One piece block/head

A friend and I were discussing how many engines tend to blow head gaskets in high power/boost situations, and wondered if anyone has ever made an engine with a one piece block/head. I'm sure there is a reason this isn't done, but it seems like it would produce a stronger unit (more rigid, stable, no gaskets to blow out).

It would be extremely difficult to setup a valvetrain in an engine through the crankcase ;) Other than messing with the valves, I don't really see any reason to have a separate head/block.

What if the valves were put in before the rod/piston assembly? Obviously the engine would need to be out of the vehicle, but I see no reason why it couldn't be done. I see a major problem with setting up the ports/valve seats, but again, no reason why it couldn't be done.

A friend and I were discussing how many engines tend to blow head gaskets in high power/boost situations, and wondered if anyone has ever made an engine with a one piece block/head. I'm sure there is a reason this isn't done, but it seems like it would produce a stronger unit (more rigid, stable, no gaskets to blow out).

This is actually very commonly done....... with small 2-stroke engines. My chain saw, weedwhacker and leaf blower all employ this technique.
It's obvious for 2 strokes since it reduces manufacturing costs and increases reliability. As 2 strokes employ ports in place of valves, there is no problem with valve gear.

This technique was also used occasionally in the very early days of automotive production, (pre-WW1). I have not seen such engines disassembled, but I think the valves were attached to cast-in ports, in an L-head or T-head arrangements.

Cast in ports wouldn't fly with today's emissions standards though. Machining the valve seats would be the hardest part of the whole operation. What would take an hour on a normal head would likely take forever on the cast engine assembly. Cost and manhours per unit would be the biggest limiting factors. Now if they could come up with some sort of port with a rotating valve that is actuated by a solenoid, that might be super feasible and cheap :)

Cast in ports wouldn't fly with today's emissions standards though. Machining the valve seats would be the hardest part of the whole operation. What would take an hour on a normal head would likely take forever on the cast engine assembly. Cost and manhours per unit would be the biggest limiting factors. Now if they could come up with some sort of port with a rotating valve that is actuated by a solenoid, that might be super feasible and cheap :)
You could do it with sleeve valves. These were quite common decades ago
http://en.wikipedia.org/wiki/Sleeve_valve

After reading a bit, another problem would be inherent and probably nearly unfixable, oil consumption. That wouldn't work at all with today's emission standards. The benefits of a sleeve system seem to be pretty damn nice though.

I think it is Ferrari (612 Scaglietti) that uses a hardened valve seat that is pressed into the head. They also press in the valve guides after shrinking them first by supercooling them with liquid nitrogen. Then they warm back up, expand, and lock themselves into place from the interference fit. This method would reduce the need to machine the valve seat...just clean-up from casting to create a flat landing for the ring. But machining could be done as long as the valve angle doesn't exceed what is achievable down the bore of the cylinder...the walls could be protected from tool nicks with plastic sleeves. It would take careful planning, but it looks achievable with today's CNC capabilities.

Casting may be another story, but the lost-foam method (employed on Saturns and the TrailBlazer/Envoy I-6) may be the most ideal for starting out to incorporate all the passageways needed for oil and coolant flow.

I don't think it is impossible, I just think it would be a cost vs profit issue. A separate head and block is more feasible and cheaper as of right now. It isn't 100% reliable on every car but i'd say it is in the 90% range for most cars. The other problem is what do you do if you bend a valve or do some sort of damage to the valvetrain? Replace the whole engine for an easily fixed bent valve? Or spend XXXX$ for full engine disassembly to replace a valve and maybe a seat.