Summing up L & R line-level for dual subs

Point taken, Lewinskih01. I do tend to be very brief, so it is good that others, like Al, take the time and patience to explain in more detail.

I've been following this thread because after years of using a single balanced out from my Velodyne SMS-1 into a Y to two HGS-15s, I discovered that the SMS-1 provides unbalanced outputs for controlling up to three subs. Velodyne recommends a Y from each of those outputs into the LR inputs of each sub.

Years ago, editors at SRI would spend hours going over our papers teaching us writing skills we hadn't leaned in doctoral programs. But by the time I became director of a research center, cost cutting reduced the staff of editors.

db

indeed, Kal. However, let me point out your posts tend to be criptic and I oftentimes don't find them helpful, despite being right on (which, like in this case, I realize after someone else has gone through the trouble of explaining). This is why Al's contributions are so valuable to many here: super-knowledgeable, takes the time to look into the question, and takes the time to explain.

Also stated in the last paragraph of the first reply. ;-)

Al,

You are right, I did mistype.

BTW, your last paragraph clarified what you meant: 2 Y connectors, each going from one pre out into both subs and letting the subs do the summing. I had not picked up on that! Thanks once again!!

Horacio

Gbart,

Thanks for the suggestion. I will look into those (especially if the Lamm outs are NOT buffered).

Cheers

You're welcome, Lewinski. I think that in your last post, though, you may have misworded something:

I have posted a question to Lamm to check if the outputs are buffered. Assuming they aren't, then it seems I could try 2 Y-connectors in series so to have L&R output > summed-up to one cable > divided into two RCA prongs using one to each sub. At least I could try it.

Didn't you mean to say "are" instead of "aren't"? Otherwise you would be summing to mono the signals that are ultimately provided to the main speakers, as was said earlier.

Also, it seems to me that a connection arrangement in which summing is done by the subs, and not with y-adapters, would be consistent with the approach recommended by Mr. Geddes. Presumably what he is recommending is simply having both subs reproducing a mono signal (with the subs having different settings, as you indicated), irrespective of how that mono signal is derived. I see no reason for the means by which the mono signal is obtained to be relevant to what his recommendations are intended to accomplish, as long as the chosen means will function properly.

Regards,
-- Al

I solve the issue of driving subs from stereo preouts by using an inexpensive 2-way pro electronic crossover (ART Pro Audio, Peavey, Behringer) with separate mono subwoofer output. A single sub can be driven from the crossover mono output or two subs can be driven from the stereo low outputs.

Al,

Thanks so much for such a thoughtful post, and taking the time to do the research and the math. Truly appreciated! I'll have to digest it.

I have posted a question to Lamm to check if the outputs are buffered. Assuming they aren't, then it seems I could try 2 Y-connectors in series so to have L&R output > summed-up to one cable > divided into two RCA prongs using one to each sub. At least I could try it.

Thanks also for the tip about Transition Audio. I'm overseas, though, so that would be a second step.

Based on that info I suspect that you would be marginally ok with respect to the effects I described.

Item 8 on this page appears to be applicable to the A370PEQ, and indicates an input impedance of 30K. The parallel combination of two 30K impedances and an 86K low frequency input impedance is 12.8K. That is not ideal in relation to the 3.3K output impedance of the preamp at 20 Hz (ideally you would want the combined load impedance to be 3.3 x 10 = 33K or more), but the result would be a rolloff of significantly less than 2 db at 20 Hz, and probably less than 1 db. There would also be some small phase shift effects. I suspect that neither of those effects would be great enough to be objectionable when you finish tweaking the sub adjustments.

I couldn't find capacitance numbers for the Neotech cabling, but I did find some anecdotal indications that it is on the low side. Between that and looking at diagrams of its construction I think it would be safe to assume that its capacitance is less than 50 pf/foot, which would mean that the total capacitance of the three cables is less than around (17 + 10 + 4) x 50 = 1550 pf. That represents a capacitive reactance (impedance) of about 5.1K at 20 kHz, which is reasonably high in relation to the 245 ohm output impedance of the preamp at high frequencies.

