Scum will come off with solvents. Solids like lint, paint, bits of thread strippings, etc. that are mechanically trapped at a restriction will dislodge if the flow is reversed.

A closed system doesn't need continuous filtering. A closed system doesn't need continuous filtering. I think a good solution would be having a fairly restrictive filter but in parallel and normally isolated, say off the legs of "T"s followed by shutoffs. Every week or month one could open the valves and kink the tube between the "T"s so flow is forced through the filter. Between cleaning cycles the filter could be detached and washed.

on a rather larger scale that is what I do, startup with the filter in and then for high flow testing switch it out
not so practical in a case even with a smallish filter, fittings, valve(s)

Are there one piece valves available that can reverse flow? It would have 2 inlets and 2 outlets, and swap the flow directions.

You could have a reservoir that spins the water and has the outlet in the middle center hanging down from the top. Debris would either fall to the bottom or float to the top.

What I don't get is why the middle piece is needed. Why not manufacture the copper base to the height the middle piece reduces it to anyway. Less copper used and no middle piece. Inlet and outlet may need a little reworking but it would accomplish the same thing.

i hope it performs well, becoz i only just placed an order yesterday

Well, you can make an assumption based on the test results that JoeC published at Overclockers while reviewing another of 1A's block, the HV3 with a Eheim 1048.

Assuming the radiator assembly has a C/W of around 0.03, which would seem to be a reasonable estimate given the fannage and radiator size, we'd be talking about a block C/W of around 0.13, or about on par with the better USA blocks. Uncertain as to what the actual flow rate is though... Joe would need to test the block separately like he does the other blocks, and included the block's pressure drop, so we can draw a better indication of C/W coupled with pressure drop and flow.

The other rather stark difference is JoeC's use of 30kgs of mounting pressure, as opposed to his more standard 6.8kgs of mounting pressure with the other blocks he tests.

Why would he do that?

That seems very irresponsible to use that mounting force, what with the maximum pressure of 11KG suggested by AMD and a lower force used with the others...

Indeed. 30kgs (66lbs) would be the recommended pressure for IHS CPU's. Can get quite a substantial performance boost by increasing the clamping pressure, especially in comparison to 6.8kgs (15lbs) on bare-die CPU's with tubing hanging off the block.

Apples to oranges comparisons again, and worse, comparisons using pressures that exceed the maximum recommended bare-die pressures by almost 3x, and standard pressures by over 4x.

As Bill once said (about JoeC's practices), BTSOOM.

JoeC is a friend, he helped me
but his thinking has become a mystery to me
yesterday (?) he had a front page blurb on a product not (yet ?) reviewed
- but sold by an advertiser
so sad

hi ho

Oh, his heart is definitely in the right place. As you say, his reasoning/thinking is a mystery at times.