Lightspeed Attenuator - Best Preamp Ever?


The question is a bit rhetorical. No preamp is the best ever, and much depends on system context. I am starting this thread beacuase there is a lot of info on this preamp in a Music First Audio Passive...thread, an Slagle AVC Modules...thread and wanted to be sure that information on this amazing product did not get lost in those threads.

I suspect that many folks may give this preamp a try at $450, direct from Australia, so I thought it would be good for current owners and future owners to have a place to describe their experience with this preamp.

It is a passive preamp that uses light LEDs, rather than mechanical contacts, to alter resistance and thereby attenuation of the source signal. It has been extremely hot in the DIY community, since the maker of this preamp provided gernerously provided information on how to make one. The trick is that while there are few parts, getting it done right, the matching of the parts is time consuming and tricky, and to boot, most of use would solder our fingers together if we tried. At $450, don't bother. It is cased in a small chassis that is fully shielded alloy, it gets it's RF sink earth via the interconnects. Vibration doesn't come into it as there is nothing to get vibrated as it's passive, even the active led's are immune as they are gas element, no filaments. The feet I attach are soft silicon/sorbethane compound anyway just in case.

This is not audio jewelry with bling, but solidly made and there is little room (if any) for audionervosa or tweaking.

So is this the best preamp ever? It might be if you have a single source (though you could use a switch box), your source is 2v or higher, your IC from pre-amp to amp is less than 2m to keep capaitance low, your amp is 5kohm input or higher (most any tube amp), and your amp is relatively sensitive (1v input sensitivity or lower v would be just right). In other words, within a passive friendly system (you do have to give this some thought), this is the finest passive preamp I have ever heard, and I have has many ranging form resistor-based to TVCs and AVCs.

In my system, with my equipment, I think it is the best I have heard passive or active, but I lean towards prefering preamp neutrality and transparency, without loosing musicality, dynamics, or the handling of low bass and highs.

If you own one, what are your impressions versus anything you have heard?

Is it the best ever? I suspect for some it may be, and to say that for a $450 product makes it stupidgood.
pubul57
Well George, hopefully you have nipped that bad idea in the bud, hard to believe what some people will do.
Selling the circuit diagram? Seriously? No April Fools joke? Never ceases to amaze me as to what some will do to make a buck.
Thanks Clio, it's just the circuit diagram they are trying to sell, no components at all.
Cheers George
Thanks George, publishing your IP has been a benefit to many. Although I suppose some of "the many" don't realize how tough it is to match those LDR's. They could sell the circuit for as cheap as they can, but the overall question is can they build it well. I appreciate the efforts of the true DIY'ers that try, but frown upon those trying to cash in on your IP.
Thos pesky little people are at it again, I've been sent a few sites where they are charging $10-$50 for the Lightspeed Attenuator circuit.
I have put it up free on DIY Forums, but they are out there trying to make money on someone else's I.P.
So here is the circuit again FREE so you can make one yourself if you up to it, and not pay for it. Hope the moderators allow it.

http://diyaudioprojects.com/Solid/DIY-Lightspeed-Passive-Attenuator/Lightspeed-Passive-Attenuator-Schematic.png

Cheers George
Hi Mcondon, if you can use the digital domain volume control within it's final 20% of full output, IE if it's 100 steps, then 80 or more you should not hear any bit stripping, if lower than 80 forget it your starting to go back to the days of good 14 bit of the first Marantz cdp which were pleasant, but no detail.
In theory if the digital domain volume control is over 80% or close to full output it should be the only thing better in transparency and honest to the source than the Lightspeed Attenuator, because you have removed another set of interconnects, and this is always a good thing.
Cheers George
Has anyone compared the Lightspeed Attenuator to a digital volume control on a digital source? If so, is the Lightspeed Attenuator substantially better? I know you theoretically "lose bits" with a digital volume control, but I want to know if others have experienced a noticeable upgrade with the LSA over going amp-direct from a digital source...Thanks.
Will be using 70 watt VAC monoblocks on them as I've brought them out of the closet for this purpose. Even took some auto detailing clay to them and added a nice coat of 3 in 1 polish. Buffed to a nice shine. cStill, 70 watts might not be enough.
Which ones? I assume the RM10 and MMGs would not be a match made in heaven, though a pair of RM10s....
Great news and report Jutl52.

