REL subs with Rowland Amplifiers


I'm having issues connecting my REL Carbon limited subwoofer to my Jeff Rowland 625 S2 amplifier via the high-level input. I connected my sub according to REL's instructions...yellow lead to positive on one channel, red lead to positive on the other channel, and floating the black wire(ground). I get minimum output when connected according to the instructions. When I use the low level input, the output is sufficient and the sub sounds great. Per my conversation with Jeff Rowland I need to ground the black wire by loosening a screw on the amplifier and connecting the black wire, but if it isn't properly grounded I may damage my amplifier. Is there anyone that owns a REL with Jeff Rowland 625 S2 amplifier or other balanced differential amplifier? If so, how do you connect your REL via the high-level input. Is there an easier way to ground the wire than unscrewing the screw on my amplifier. I just don't want to unscrew the screw and prefer another method of grounding the sub.   
ricred1
daveyf,
"If you are not connecting the black wire, I would be concerned that you are not grounding the sub." If it wasn't properly grounded I'm assuming I would hear a hum? I assume REL's Chief Engineer wouldn't tell me to do something that would damage my amp or sub.
Al,
 I initially connected the red wire to the right positive channel and the yellow to the left positive channel, with the black wire floated. Now I have it connected to the right positive channel, but the black is still floated. I was told that if I don't hear a hum, the amp and sub are properly grounded and I won't have any issues. 
@ricred1 Al’s comment above about the balanced aspect of your amp is correct. I use both a balanced output amp ( the Rowland) and a non-balanced mono amp ( Jadis), The points that Al brought up are 100% correct with the balanced design. Nonetheless, I think you are not hearing the sub in a way that would be beneficial with the set up you now describe. The sub is only receiving the mono signal from the left ( or right, depending on which mono block you use) channel.
Quick question for Al, if I were to hook up my two (2) subs to my Rowland stereo amp, would I actually utilize the summed red and yellow cables for each left and right connector and the black cable ( now two) for the ground connector ( that Jeff installed)....or leave those disconnected entirely? ( like what ricred 1 is doing).

daveyf,

Are you suggesting that there's no benefit to connecting DUAL REL subs, one to the left channel and the other to the right channel, with the red and yellow wire connected to the positive speaker post on each and the black wire floating? 
Quick question for Al, if I were to hook up my two (2) subs to my Rowland stereo amp, would I actually utilize the summed red and yellow cables for each left and right connector and the black cable ( now two) for the ground connector ( that Jeff installed)....or leave those disconnected entirely? ( like what ricred 1 is doing).

@daveyf
Connecting the black wire of both subs to the ground connector on the amp seems to me to be an ideal approach. While two ground paths/signal return paths will then exist between each of your subs and the amp (one via connection of the black wire to the ground connector, and one via the convoluted path I described earlier involving the AC safety grounds), the presumably much lower impedance of the direct path to the ground connector will result in nearly all of the signal return currents utilizing that path.

That said, regarding Richard’s situation, when he receives the second sub he has ordered, and he has the red and yellow wires from one sub connected to the + output of one channel of his amp, and he has the red and yellow wires from the other sub connected to the + output of the other channel of his amp, with the black wires of both subs unconnected, based on his latest findings with one sub it appears likely that he will have fine results. So while having a ground terminal installed on his amp is theoretically preferable, if he were to do that my guess (and it’s just a guess) is that the resulting sonic benefit would fall somewhere between minimal and none.

Best regards,
-- Al