sfxf:
Don't supply cooling air into a mechanical room acting as a return plenum! There is plenty of cooling air from the return volume passing through the space.
I estimate this is an ~8,000 cfm unit based on a rough thumbrule of 400 cfm per ton. Also based on a standard (estimated) 20% outside air, your return volume might be 6,400 cfm. Using the formula:
Q = 1.08 * V * dT, where
Q = heat gain in BTU/Hr,
V = flow in cubic feet per minute (cfm), and
dT is the temperature rise (°F) from sensible heat,
and the Q is given as heat output of 4 KW (13,652 BTU/Hr), the 1.08 a conversion, and the cfm of return air passing through the space an estimated 6,400 BTU/Hr,
I would predict a 2°F temperature rise in the return air. Rough. Subject to factors that help make HVAC an inexact science; subject to ACTUAL heat output, not manufacturer-stated output, of the PDU (note that often a typical operating heat output of an electric piece of equipment might be 15-30% of the manufacturer-stated design heat output, which might bring us down to a half a degree or so).
There's no need to condition this mechanical room with its current air exchange rate. Let the return draw past the panel and accept that there may be a 2°F temperature rise from the return plenum to the AHU mixing chamber. This just means that the AHU will have to work a little harder to do its job in cooling mode. And it means that it will need to work less hard to heat the air during the winter and will allow you to take in more fresh air for ventilating.
If there is a temperature economizer cycle, make sure the return air sensor is inside the unit intake and not in the plenum or mechanical space.
Best of luck, -CB