BTW - I'll be testing out some Maggies soon. Looking forward to hearing them with the LSA in the chain.
I have a combined $2.5k sunk into my components between transport and speakers (ex-cables) and I think I'm done. Beat that price/performance ratio!

I accept the challenge: Allnic h1200 phono stage>LSA>Music Reference rm10, bought for total $2550 (used of course, except the LSA). I must admit that an itch has emerged for a new phono stage, but the LSA stays put--perhaps forever.

Jult52
1) With my 1.8v DAC feeding my 2.1v input amp, I find most of my listening to be in the 12-2 o'clock range.
2) I swear there was some break-in in the bass with the LSA, lasting maybe 5 hours.
3) I am a newbie compared to the very experienced audiophiles here. I think suggesting a DAC/passive preamp/high impedance amp combo is a must whenever recommending a path forward for a newbie or someone with a less expensive system. I have a combined $2.5k sunk into my components between transport and speakers (ex-cables) and I think I'm done. Beat that price/performance ratio!
Jult52

1: At 1.8k your dac does not reach the 2v Redbook CD standard, if it did then your volume on the Lightspeed Attenuator would be around 11 o'clock, but there is no problem with where it is on your system at 12-2 o'clock.
2: I don't have an answer to this, but maybe you just got used to now having bass that is tighter and goes down to 0hz (dc) instead of what you were using for a pre before.
3: Yes the Lightspeed Attenuator represents great dollar value as seen in the latest recommended components in the April issue of Stereophile, as the $25,000.00 Dartzeel NHB-18ns preamp uses a Lightspeed Attenuator volume control with a lot of switching facilities and a lot of glitz.

Cheers George
Jult52, glad the LSA worked for you, no audionervosa with this piece:) I think Clio and I have been recommending the LSA/Music Reference RM10 combo for world class sound for the cost of some folks speaker cables. If 35 watts is enough power, this is an awfully good example of great sound the real thoughtful engineering. You got yourself a very nice system there.
I want to add my very positive initial experience with the Lightspeed here. After receiving some good advice on amp input voltage here, I purchased and plugged in the LSA into my system about a month ago and am very impressed with it in terms of:

1) detail - every recording I listen to now has added lines, inflections and colors;

2) color - the diversity of sound colors I now hear is maybe the biggest single change I've experienced. I listened to Philippe Herreweghe's sonically-excellent recording of the Mozart c-minor serenade for 8 wind instruments soon after getting the LSA. It was like listening to a kaleidoscope of timbres, with each of the instruments differentiated by register, type of attack and dynamic. Amazing.

3) musicality - this is maybe the most surprising effect, for which I was totally unprepared. The performers actually seem to be more "human". There's also a "spaciousness in time", in that the melodic lines "breathe" more (boy, is writing about music hard). My working hypothesis is that this is a result of the added detail: one can now hear more of the inflections given to the notes, especially during the initial attack, and this decrease in blurring at the transient edges of sound adds ease and expressivity to the music.

So I want to thank George, Pubul and Clio for turning me onto a component that is the end-state for my system (off the merry-go-round!). A few technical comments:

1) With my 1.8v DAC feeding my 2.1v input amp, I find most of my listening to be in the 12-2 o'clock range.
2) I swear there was some break-in in the bass with the LSA, lasting maybe 5 hours.
3) I am a newbie compared to the very experienced audiophiles here. I think suggesting a DAC/passive preamp/high impedance amp combo is a must whenever recommending a path forward for a newbie or someone with a less expensive system. I have a combined $2.5k sunk into my components between transport and speakers (ex-cables) and I think I'm done. Beat that price/performance ratio!

