Which subwoofers have at least a 20k input impedance as required for ARC preamp

Thanks for all the input. Actually the issue is very complicated after I spent about an hour on the phone w ARC yesterday. In fact we didn’t really come up with a comfortable combination of amp(ref 75se) and sub.
Running balanced into the amp and unbalanced into the sub is bad regardless of the subs impedance spec such as JL’s 50k spec.
Unless the sub is true differential balanced it’s really single ended. Don’t know about the JL.
The solution might be balanced out to a sub with true balanced inputs and a reasonably high input impedance. Good luck on that one.
To try to understand all of this a potential ARC buyer should call them. They were working the math and it was complicated with no real conclusion.
Obvious answer is ARC preamps are not really designed to drive subwoofers.
Almost forgot. He thought the REL speaker level connection might work.
Im sorry that I’m not technically astute enough to better explain but I personally won’t be buying an ARC Ref pre due to this and the fact that ARC indicated that the 6550 tubes in the power supply of a Ref preamp could be very problematic should one fail in a bad way. Then there’s the amp output tubes.
Regardless of what is posted here there is apparently is a lot of shipping back and forth to ARC for service. For me there’s too much great SS now to suffer the grief unless you’re lucky.
My apologies for the length of the post.

Thanks for the follow-up, @4425. Running balanced into the amp and unbalanced into the sub will result in a very slight imbalance between the impedances of the two signal lines in the balanced signal pair, relative to ground. That will result in a slight degradation of the ability of the amp to reject common mode noise that may be present at its inputs. I’d be surprised, though, if they would consider that effect to be great enough in degree to be audibly significant, since those impedances will be dominated by the output impedance of the preamp, which is vastly lower than the 50K unbalanced input impedance of a JL sub.

Or perhaps they were envisioning the possibility of ground loop issues. But that would be a possibility, depending on the specific designs, even if the unbalanced outputs were being used to drive a power amp and the balanced outputs weren’t used at all.

Those are the only two reasons I can think of that might lead them to recommend against driving a sub having suitably high input impedance with the preamp’s unbalanced outputs, while driving the power amp balanced.

Tketcham’s post reminds me that another member here who uses ARC Ref electronics and a sub had a tube buffer stage custom made for him some years ago by Tom Tutay of Transition Audio Design in Florida, which worked out very nicely and didn’t cost a great deal (well under $1K if I recall correctly). I don’t think Tom has a website but his contact info can be found via Google.

It sounds like you’ve settled on a course of action to which none of this applies, but I’m mentioning these things in case others find themselves in a similar situation.

Best regards,
-- Al

Al: I wish I had your knowledge. I’m too simple on audio and don’t want to go a little funky or compromise in any way if I was buying uber ARC products. That’s my problem. 

Is replacing the preamp a possibility? There are other balanced tube preamps that have no problems driving loads like that.

I have an LS-28 and have ran into the same problem. Switched to a REL sub using the high level input (driven from the speaker outputs) and it works great.  I’m running a Rogue Stereo 100 amp with KT-120s.