constantly grabbing and pulling at that pulley, lots of torque. IMO, the biggest forces would be from sudden acceleration and if the idler bearing fails, so my "design" needed to be robust enough to resist these forces. Weakest link is the bolt holding the pulley onto the bracket which is ok, very easy to change to a higher grade which I plan to do to replace the 8.8 graded chineseum bolt. Pulley center bearing is replaceable and is the almost universal 6203 if I remember right.
I decided against a common serpentine belt spring tensioner even though I could have installed a v belt pulley onto it, because I wasn't sure of the amount of tension. Serpentine is thinner and routes all over the place so my assumption was perhaps less tension. V belts have been tensioned by feel for decades, so keep it simple and don't spend a fortune. With that in mind I poked around and came across the Hayden 5950 at about $20 that was all steel and included its own mounting bracket. It fit into the left side space so I needed to come up with a mount that would work. I'd have to do this regardless.
So from that mindset I used material mass and triangular shapes for stiffness, assuming that there would still be some level of acceptable deflection. I decided on .25 plate for the mass and then looked to see how I could triangulate the bracket. One bolt at the WP housing and 2 bolts at the engine block/head gave me this triangle. The top part of the bracket is bolted and the bottom half has the two braces welded into place, its all very rigid and of course allows the idler bracket to arc back and forth to install and tension the belt.
The proof will be in the operation, but I wanted to come up with a reasonable design that addressed what I considered as possible failure points. I get it - its just a little belt pulley - but I'd rather not throw a belt or worse, have this thing get pulled into the balancer pulley. This whole (tedious) second-time-around build has been based on ensuring I've done everything I can to update/correct/check both function and appearance.
As mentioned, a long-winded answer!
No plasma, I've got a gas rig, MIG and stick welders along with drill press, grinders, cutting wheels etc. The bracket pieces were cut with a cutting wheel, ground smooth and sanded. Slot was done with drill press, 5 holes connected with jig saw then smoothed with a small sanding roller.
I tack welded the pieces on the engine to ensure fit, then stick welded it all, smoothed the welds, primed and painted. I have a perverse enjoyment making stuff like this.
I've got engineering/construction ingrained from my career, so I apply the mindset where reasonable. Even a little pulley has factors to consider, so here's what I thought about....a peek into my brain....
I wanted something effective, relatively simple (considering the engine has no real provision for a left side idler), and low cost. From a load standpoint, the idler has to resist twisting and load pull from the balancer with consideration of clockwise rotation. The balancer is constantly grabbing and pulling at that pulley, lots of torque. IMO, the biggest forces would be from sudden acceleration and if the idler bearing fails, so my "design" needed to be robust enough to resist these forces. Weakest link is the bolt holding the pulley onto the bracket which is ok, very easy to change to a higher grade which I plan to do to replace the 8.8 graded chineseum bolt. Pulley center bearing is replaceable and is the almost universal 6203 if I remember right.
I decided against a common serpentine belt spring tensioner even though I could have installed a v belt pulley onto it, because I wasn't sure of the amount of tension. Serpentine is thinner and routes all over the place so my assumption was perhaps less tension. V belts have been tensioned by feel for decades, so keep it simple and don't spend a fortune. With that in mind I poked around and came across the Hayden 5950 at about $20 that was all steel and included its own mounting bracket. It fit into the left side space so I needed to come up with a mount that would work. I'd have to do this regardless.
So from that mindset I used material mass and triangular shapes for stiffness, assuming that there would still be some level of acceptable deflection. I decided on .25 plate for the mass and then looked to see how I could triangulate the bracket. One bolt at the WP housing and 2 bolts at the engine block/head gave me this triangle. The top part of the bracket is bolted and the bottom half has the two braces welded into place, its all very rigid and of course allows the idler bracket to arc back and forth to install and tension the belt.
The proof will be in the operation, but I wanted to come up with a reasonable design that addressed what I considered as possible failure points. I get it - its just a little belt pulley - but I'd rather not throw a belt or worse, have this thing get pulled into the balancer pulley. This whole (tedious) second-time-around build has been based on ensuring I've done everything I can to update/correct/check both function and appearance.
As mentioned, a long-winded answer!
......
