KeithR's "Dream Speaker" Search

[213Cobra]

Many loose threads introduced, so let me try to unify them by stepping through the distractions.

First, let's sort through the TVC issue. Some of you characterize this as a passive preamp matter, but a TVC is not a resistive volume control. It's not the TVC you are wondering about, it's the fact that via a TVC, the SOURCE is seeing the amp input impedance, though via modification through the TVC transformers. So, is Keith's MSB DAC and his phono pre unable to deal with what they see in power amp input factors via the TVC? Let's examine that.

The 160m ARC has an input impedance of 200kOhms. The REF75SE has an input impedance of 300kOhms. The m2tech crosby in bridged mode has an input impedance of 20kOhms. The Luxman m900u has a BAL input impedance of 34kOhms. Now, the MSB DAC is completely capable of driving the impedance it sees through the TVC when driving the 300kOhm input ARC REF75SE in turn driving Devore Gibbon X. It's the best I've ever heard that troubled speaker, by far. But throw the SAME electronic chain against the 4 ohms, 87 db/w/m YGs and it collapses. Let's be serious, we are talking about just a 4 db difference of efficiency. It's not like 87db vs the 101 db of my Zu speakers. Enough between the YG and Devore, but not dramatic. The REF75SE which is THE BEST machine I've heard on Devores for instilling some coherence, great dynamics and good tonality becomes a blithering incompetent on the 87 db/w/m YG. This is not the TVC failing. It's the amp-speaker interface and lack of power. The result would show up in a closet.

Room. Alma thinks it's the room. Someone else thinks it's the TVC. Balderdash. I fully expected the 160m to sound great, on the basis of what I know from the REF75SE. I never suspected the 160m wouldn't sound fabulous. And I had nothing to do with the ARC 160m showing up in the first place. I was PREDISPOSED to like it. It was a sloppy mess with the YGs. It was not the room. Sound sucked at *any* volume. I had suggested the Bongiorno amps as a likely solution and Keith asked about the Luxman. I expressed some reservations but encouraged him to try it if Alma was willing to bring it to him. The Luxman can sound great but I noted that I found all Luxman amps in 45 years of listening to them to be dynamically disappointing. It indeed disappointed. No balls. No depth. Lean beef. Keith found redeeming qualities in its texture.

The REF75SE had less smearing than the 160m at any volume, but it had leven ess ooomph. Yet it is higher impedance than the 160m. On the other hand, the 34kOhms input Luxman had less smearing than the highest impedance ARC, which was better than the next lower impedance ARC. And the 20kOhm impedance M2tech had the most clarity and dynamics of the lot. Many contradictions in that data.

The room's ability to absorb or support dynamics wasn't the causal element. The smearing of events from the ARCs and less in the Luxman was evident at *any* volume level. The room will introduce its limits, but that wasn't the ceiling here.

Re: the Luxman: I've heard the YGs on Luxmans in larger rooms. When Philip O'Hanlon has demo'd Luxman on YG at shows in much larger and quiet rooms, I never found them to be a competent or synergistic match. I could go to Alma and listen to what Alex contends is true for him, and maybe likely not agree. MAYBE two Luxman m900u amps in monoblock config would be convincing on YG Hailey, but I don't think Keith is open to that bulk and expenditure.

So, we have ARC REF75Se proven to work with MSB driving the amp via TVC, resulting in great results with another speaker. We have improvement on YG moving up AND down in amp input impedance. We have event smearing happening at low, modest, high and blasting volume levels consistently on ARC and Luxman. And we have bass anemia except Class D amps with stiff power supplies, over 400w.

I have used TVCs on and off for over 40 years. I have only found one instance where a TVC couldn't compete with an excellent active preamp and that was with Audion triode amplifiers specifically, and there was no evident electrical explanation. Nor was there a dramatic difference. The TVC was quite good. The active preamps were discernibly better. Nothing in our work here indicates that the amp mismatches are anything other than amp/speaker interface anomalies. The room can get excited, but it's not introducing this blatant smearing at 70db.

Phil

 

[213Cobra]

I was scanning the thread to see if the very simple and dispositive issue about the low sensitivity of the ARC 160M input stage had yet been mentioned.

The subjective discussion about the sonic inadequacy of the ARC 160M is obviated by focusing on this objective specification. With double the number of tubes per channel compared to Keith's REF75SE, according to ARC specifications, the gain of the 160M exceeds the gain of the REF75SE by only .5dB. So with essentially the same gain, and with half the input sensitivity, it is predictable that the 160M will not sound as good on the YGs as even Keith's REF75SE.

The 160M's low input sensitivity (and where is what should be almost a 3dB increase in gain hiding?) should have made the 160M a nonstarter for Keith's application.

