Sub connection to CARY SLP-05 preamp


I have a velodyne DD-12 that would like to add to the system with a CARY SLP-05 preamp to support my Thiel CS 3.7, can anyone tell me what is the best way to do it?
grandetech
Are you running balanced or unbalanced cables to your amplifier?

The SLP-05 has both for outputs. If you are running balanced to your amp, then use the unbalanced for your connection to the sub.
The input impedance of the DD-12's line-level rca inputs doesn't appear to be specified anywhere. The SLP-05's output impedance rises substantially at deep bass frequencies, to 3400 ohms at 20Hz, as indicated in Stereophile's review.

The concern that raises is that if the sub's input impedance is not high (e.g. 30K or more), the sub's deep bass response will be perceptibly rolled off.

The best approach, which would eliminate that concern, would be to connect your power amplifier outputs to the sub's speaker-level inputs. You don't need heavy gauge high quality speaker cables to do that, since what you would be driving is the input to the sub's amplifier, which undoubtedly draws a negligible amount of current compared to a speaker.

Regards,
-- Al
Almarg,

Just being notice of this!

My Sim audio Titan (5ch) power amp has an input impedance of 25K, according to what you said, is that causing my speaker rolling off the bass significantly? If yes, does it mean a high input impedance good quality amp will have a great improvement? Any suggestion?

Besides, do I need to biwire it back to from sub to poweramp according to some threads?
25K is sufficiently higher than 3.4K to be good enough. The worst case rolloff, which would occur at 20Hz, would be less than 1db, probably significantly less than 1db.

The reason I mentioned the issue is that a lot of subs have line-level input impedances in the range of 5K to 20K or so, which would be low enough to cause significant rolloff in the bottom octave if driven by the SLP-05.
do I need to biwire it back to from sub to poweramp according to some threads?
You may be thinking of the configuration in which the preamp is connected only to the sub, and the sub's high pass outputs are connected to the inputs of the main power amp. That would prevent deep bass frequencies from reaching the main power amp and the main speakers, which is sometimes done, although it puts the sub's circuitry into the main signal path. Given the high quality of your system, as well as the uncertainty about the sub's input impedance, that approach is most likely not preferable.

If you connect the main power amp's outputs to the sub's speaker-level inputs, as I suggested, you would continue using the existing cables and connections between the power amp and the main speakers, but you would add a second set of wires between the power amp and the sub's speaker-level inputs. Those wires could be narrow gauge inexpensive wire.

Regards,
-- Al
SLP-05 is running XLR to AMP.
Almarg,

Sorry I am a bit confused with the ohm thing.

You said
1. SLP-05 3.4k ohm at 20Hz, if sub input impedance < 30k ohm = bass possibly roll off. (10 times rule)

2. When Titan amp input impedance 25k < 30k ohm, you then compare to 25kohm to 3.4k directly saying its high enough (direct 1:1 compare?)

Did I miss anything? Or amp vs sub is different story when matching the impedance 10times rule?

I hope I made myself clear with these questions, otherwise, please feel free to ask me to elaborate again. Thanks very much for your help.

I will try your Sub connection method when I get back a pair of speaker cable loan to a friend.
The 30K figure I originally cited was a conservative rough approximation, cited as an example (the term "e.g." which I used means "for example").

Prior to submitting my subsequent response, in which I indicated that 25K would also be ok, I did some calculations which confirmed that.

The same principles apply whether the preamp is driving a sub or a power amp, except that in general it will be a more critical issue when a sub is involved, since main speakers will typically be greatly rolled off anyway at frequencies approaching 20Hz, where the preamp output impedance is highest.

The 10x rule incorporates conservative margins, IMO, provided that it is applied at the worst case frequency, where the preamp's output impedance is highest. Falling a little short of satisfying that rule at 20Hz, as a result of a rise in preamp output impedance caused by a coupling capacitor (as in this case) can be shown by analysis to be insignificant.