System: Transport -> Nixon TD-2 DAC with added power supply -> LSA -> W4S ST-500 -> framed Maggie MMGs.
The LSA replaced a modded Forte F-40 ss preamp which I think is very musical but which I now believes is grainy (not its rep).
2549 and they are low capacitance. Just a simple shielded cable design. This connects the LSA to the S-30.

Roger Modjeski first introduced me to Litz as he uses this geometry in is top line cables. Litz in general yield low capacitance and superior noise rejection. I use a Litz cable between the LSA and RM-10 and EM7 amps.
Anthony do you know if the cable in that RCA is Mogami 2534 or 2549? Capacitance for both seems very low:)
I use Mogami:

http://www.audiogon.com/cgi-bin/cls.pl?cablintr&1303312915&/Mogami-RCA-1m---studio-standar

Alternately I recommend a cable that uses a Litz geometry as it is noise canceling by design. One such cable is the one offered from Q Audio.

The TelWire power cords are the best I have found for my systems needs. I use the HC cords on my amps. There are more people who use these cords than you may suspect and for very good reason. Extremely neutral and simple design.

Dealer Disclaimer: I resell Q Audio and TelWire cables.
I'll check into the shielded vs. non shielded issue. I know my Morrow's are noisier than my AU24.What shielded ic's do you recommend? Hey Clio I saw you have have the Tel Wire powercords. I have them on my preamp and CDP. They are very,very good. Don't come across too many guys who have Tel Wires. What do you think of the HC powercord. I currently use original Lesslosses on my high current applications.
They are both solid state CDPs, should be fine. I'd recommend a shielded IC between the LSA and amp. I think the Reality cables are shielded, not sure about Audience. Morrow are not shielded.
I am asking because the output impedance of the DAC or CD Player is also an important part of the passive equation. Most SS players pass the test, a tube output stage could present some issues, but George would be better on commenting on that.
Have a level one plus modded Music Hall 25.2. Gain is 2.5V. Just ordered a NAD 565 BEE. Gain is 2.2V. IC's are AU24,Straley Reality and Morrow.
You have no gain issues. You should no IC capacitance issues (with most ICs) and you should be ok with the input imepdance of the amp. The higher the better, but I think you are on that safe side there too.
Just picked up one of the preamps today on the 'gon, mkII. I will be using it with 1 meter ic's into a Krell KSA 50S amp. It has 49 kohms input impedance, output impedance .28 ohms, max output voltage 75.8V and source is 2.5V. Do these stats add up to a winning combo with the Lightspeed pre?
Of course, if your amp has its inputs in front.... Anybody on this thread have any leads on a monoblock w the inputs in front???? Paul, Ralph, anyone????
Like I said two lager than cigarette size boxes hanging off each rca on the back of the amp/s, and a third power supply unit, you need a remote volume control as they would be behind the amps, then there's the problem of the remote receiver in both units not picking up the remote controls signals because they're being block by the amp itself.
Then there's the price, definatly tipple or more, I doubt anyone would buy it after you've tooled up for all this, like I said a nightmare, but for DIY'ers worth a shot to get rid of one set of interconnects.
Cheers George
George, I appreciate that you want to keep it simple but you already have that product. If something was created pretty much like you and I have described, it would be better, but it would not be as simple. But that's OK- there would still be a market.
The only other possibility is to put it inside the actual amp at the input. It would require enough chassis space, but it can be done. This is Ralph's approach with the passive attenuator upgrade for the M-60 amps and up in his line. Unfortunately not enough room in the S-30 or Iwould have opted for it.
To reconfigure it and have it so it plugs directly into the poweramps rca's, would be a nightmare to do. As there is no standard distance between left and right rca's on a stereo poweramps and then there's monoblocks to contend with.
You would have to make two independent Lightspeeds each with their own power supply, they would need to be about the size of a cigarette packet each, this would be hanging of the amps rca's, and because it's on the back of the said poweramps, they would need to be remote volume, so double the size of those cigarette packets and you have your nightmare.
KISS KISS KISS KISS KISS

