Hudson / GE / GEA Rainey would be good units and fully API-661 compliant. Like EmmanuelTop, I also am partial to Moore fans.
My most memorable pearls of wisdom would be:
* Especially if you are in a climate that can get cold, make sure to keep any automated louver actuators away from the warm air stream. They can freeze up very easily. Also, I like the fuses in the MCC - especially where VFDs are used for the fan motors - to be rated for as high a power draw (current) as you can safely get away with. Power consumption goes up as the cube of the fan speed. In combination with cold winter air, if you run into an increased cooling load for whatever reason, it's a bit of a drag to blow the fuses and lose your cooling altogether, especially in the middle of the night in January at a gas plant 200 kilometres from the nearest fur trapper, let alone oilfield supply store. You could find yourself on the header access platform shoveling snow from 45-gallon drums onto the bundle until dawn in an effort to avert an unscheduled compressor outage while waiting for the courier to arrive. (Not that I would know.)
* I see many more FD designs than ID designs. I suppose it depends on whether you want the cooler to blow or suck. Most people think that a cooler that sucks isn't an especially good cooler.
* I am inclined to recommend a pull test for any tube to tubesheet joint that is made from anything other than carbon steel rolled into carbon steel, especially if the material is prone to work-hardening. 10% tube wall reduction is typical for a carbon steel tube to tubesheet joint for optimum strength, but I have in the past been involved in testing some units where, for example with 410 SS, the optimum strength appears to occur at about 4% to 6% tube wall reduction.
* Keep your tube length to about 30 feet maximum or less to avoid too much sagging in the bundle.
* Bug screens and hail guards should be purchased and their effects should be factored into the air side performance.
* 100% UT on header box corner welds. I would also have the shop inspector UT the header box face to verify correct positioning (even existence) of pass partitions, if he or she did not witness the fabrication prior to closure.
* I personally prefer L-footed fins to mechanically embedded fins, and SA-179 seamless tubes over SA-178 ERW.
* If you ever need to do a performance evaluation on a cooler in service and are having trouble with the air side heat transfer calculations, remember to correct for the bundle face velocity by subtracting the portion of area occupied by louvers and structural framework etc. above the bundle. The sum of these areas can be 10% of the total face area that is calculated based solely on outside dimensions. Sometimes you take an anemometer and it looks like you are getting enough air when, in reality, you are getting less air but at the same or higher velocity than expected.
That's a start...
Regards,
SNORGY.
![[smile]](/data/assets/smilies/smile.gif "[smile] [smile]")