Regards,
-- Al
So say if an amp with 100k input impedance, is it always better to produce the bass response than one with only 25k? Or it is just a matter of enough, excess is useless?
Or its better to ask in a broad manner, how is an amp with 100k better than one with only 25k, of course there are other factors for overall quality, but when speaking of input impedance level how it affects people's choice? Because at least you wont need to bother too much of insufficiency, other than that, whatelse?
Once the 10x rule is satisfied, or comes close enough to being satisfied (as in this case), then further increases in the load impedance seen by the preamp will provide no further benefit. The choice of an amp should be made based on other factors, obviously including listening comparisons if at all possible.

Another point to keep in mind, though, is that if a preamp has to simultaneously drive both a power amp and the line-level inputs of a sub, and if the preamp drives both of them from the same output stage, that output stage will see a load impedance that is much lower than either of the two individual input impedances. That PROBABLY wouldn't apply in the case of the SLP-05, which has both an xlr output and an rca output, but it would apply in the case of a preamp that provides two pairs of rca output jacks that are simply wired together inside the rear panel, as opposed to being individually buffered. That is a common situation. It would also apply in the situation where a y-adapter is used.

In those situations the combined input impedance is equal to the product (multiplication) of the two individual input impedances, divided by their sum. For example, if the power amp's input impedance is 50K, and the sub's input impedance is 25K, and both were driven by the same output stage in the preamp, the preamp would see a combined load of (25 x 50)/(25 + 50) = 16.7K.

Regards,
-- Al
Impedance is usually well wihtin range with parallel downstream loads. However, adding to the complexity, I am using the fully balanced modwright 36.5 to drive my subs and mains, and the manufacturer explicitly advises against using RCA and XLR at the same time, because it will create an uneven load on the output transformer. I presume the Cary would have the same issue. So you either use 2 RCA outputs or get a high grade XLR splitter to split your XLR output (only possible if you subs have XLR inputs of course). Purist audio design makes a beautiful XLR splitter, which is an adaptor with no cable.
Edorr,

Don't you agree what Almarg suggested about feeding a pair of speaker cables from Amp to Sub speaker level input? And also why he does not suggest to do what just say. Can you explain in case you think your way is better?
I think that the problem with tapping a DD sub off the main power amp is that (AFAIK) you can't use the RTA/Room EQ function that the sub offers in that configuration - unless I'm reading this wrong (always possible!) In my book, the Room EQ is probably the single best reason to use the sub in the first place (particularly with your speakers).

Incidentally, I assume that the 2 channel pre-amp indicates a music only system. In this case, I'd think that any roll off at the bottom of the sub's passband is unlikely to be very important, since there isn't a lot of music program info that low in pitch (and it can be EQ'd back, anyway, if you really want it).

If Cary says that running RCAs and XLRs concurrently is a no-no, I'd go the XLR splitter route described by Edorr.

Just my take.

Marty
09-12-11: Edorr
... I am using the fully balanced modwright 36.5 to drive my subs and mains, and the manufacturer explicitly advises against using RCA and XLR at the same time, because it will create an uneven load on the output transformer. I presume the Cary would have the same issue.
My posts about the SLP-05 were based on the assumption that its outputs are capacitor coupled, as is the case with most tube preamps, and not transformer coupled, as is the case with your Modwright. I was also making the assumption that its rca and xlr outputs are driven by independent output stages, which is why I used the word "probably," in capital letters, when I referred to the impedance reduction that would result from driving two line level loads in parallel as "probably" not being applicable to the SLP-05.

If the Modwright drives rca and xlr outputs from a common output transformer, presumably with a grounded center tap on the secondary winding, and one end of that winding going to the rca center pin, that is a completely different story, and you were correctly advised.

Regards,
-- Al
Al, thanks for explaining.

Grandtech, probaby wise to check with Cary as Marty suggested. Regarding your other question, I'm also with Marty on using speaker level inputs. You want to use EQ with the subs if at all possible which such a setup would preclude you from doing. I am not even sure how you would use the Xover build into the DD-12 on a speaker level signal. I could be wrong here but I believe the Xover works on the line level input. This would make this a complete non-starter, since you need to apply a low pass filter.
Marty & Edorr, good points about RTA and equalization, with which I have no particular familiarity. I would expect, though (correct me if I'm wrong), that driving the sub and the power amp with parallel line-level signals from the preamp would be no different with respect to those functions than driving the sub from the power amp outputs, since in both configurations the sub has no control over the signals that ultimately get to the main speakers.