Except that the 160M wasn't really lacking gain. The system drove plenty of loudness. It just did so absent of *any* jump factor or tone. The TVC could have added +6db more gain but it wasn't needed. This was not a gain mismatch issue. -Phil

 

[spiritofmusic]

So Phil, you're saying no matter how much Keith likes the YG, it's particularly difficult to drive, and he will have his work cut out finding a great synergy that doesnt involve
1- big Watts
2- big boxes
3- big Bucks

 

[213Cobra]

So Phil, you're saying no matter how much Keith likes the YG, it's particularly difficult to drive, and he will have his work cut out finding a great synergy that doesnt involve
1- big Watts
2- big boxes
3- big Bucks

Not quite what I am saying. Look, when Keith transitioned from Zu to Devore, I did not agree but I all the same worked with him to find the right amplification. I suggested the ARC REF75SE as the best option after several were eliminated. Now he has decided to move on to YG from Devore. I thought that had the potential to further his interests but that finding the right driving amp would be more challenging. I again got involved in helping to resolve that. I loaned him a pair of amps that at least put a floor under the speakers' performance so he could make a decision about whether to buy or not. That done, I've suggested amps that I think will solve the dilemmas resulting from Keith venturing into low efficiency crossover speakers. I think this is going to be solved in a week. If I am wrong, he has another search ahead of him.

BUT, yes. 87db/w/m, gnarly crossover, three drivers = big watts. Maybe not big boxes. The m2techs sound pretty good but he can do better. The Bongiorno amps are quite compact, but heavy. Big bucks? Not if the Bongiorno amps match as I think. If not then amp dollars escalate. There are *very* few musically-credible high power solid state amps (I don't consider D'agostino among them). Expense builds quickly. This is the penalty of inefficiency. Keith isn't going to add $118,500 MSB monoblocks to his system. Not this year anyway. So the challenge is how to make an acutely (circa 1980s) inefficient speaker musically amplified? There are only a tiny handful of amps worth considering. The Bongiorno SST is the best place to start.

Phil

 

[213Cobra]

To be clear, I think the YG can be made to sound quite good, for a crossover / multi-drivers speaker, with synergistic amplification. Synergistic amplification just doesn't happen to be anything ARC or Luxman make, and that's OK. -Phil

 

[Mdp632]

To be clear, I think the YG can be made to sound quite good, for a crossover / multi-drivers speaker, with synergistic amplification. Synergistic amplification just doesn't happen to be anything ARC or Luxman make, and that's OK. -Phil

Thanks for sharing. I can tell you as as Carmel 2 owner I’ve tried three electronics combos with my speakers.

Just an FYI that I don’t subscribe to the mixing and matching of pre+amp across brands. I buy into the system approach and see the pre+amp as one component.

I know many don’t agree with that or that’s not what you’re looking for.

To that end I’ve tried Pass , Audionet and Spectral with Yg.

To my ears and Taste , Spectral was the winner. Audionet a close second.

Pass warm but too slow and couldn’t control the Carmel 2 with the precision of Audionet and Spectral.

To me that is my preferred synergy with Yg and you don’t see this combo because there aren’t any Yg dealers that sell spectral and vice versa.

One of the largest Yg dealers is also the North American distributor of Audionet electronics.

 

[morricab]

We

The first amps to try on the Hailey 2s will be the Ampzilla monos, designed by the late James Bongiorno:

https://www.sst.audio/ampzilla-1

They are quite reasonable, quite powerful, and allegedly warm, tube-like SS. In my journey so far, I feel YGs need 400 watts into 4 ohms minimum. Bongiorno also didn't believe in audiophile power cords, so they are welded in the chassis

we tried these amps long ago on Apogee Divas. They are warmish but not tubelike.

 

[morricab]

Many loose threads introduced, so let me try to unify them by stepping through the distractions.

First, let's sort through the TVC issue. Some of you characterize this as a passive preamp matter, but a TVC is not a resistive volume control. It's not the TVC you are wondering about, it's the fact that via a TVC, the SOURCE is seeing the amp input impedance, though via modification through the TVC transformers. So, is Keith's MSB DAC and his phono pre unable to deal with what they see in power amp input factors via the TVC? Let's examine that.

The 160m ARC has an input impedance of 200kOhms. The REF75SE has an input impedance of 300kOhms. The m2tech crosby in bridged mode has an input impedance of 20kOhms. The Luxman m900u has a BAL input impedance of 34kOhms. Now, the MSB DAC is completely capable of driving the impedance it sees through the TVC when driving the 300kOhm input ARC REF75SE in turn driving Devore Gibbon X. It's the best I've ever heard that troubled speaker, by far. But throw the SAME electronic chain against the 4 ohms, 87 db/w/m YGs and it collapses. Let's be serious, we are talking about just a 4 db difference of efficiency. It's not like 87db vs the 101 db of my Zu speakers. Enough between the YG and Devore, but not dramatic. The REF75SE which is THE BEST machine I've heard on Devores for instilling some coherence, great dynamics and good tonality becomes a blithering incompetent on the 87 db/w/m YG. This is not the TVC failing. It's the amp-speaker interface and lack of power. The result would show up in a closet.

