Ref 3 Impedence Mismatch HELP please
Sgr, FWIW, you don't have to go to transistors to run 25 feet of balanced cable to a crossover. You just need a preamp that can do that. Our preamps can drive over 100 feet of cable, and they are all-tube; plus they can drive input impedances as low as 600 ohms with no problem- 10K is a piece of cake :) I have held the opinion for over 25 years that if a preamp or other device like a DAC has an XLR output, then it should also support the associated balanced standard. The standard is simple: in addition to proper pin connections on the XLR, it requires that the device be able to drive a low impedance, 600 ohms being typical (meaning that 10K should be no worries). It also requires the the ground (shield) not carry any signal current (IOW, the signal occurs entirely between the inverted and non-inverted phases, ignoring ground). Most modern preamps in high end audio with balanced outputs don't support the standard, being unable to drive low impedances and also using the ground connection as part of the signal (which increases susceptibility to cable problems). ARC is one of them. This causes confusion, because with balanced operation you are supposed to be able to use long interconnects without loss. This will not be true if the preamp does not support the standard! To my knowledge, we are the only tube manufacturer that supports the balanced standard (BTW I would love to learn that I am wrong). OTOH there are a good number of transistor preamp manufacturers that do support the standard. If there is ever any question, just ask 'What is the longest cable this preamp will drive?' If 50 feet is out of the question, the preamp does not support the standard. |
Hi, Thanks for all the suggestions. There is indeed a bass roll off using the REF 3 with the Xilica Crossover. The spokesman at ARC, confirmed there would be a roll off in the treble and bass, with a loss of dynamics. I am using a 25 foot balanced interconnect between preamps and crossover so that I could keep the source and preamp from being on a rack inbetween the speakers. I have recently tried two different solid state preamps in my system and this problem has been confirmed, the other preamps sound much better than my ARC. Darn, I like ARC preamps and never suspected there could be a problem. This does solve one problem though, why wasn't I getting great bass and treble response from my system. My solution is to look for a great solid state preamp to install in my system. Right now I'm using a friends Gryphon Sonata Allegro and it really has taken the system to a new level. If I could find another of these that's what I'd purchase. Darn, they are not imported into the states any more. Right now I'm considering a Pass Labs XP 20 preamp. Anyone else have any more suggestions? Thanks, Steve |
Herman, Bifwynne's suggestion is a good one- Tom is a good guy and knows his stuff. If you install larger coupling caps, the caps themselves will color the sound to a greater degree. The bottom line here though will be that if you use this preamp with this crossover, you are likely to wind up with some sort of coloration or artifact- this is a tricky problem to get around. |
Sgr, pull my old OPs that considered a very similar problem, in some ways even worse than what you have. I used to own a Ref 3 and now a Ref 5. A fellow by the name of Tom Tutay built a custom impedance buffer for me. Admittedly, it is principally for the Ref 3/5's Main #2 output that powers my sub, not my main outputs. I would call Tom and see if he can custom build something for you. My only concern is that inserting an artifact into your main signal path may cause a different set of problems. But do give Tom a call. He is top drawer. His number is listed on my OPs. |
Any idea how much that would be?http://www.stereophile.com/content/audio-research-reference-3-line-preamplifier-measurements: The output impedance was also to spec., at 635 ohms balanced and 326 ohms unbalanced in the treble and midrange, but rose to 1437 ohms and 625 ohms, respectively, at 20Hz.The OP's processor has only balanced inputs. A rise in preamp output impedance from 635 ohms in the mids and treble to 1437 ohms at 20Hz is non-ideal in relation to a 10K load (assuming that spec is accurate), but on the other hand doesn't seem likely to cause more than a very slight bottom octave rolloff. Not sure why there would be issues with the highs, as the OP reported, unless the interconnect cables are particularly long and do not have low capacitance per unit length. In that situation, perceptible upper treble rolloff could result from the interaction of cable capacitance and preamp output impedance. Best regards, -- Al |
. the output impedance of the ARC goes higher as the frequency goes lower. Good point, any idea how much that would be? I have no idea what size caps they use. There is no good way to fix this- either Bigger caps? I realize that most would be reluctant to do surgery on such an expensive pre but if not then why not just up the value? . |
The problem here is that the output impedance of the ARC goes higher as the frequency goes lower. If driving a high impedance this is no problem, but a middle impedance like 10K (less than 2000 ohms being 'low impedance'...) is problematic as the coupling caps at the output of the preamp are rolling off the bass with the reduced load at the output. There is no good way to fix this- either you somehow find a way for the preamp to drive a higher impedance or change out the preamp. We got around this problem by direct-coupling the output, so there is no bass roll off. Herman is right, there is far too much variance in high end audio so there is not likely to be much in the way of standards, about all you can expect is that tube amps will likely have an input impedance of 100K or higher, whereas its likely that with most solid state components 10K is more likely. |
.. So 600 out to 10K in? Lower than I would like but not a gross mismatch as it is about 1 to 15. Most recommend a minimum of 1 to 10. With your system it seems a shame to add a buffer, why not get a Krell preamp? I doubt you will find a tube line stage with lower Zout. If you have more than enough gain you could try some step down transformers between the pre and Xover. Dave Slagle at Intact Audio could wind you a custom pair to meet your needs. If you can DIY or know somebody who can the Zin of the crossover is probably set by a single resistor for each channel which could be changed. You might contact them to see. As for a standard I doubt that will ever happen. . |