And the alternative of driving the power amp's inputs with the high pass outputs of the sub's crossover necessitates a tradeoff against the possible adverse sonic effects of putting a lot of sub circuitry into the main signal path. As well as the need to make sure that there is a reasonable impedance match between preamp and sub, which without further information on the sub we can't say with certainty.

Regards,
-- Al
I don't know how the DD-12 works, but I presume it has a separe line level input and speaker level input and the EQ and Xover can only be applied to the line level input - I could be completely wrong on this. If this is the case you cannot use the speaker level input, because your subs would run on a full range signal, which would be a complete mess with your mains.

I also would personally never run my mains with a high pass signal out of a relatively low grade Xover build into the sub. (not to mention long runs of cabling from preamp to cabling back into your main power amp.

So to me the only option would be to split the line level signal from the preamp and run parallel into mains and sub, applying EQ in the sub and using the sub xover for low pass filtering. If there are no impedance issues with this approach this would be preferable IMO. Of course, you run your mains at full range in this setup. In my (limited) experience this is not an issue, as long as you use a low crossover point for your subs.
I could be wrong here but I believe the Xover works on the line level input. This would make this a complete non-starter, since you need to apply a low pass filter.
I would feel pretty certain that the sub low pass filters its speaker-level inputs no differently than it does with its line-level inputs, aside from providing different gains for the two paths.

Regards,
-- Al
Could very well be. Lack of EQ would still be an issue, but may be even EQ works on speaker level inputs as well.
If the DD works like my SMS-1 (and I'm told that it's functionally the same), there's a very versatile (fine control of slope, phase, frequency, output level) line level high cut filter to the sub. There's also a crappy, fixed 80hz, 6db/octave low cut to the mains.

The low passed line level signal undergoes ADC, and is reflected on a video screen by the RTA as a sweep tone is generated. The user then applies up to 8 bands of parametric EQ to flatten response and level match the low pass to the main signal (response is displayed up to 200hz). I can assure you guys that this is a very effective way to ensure seamless "hand-off" from main to sub.

OTOH, if you pass at high level, I assume that you keep the crappy 80hz, 6db low cut to the mains, but...

You lose all of that good high cut control for the sub. Basically, you keep the bad and lose the good - I think.

Marty

More good news: The highcut to the mains ain't great sounding either. I use a separate active NHT X2 x-over unit for rolling off the bass to the mains and take the low pass out of the X2 into the SMS for high cutting the sub. The SMS is out of the main path entirely - a good thing IMHO.
Marty, I use the NHT X2 as well, but strictly as a low pass filter for my 2 JL113s. The low pass also runs through a SVS sub EQ unit before reaching the subs. My mains run at full range as I am a firm believer of keeping any circuitry (in this case an active high pass filter) out of the mains signal path if you can help it.

If I were to try a high pass filter for mains I would get a Marchand instead of using the NHT x2 (which has very limited slope and cross over frequency options), but I confess that part of that decision would be an irrational suspicion of a $400 device in my mains signal path doing high pass filtering (i.e. the NHT x2).

What we seem to agree on not to use the high pass filter of the sub's Xover for your mains.
Marty, take a look at the manual for the Velodyne DD subs. Its descriptions of the RTA/EQ processes are similar to what you have described, but I see absolutely nothing in the manual to suggest that those processes would not occur in exactly the same manner if the sub's inputs are at speaker level, driven from the outputs of the main power amp, compared to the configuration in which the sub's line-level inputs are driven from the outputs of the preamplifier.

For the design to be otherwise would make no sense, IMO. At a conceptual level I envision the design as simply dividing down the speaker-level input signals to line-level amplitudes, and then putting them through exactly the same circuitry that is used to process the line-level inputs.

The only difference between the two configurations, I believe, is that if the speaker-level inputs are used, driven by the main power amp, obviously the sub's high pass outputs cannot be connected into the main power amp's inputs. But we are all agreed that using the sub's high pass outputs is undesirable in most cases anyway.