Room. Alma thinks it's the room. Someone else thinks it's the TVC. Balderdash. I fully expected the 160m to sound great, on the basis of what I know from the REF75SE. I never suspected the 160m wouldn't sound fabulous. And I had nothing to do with the ARC 160m showing up in the first place. I was PREDISPOSED to like it. It was a sloppy mess with the YGs. It was not the room. Sound sucked at *any* volume. I had suggested the Bongiorno amps as a likely solution and Keith asked about the Luxman. I expressed some reservations but encouraged him to try it if Alma was willing to bring it to him. The Luxman can sound great but I noted that I found all Luxman amps in 45 years of listening to them to be dynamically disappointing. It indeed disappointed. No balls. No depth. Lean beef. Keith found redeeming qualities in its texture.

The REF75SE had less smearing than the 160m at any volume, but it had leven ess ooomph. Yet it is higher impedance than the 160m. On the other hand, the 34kOhms input Luxman had less smearing than the highest impedance ARC, which was better than the next lower impedance ARC. And the 20kOhm impedance M2tech had the most clarity and dynamics of the lot. Many contradictions in that data.

The room's ability to absorb or support dynamics wasn't the causal element. The smearing of events from the ARCs and less in the Luxman was evident at *any* volume level. The room will introduce its limits, but that wasn't the ceiling here.

Re: the Luxman: I've heard the YGs on Luxmans in larger rooms. When Philip O'Hanlon has demo'd Luxman on YG at shows in much larger and quiet rooms, I never found them to be a competent or synergistic match. I could go to Alma and listen to what Alex contends is true for him, and maybe likely not agree. MAYBE two Luxman m900u amps in monoblock config would be convincing on YG Hailey, but I don't think Keith is open to that bulk and expenditure.

So, we have ARC REF75Se proven to work with MSB driving the amp via TVC, resulting in great results with another speaker. We have improvement on YG moving up AND down in amp input impedance. We have event smearing happening at low, modest, high and blasting volume levels consistently on ARC and Luxman. And we have bass anemia except Class D amps with stiff power supplies, over 400w.

I have used TVCs on and off for over 40 years. I have only found one instance where a TVC couldn't compete with an excellent active preamp and that was with Audion triode amplifiers specifically, and there was no evident electrical explanation. Nor was there a dramatic difference. The TVC was quite good. The active preamps were discernibly better. Nothing in our work here indicates that the amp mismatches are anything other than amp/speaker interface anomalies. The room can get excited, but it's not introducing this blatant smearing at 70db.

Phil

Yes, TVCs are great when coupled to a great tube stage ...on their own they have matching issues.

 

[morricab]

I did a demo once in London for the Wilson Benesch Arc, coupled to ARC REF3se pre and Ref 210 monos. Source was Krell SACD Standard. The sound was a mess. The dealer actually blamed the big monos!! I was more than a little surprised. I came back a week later with my KR Audio.VA350i and feed with same Krell (no ARC in the chain). That sounded so good another guy in the shop asked to join me and try his music. Ultimately I went with Small Odeons but that 30 watt SET did what the 200 watt ARC monos could not...sound coherent and alive.

Modern ARC seems to need an easy load and I wonder if they are skimping too much on the output iron.

 

[DaveC]

Sounds like impedance match is not right rather than gain, this is very common.

Some years ago Burson gave away some "active cables" in exchange for reviews, it had an active buffer with USB power input, and I was shocked how many people reported improvements, and the result sounds a lot like the issue with the amp here. This was a super-cheap buffer with Canary cable parts attached to it... less than $200 retail... if this active preamp with cheap cables attached makes an improvement, this should be seen as an issue because the only way something like that is an improvement in an otherwise high end system is if something is WRONG.

The amp may not be perfect in terms of output impedance vs the speaker's input impedance too, but that wouldn't cause the issues described here, it would cause frequency response issues, usually boomy bass but that doesn't sound like the issue.

Either the amp needs to be easier to drive or the TVC needs a buffer after it, imo.

 

[bazelio]

I agree that the TVC is unlikely to be at fault here. It is not an impedance problem. A TVC or AVC is used for exactly that reason; that is, the very reason they make such good, transparent devices is they work around the need for a powered buffer in most situations. I have found one situation where a powered preamp was electrically necessary - with certain tubed phono devices that don't have reasonable output drive capability and are designed to mate with their own preamp. Zanden and Lamm phonos are likely two examples.

