Hi Andrew.
I'm always amazed by the Wren group's kit balancing technique, although I still maintain there's nothing better than true dynamic balancing on a machine. Your motor's reliability does speak for itself however, as do many more. You also have a very good point about the cost of the kits vs scratch, although Jon's first engine was completely scratch built with the exception of the KKK 2038 compressor only. Amazing little engine... ~$130 for the compressor, plus cost of inconel, stainless and Al bar stock. He even made a primitive ECU for it.
I just picked up one of the balancer packages out of the UK (shown below), I'm currently setting up the shaft mounting hardware and wiring up the motor/sensors. This is the exact same package PST and TJT use in production and I've heard it can pick up a rather thick felt pen mark on the shaft if set up properly.
We'll see, looking forward to see just what level of repeatable performance this unit is capable of.
Next time Marc Thomson heads out this way, we are thinking of tearing down his RAM-500 just to see how close he got it. I think he's been thought the Wren balance instructions about 5-6 times on that engine and it sounds like he's got it running really nice. This was one of the engines released with the improperly ground turbine wheel so he had to get a replacement from Wren and fix it himself when RAM went under. He's had a bit of a dark cloud over his head relating to his turbines, great to see he's starting to get better results.
As for machining, only a few parts really require high tolerance. Depending on how you make it, the diffuser really isn't that big a deal, it just takes way too much time on to machine. I made 4 of them before I got one right, each taking ~30-35 hours on a manual mill. On top of that, I screwed up one of them on
the very last step I don't need to tell you how much colorful language came out in the shop that day...
If you were to make the shaft tunnel with bearing bores in both ends, you'll need to be careful. Most of the tunnel is made in a single 'clamping' but you can't easily do both bearing bores in a single operation. You'll need to ensure that the runout and concentricity are as close as possible when machining the other bore, or design in some sort of auto alignment system. I stuck to the old design in my KJ with the bearing press fit into the diffuser, which has its own inherent problems if its not done just right.
IMO, making a twin IC engines quite an achievement! Got any photos to share? I once read an article in RCM about a guy who made a pair of 4-cycle twins. Even worse, he crashed on the first flight because he forgot to run his RX antenna, ouch. The engines were very impressive.
Thx,
Kelly
PS: How's your Pizza box flying? Mine's almost done!