Best regards,
-- Al
If the DD works like my SMS-1 (and I'm told that it's functionally the same), there's a very versatile (fine control of slope, phase, frequency, output level) line level high cut filter to the sub. There's also a crappy, fixed 80hz, 6db/octave low cut to the mains.

The low passed line level signal undergoes ADC, and is reflected on a video screen by the RTA as a sweep tone is generated. The user then applies up to 8 bands of parametric EQ to flatten response and level match the low pass to the main signal (response is displayed up to 200hz). I can assure you guys that this is a very effective way to ensure seamless "hand-off" from main to sub.

OTOH, if you pass at high level, I assume that you keep the crappy 80hz, 6db low cut to the mains, but...

You lose all of that good high cut control for the sub. Basically, you keep the bad and lose the good - I think.

Marty

More good news: The highcut to the mains ain't great sounding either. I use a separate active NHT X2 x-over unit for rolling off the bass to the mains and take the low pass out of the X2 into the SMS for high cutting the sub. The SMS is out of the main path entirely - a good thing IMHO.
Don't know why my post turned up twice. Strange.

You may well be right Al, I was just guessing. The speaker level may go thru the same processing, but intutively, I reached the opposite conclusion....I thought it unlikely that the speaker level signal would see anything but the high pass.

Maybe someone should put the question to Velodyne.

Marty
Edorr,

I tried a Marchand (used) from a local dealer, but the example I got was too noisy to keep. While looking around for a replacement, I tried the NHT on a money back trial basis and ended up keeping it.

Marty
I would second what Almarg thinks in terms of how the velodyne works on both input. It comes in more handy for me to go for connecting the subs from Poweramp since I dont have the XLR splitter while RCA route also uncertain if a good to go. If there is no risk on trying, I will test Almarg way soon. Please confirm. I will also find out if DD-15 (actually its 15") EQ works on speaker input, which I suppose it would to make sense - I mean how would velodyne think people would like to use their sub as a full range speaker on earth....
Marty, the irrational suspicion of cheap components which a share with many an audiophile, no doubt greatly contributes to the survival of the high end audio industry. Nonetheless I have been very happy with my NHT x2. The reason I bought mine is it was the only Xover I could find that allowed me to integrate discrete left and right channels with the LFE channel for multi channel applications, which is precisely what I am doing. Marchand could custom build me one, but based on your experience I am glad I stayed away from it.
My SLP-05 can output from RCA and XLR simultaneously. Simply run XLR to amp and RCA to sub. Problem solved.

If you want a fancier setup, check out my system.
Amfibius,

I am glad a Cary SLP-05 user shown up, driving the Sub with RCA means the SLP05 has to drive both the XLR and RCA simultaneously, will it cause uneven load on the output transformer and all the impedance issue as others suspecting above? Some advise to use Y-adapter for XLR, what do you think?

Richard
Richard, as I indicated above my assumption has been that there is no output transformer in the SLP-05, that it is capacitively coupled instead (as is the case with most tube preamps). You might want to ask Cary to confirm that, and also to confirm that the rca and xlr outputs are independently buffered.

As far as the impedance issue is concerned, the only way to answer that with certainty is to determine what the input impedance of the particular sub is. That figure appears to be unavailable on the web, so an email to Velodyne would seem to be in order, if you want to pursue the line-level approach rather than the speaker-level approach.

Regards,
-- Al
I just sent to Cary and Vel, thanks Al, will post it when I got answers.
Almarg,

Got reply from Vel. "The Line level input impedance is approx 15K on DD15". Is it good?

Richard
15K is too low to be ideal, given the impedance rise of the SLP-05's output to 3.4K at 20Hz. That is less than a 5x ratio. On the other hand, depending on the placement of the sub and speakers, the settings on the sub that you end up with, room acoustics, and certainly the source material, the effects of that will not be noticeable on a lot of material, and may not be bothersome on any material. Assuming, again, that the SLP-05's output is capacitively coupled, and that its rca and xlr outputs are independently buffered.