Cheers George
You know Anthony, it just might be that the LSA is what it is and the way it must be:) Some of the DIY community seems to be hell bent on "improving" the LSA - it's just possible it can't be done - without trade off - thought it would be nice to get rid of the ICs. One less thing, but as you suggest an added element of complexity.
I have the EVS attenuators and the concept is pretty sound in terms of being able to put them at the amps inputs. Not sure you could get the LSA to accomplish the same thing since each unit would probably be larger in size than an EVS or Endler making them a tight or impossible fit on some stereo amp inputs (monoblocks should be much less problematic). Not to mention I assume each will require it's own power supply or share a power supply that sits off to the side somewhere.
Reliability is golden. I certainly would not be interested in paying 3x for remote - I'm really not that lazy:) What about the idea of LDRs at the amp input - no IC? Endler Style Attenuators with the superior Lightspeed LDR attenuator.
On the famous DIY thread, it appears that have some tried the remote route, but it requires some drift from the purity of sound you get without it. I would be willing to go with a manual "Endler" approach, but it would be a bit inconvenient getting changing volume levels and rebalancing channels.
No negative impact using a remote, but no improvement in sound either, but there is a negative impact using any form of input or output switching.
It has has always crossed my mind to do a remote volume, but as it is the Lightspeed Attenuator has 99.9% reliability for minimum dollar expenditure.
Put in a remote receiver sensor circuit, a motorized pot and a good quality remote control, plus another 2 power supplies one for the receiver circuit and one for the motorized pot. And you have diminished that reliability by quite a bit. Plus you need a bigger chassis to house it all, and you have increased the price 3 fold. The cheap skates are already complaining about the $470 shipped price, can you imagine if it were to go over the grand mark.
I have a "KISS" reflection every now and again to bring me back to earth to make me realize why it is such a good product to begin with, and if it ain't broke don't ---- with it.

Cheers George
Sounds good to me. Seems like remote LDR options exist out there, but wonder if there is any negative impact on the sound using a remote feature.

I can think of one: If a remote were built into the product, the actual volume control system (as a module) could plug directly into the input of the amplifier (with separate left and right hand channels) rather than being in a box that needs an interconnect at its output.

In this way, the interconnect would play a lesser role in the resulting sound as it would only be used at the input and not at the output. IOW you would then be able to get around the traditional problem that affect all passives- the inability to drive a cable. In this system the remote would be mandatory, but it would be easy to build balance and mute functions into the system.
None really, it's maxed out because of it's simplicity, the siganl travels through one soldered resitor only with no volume pot contacts or other contacts in the signal path.
All that could be done is to 12vdc battery power it, but that is a very very small gain, most cannot hear a difference over the shipped wall wart.
The difference in supply noise between the battery and wall wart at the Lightspeed's output is microscopic using the very best test gear. "It is in effect noiseless"
Cheers George
George, this may be an odd question to ask a designer, but can you envision any future implementation that might sound even better in a passive friendly environment (no need for gain or buffer)on the LSA? I mean sonics. Sure cosmetics could evolve, perhaps some more I/O flexibility, but given that this is the best sounding "preamplifier" I have heard in my system, and that you have been sorting it out for 30+ years, is the any room for improvement that you can envision?
Yes, you want the linear regulated PS, not a switching PS. 400mA is plenty. No need to go the super beefy route with the LSA. Of course you need the correct pin adapter with center pin positive too.
I have a wallwart I purchased from Circuit-Test. It is 12VDC, 400mA, and is a linear power supply. Absent a battery PS which may sound better, is that all I need in an AC adapter?
That has to be the lowest capacitance if any cable - no?
The extremely low capacitance results from the flat construction with wide spacing of the conductors. 300 ohm antenna twinlead has similarly low capacitance.

The lack of shielding could be a problem in some setups, though. Also, the relatively high resistance and inductance of the return conductors (compared to what a shield would provide) might also contribute to ground loop issues in some systems.

Best regards,
-- Al