coleman 1/2 gallon jug holding gel packs chilled to 50f(10c) degrees. Gel packs held with both hands for 3-4 minutes in between working sets of squats.

based on the following papers: http://corecontrolcooling.com/scientific-research/peer-reviewed-scientific-studies/palm-cooling-delays-fatigue-during-high-intensity-bench-press-exercise-2/

http://www.avacore.com/sites/default/files/Grahn_Dillon_Heller_JBE_2009.pdf

the first study details using vacuum cooling for traditional resistance exercise, the second details using non vacuum cooling compared to vacuum cooling in general heightened core temperature conditions.

Numbers on non vacuum cooling:

by 0.80 +-.3°C
n= 12 with cooling only, and by 1.00 +-.2°C
n= 12 with cooling and subatmospheric pressure
Fig. 3
a

indicating 80% effectiveness without vacuum.

there is however a cohort issue:

A more detailed analysis revealed that cooling-alone treatment yielded two discrete Tes response patterns, one resembling that of cooling with subatmospheric pressure and one resembling that of no cooling
Fig. 3
b. In 8 of the 12 subjects, Tes decreased by 1.00.3°C with cooling-alone compared with 1.00.2°C with cooling and pressure differential and 0.30.2°C with no cooling
control treatment. In these eight subjects, post hoc t-tests revealed that the data from cooling alone were significantly different from control
p= 0.001 but was not significantly different from the cooling with pressure differential
p= 0.53. In 4 of the 12 subjects, Tes decreased by 0.50.2°C with cooling-alone, compared with 0.90.2°C with cooling and a pressure differential and 0.50.1°C with control treatment. For these four subjects cooling-alone and control treatments were not different
p = 0.38, while cooling-alone and cooling with a pressure differential trended to be different
p= 0.07

The non-vacuum test was single hand. Cooling of both hands looks to be around a 44% increase in effectiveness, which I'm hoping translates to non-vacuum cooling.