Also, the 15K value pretty much rules out the possibility of using a y-adapter on either of the SLP-05's outputs, because the combined parallel impedance of the sub and your main amp is only (15 x 25)/(15 + 25) = 9.4K

But I'd start with the power amp to speaker-level input approach first, which (assuming it works as we suspect) avoids all of those issues, and has the additional benefit of providing the same signal to both the sub and the main speakers. REL, in the literature on their subwoofers, makes a point of emphasizing the benefits that can provide in terms of overall coherency, since the sonic effects of the main power amplifier will be reflected in the signals to both the sub and the main speakers.

Good luck!

-- Al
Richard, sorry for the late response. All I can say is that there is no problem asking the SLP-05 to output both XLR and RCA at the same time. If you want an even better solution, you should consider an active crossover to remove bass frequencies from your main speaker.
No problem at all, Amfibius. I have put up a pair of speaker cable to the sub from amp tonight, it seems working quite good now. I found the xover at 53hz quite fit to my speaker, still tweaking.... just wondering what it means by tweaking those SLOPE things...., there are LOW PASS XOVER SLOPE, SUBSONIC FREQ, SUBSONIC SLOPE, PHASE, CONTOUR FREQ, CONTOUR LEVEL. Nothing from the manual....
BTW, thanks for the help from Almarg! Thumbs up!Music is so much involving now.....
Here you go Cary's reply.....Any thoughts?

Richard.



The best thing to do would be to run a pair of cables from the second pair on main outputs to you subwoofer. Do not use a Y-adapter to a mono sub input. The connections should be stereo (R/L) connections.



The outputs are capacitor coupled but the positive phase from the XLR is shared with the RCA positive signal. They are not isolated from each other. This type of subwoofer setup is done a lot with our equipment and unless there is a very low input impedance at the amplifier inputs, like 10K or below you shouldn’t have any issues doing this. The SLP-05 has plenty of gain to drive this.



Hope this helps!



Dan Wemmer

Technician

CARY AUDIO DESIGN
The Cary person provided excellent information. However, as I see it the information he provided makes the case for connecting via speaker-level much more compelling.

Since the positive polarity xlr signal and the rca signal are common, if you were to connect the xlr output to the main power amp and the rca output to the sub, that signal would have a load of 9.4K (see the calculation in my previous post), which would violate the 10x rule by a wide margin, and would also violate the 10K criterion Mr. Wemmer specified.

There would be other downsides to doing that, as well. At low frequencies the balanced inputs to your main amp would no longer be balanced, in terms of the amplitudes of the two signal polarities. There would also be a significant impedance imbalance on the two polarities, which would substantially degrade the noise rejection capability of the balanced interface to your main amp.

The good results Amfibius has obtained with that configuration are, I suspect, attributable to having considerably higher input impedances on both his amp and his sub. I don't know what the input impedance of his JL Audio sub is, but I see that the balanced input of his CAD-211AE amp is specified as 91K.

Best regards,
-- Al
Grandtech,

Are you using a video monitor to view response during the set-up tone sweeps?

If so, you can see the impact of adjusting any of those controls. As you adjust them, the change in the output of the sub will be reflected on screen. Once you see the change, you can listen for it. In my case, most changes were clearly audible, tho sometimes an adjustment that produced a visible result did not produce an audible result (that I could detect, anyway).

Fine adjustment of phase may help reduce some response anomolies prior to EQ and therefore reduce the amount of EQ you'll need to "fix" the response. Similarly, you can adjust placement of the subs while viewing response to the same end - optimal placement reduces the amount of EQ needed. That way, you'll be able to "surgically" use those 8 bands to get really smooth bass response. A mammoth PITA, but worth the effort.

The slope and level controls are essential to acheiving smoothest hand-off from main to sub. That one you will see and hear.

Marty

BTW, if the software hasn't been updated since I bought my SMS-1 (it's been +/- 3 years, so this may or may not still be true):

I would strongly advise against just using the AUTO EQ function. It does not even begin to utilize the parametric capabilities in the sub and you can do much, much better via manual adjustment w/ the video readout, per the above.

Good Luck,

Marty
Post removed