Been playing with the PA160 radiator here which arrived early today, and have been getting some good results with it which pretty much shows that in every way it meets its design expectations or even exceeds them slightly, but I really need to cross-correlate my own results with a known quantity (i.e. measure a radiator myself that someone like Bill has and attempt to correlate/verify). Am truly doing a poor-man's version here though - no pressure testing and no actual air-flow measurements - just measuring performance with various fans attached and what I believe to be a fairly constant heat-load (110W +/- 3W or so), and measuring the delta between water-discharge and air-flow in.
With the Tricod fan on-board, which I don't have an exact PQ curve for but I would estimate it to be around a 45CFM free-air flow fan and in many respects seems quite similar to the L1A at 7V (since I have both fans and can run the L1A at 7V side-by-side with the Tricod at 12v).
Anyways, with the Tricod at 12.0v the radiator achieved a C/W of ~0.044, and with the L1A (~69CFM rated fan) at 12v, the radiator achieved a C/W of 0.031. Given that the L1A has a rated 30dBA noise level, I would say that the Tricod is perhaps ~20dBA in comparison. These were both achieved at a ~7LPM flow rate.
Would love to see the radiator tested on a more robust and accurate testbed. Wasn't Phaestus setting himself up to measure radiators?
As for the PA160's size, when I pulled it out of the box my first thought was "This isn't going to fit", but I dry-placed it in the 3 computer cases I had here and it snugly fit in the lower front of all 3, providing close to maximal use of the available lower-front space.
One thing I noticed was that it came with barbs/fittings of extraordinary length, at least a full 5cm (2") long in the barbed section, not even including the thread or spanner knurl. Needless to say I did not use the supplied barbs, but instead used some 1" barbed length fittings, and this made for a much more compact look.
|