There is always, of course, the possibility that noise or hum problems can arise when long unbalanced cables are used, but that can't be predicted.

Also, fyi, member Bifwynne has mentioned in past threads that a gentleman named Tom Tutay, I believe of Transition Audio Design in Florida, has custom made for him at a reasonable price a combination mixer/buffer for addressing this kind of situation. It presents the preamp outputs with a high impedance, sums the two channels together, and provides a low output impedance.

Regards,
-- Al

Al,

Thanks for the very thoughtful post. I'll do my best to answer:

1) input impedance of the line level inputs of the subs: Rythmik 12" with A370PEQ amps. Impedance is not listed. Input is through coax, though. Should this be a key data point I would ask the manufacturer - let me know.

2) input impedance of the main power amp: McIntosh MC275 mk IV. According to Stereophile measurements on the "RCA jacks was a high 86k ohms at low and middle frequencies, dropping slightly and inconsequentially to 62k ohms at 20kHz"

3) how long the interconnect cables would be, from the preamp outputs to each of the subs and to the main power amp: all of them are RCAs. From pre to amp: 4 feet of PUrist Audio Museaus - I don't know their impedance. From pre to subs: one coax is 10 feet, the other one is 17 feet of Neotech bulk OFC interconnect cable.

Thank you!

Some very low frequencies are, still, stereophonic in nature.
If you have two subs, why sum the channels to mono?
I guess I don't understand the theory behind this.

My subs have L & R inputs, but how should I connect them independently? From preamp left output #2 have a Y connector to go to sub # 1 and to sub #2 connecting to left input in each one, and do the same on the right?

What is the input impedance of the line level inputs of the subs? Or, if you can't readily determine that, what model subs are they?

It is common for the line level inputs of powered subs to have input impedances of as little as 10K, or even 5K in a few cases. Driving two of them in parallel would cut that value in half, from the perspective of the preamp.

According to JA's measurements in Stereophile the specified 250 ohm output impedance of the LL2 rises to 3.3K at 20 Hz, which is too high to drive that kind of a load without significant rolloff of the deep bass occurring.

Also, if the LL2's two sets of outputs are not independently buffered, which I suspect is probable, and you use a y-adapter to connect each of the two channels independently to both of the subs (letting the subs do the summing), the load impedance seen by the preamp's output stage would be reduced even further by the input impedance of the main power amp. In other words, in that arrangement the left and right channel output stages of the preamp would each be driving three loads, the two subs and the main power amp.

Let us know what the input impedance of the main power amp is, as well as the input impedance of the subs.

Also let us know how long the interconnect cables would be, from the preamp outputs to each of the subs and to the main power amp. If you were to drive all three cables from the same preamp output stage, the combined capacitance of all three cables would affect the signal that is seen by the main power amp and speakers, potentially resulting in perceptible rolloff of the upper treble if any or all of the three cables are particularly long and do not have low capacitance per unit length.

Regards,
-- Al

Thanks for the inputs. I'll ask Lamm if the outputs are buffered or not. Good tip!

Karl, you wrote "If your subs have L/R inputs, you should connect to them independently and let each sub do the summing". Yes, my subs have L & R inputs, but how should I connect them independently? From preamp left output #2 have a Y connector to go to sub # 1 and to sub #2 connecting to left input in each one, and do the same on the right?

Csontos, the reason being Mr Geddes has put much more time than me into this and that's his prescribed way...so if I want to give the approach a fair shot...it follows I should follow the prescription. It is indeed an intriguing approach.

Thanks all for the input!

If the second set of outputs on the Lamm LL2 are buffered than you can use a Y cord. If not you will have to add a small buffer stage.

You could always try a Y cord (as it will not hurt anything) and if the main system ends up in mono you will know that the Lamm does not have separate buffered outputs.

Why would you not want your subs to be in stereo as extensions of L/R main speakers?

Probably too well. Combining the L/R out of the Lamm will probably sum the output to the main amp/speakers. Result: No stereo.

If your subs have L/R inputs, you should connect to them independently and let each sub do the summing. Otherwise, you need an isolated or buffered Y connection.