All that said, I'm still puzzled by the poor performance of the Luxman, which is 300W into 4 ohms. Any speaker which necessitates 500W+ in a smallish room is a real oddball to me.

 

[KeithR]

@Folsom Here is my phono - believe it's set for 68db gain right now.

http://www.octave.de/en/htdocs/verstaerker/phono_eq_2.php

 

[DaveC]

I agree that the TVC is unlikely to be at fault here. It is not an impedance problem. A TVC or AVC is used for exactly that reason; that is, the very reason they make such good, transparent devices is they work around the need for a powered buffer in most situations. I have found one situation where a powered preamp was electrically necessary - with certain tubed phono devices that don't have reasonable output drive capability and are designed to mate with their own preamp. Zanden and Lamm phonos are likely two examples.

All that said, I'm still puzzled by the poor performance of the Luxman, which is 300W into 4 ohms. Any speaker which necessitates 500W+ in a smallish room is a real oddball to me.

IMO, it's likely the source's output section.... and the sonics described sounds exactly like an impedance problem.

I said "sounds like"... without knowing more it's hard to say with certainty, a source or preamp's output specs in terms of ability to source current aren't always fully given, the specs for the octave phono give you voltage but not current capability.

In any case, I'd be willing to bet if a buffer with high input impedance and sufficient current drive were added after the TVC all these issues described would go away.

 

[bazelio]

Very doubtful. 100 ohms output impedance for the phono and high gain. It's going to have voltage swing. And please read the other anecdotal evidence to the contrary with the same source in Phil's notes.

 

[DaveC]

Very doubtful. 100 ohms output impedance for the phono and high gain. It's going to have voltage swing. And please read the other anecdotal evidence to the contrary with the same source in Phil's notes.

It's really hard to say without more info, but from my own experience it sounds like that's the issue, and not the amp/speaker interface. AFAIK YG's impedance doesn't dip below ~4 ohms anywhere and typically the big issue with a tube amp and this kind of speakers is a bass bump ~100 Hz or so. I don't think the amp is running out of current capability driving the YGs given that info...

Output impedance on the Octave doesn't give you info on how much current it can supply.

I'd see if an inexpensive buffer can be borrowed to check for sure, or just buy one of those Burson cables with the USB buffer for $100 or so, it does have gain though, so gotta be careful to match SPLs.

 

[bazelio]

You can calculate it simply. . And this system had done just fine with loaner amps Phil brought.

 

[DaveC]

You can calculate it simply. . And this system had done just fine with loaner amps Phil brought.

Ohm's law won't do it... you need specs of the output devices being used. The output impedance of the device does not equal it's ability to source current.

As I said, I'm mostly guessing, it's impossible to know without more data, but that's what it sounds like to me. I don't see any reason the amp can't drive those speakers at reasonable SPLs without major issues.

 

[bazelio]

I mean you have Vsource from the spec, I believe. Maybe I'm mistaken. But again, I'm just saying it's very doubtful to be the issue. And many people in this thread are conflating passive resistor ladder volume controls with passive TVCs. I can't tell whether or not you're doing this. But a buffer after a TVC for impedance matching makes zero sense. If the pre amplifier needs to be active, as in a few rare cases in practice, then you should just use a decent resistor ladder for volume because you don't need the impedance matching from the transformer. Buffering a TVC for "sound" might be your own perogative, but now you're just adding flavor vs a simple resistor.

 

[DaveC]

I mean you have Vsource from the spec, I believe. Maybe I'm mistaken. But again, I'm just saying it's very doubtful to be the issue. And many people in this thread are conflating passive resistor ladder volume controls with passive TVCs. I can't tell whether or not you're doing this. But a buffer after a TVC for impedance matching makes zero sense. If the pre amplifier needs to be active, as in a few rare cases in practice, then you should just use a decent resistor ladder for volume because you don't need the impedance matching from the transformer. Buffering a TVC for "sound" might be your own perogative, but now you're just adding flavor vs a simple resistor.

Maybe, but I'm in good company as I believe Tomas Mayer does exactly that!

I'm using LDR right now but looking forward to buying silver TVCs as I believe that sounds better, and while not necessary for my present source component, it'll surely have a buffer after it.

I do understand what you're saying but source output sections are variable, amp/preamp input sections are variable, and if you want to make sure your preamp works in all situations you need an active buffer... this is why there are so few passives on the market.

Here's another example, my LDR has variable input impedance and many sources sound different depending on what impedance you choose... but they really shouldn't... and no surprise my Sony HAP-Z1ES can go from driving 10k to 100k with no change in sound at all, it's a Sony and they are competent engineers. Other sources aren't designed as well, or the designer thinks it sounds best to use a design that is less robust, like tube amps where the voltage gain stage doesn't have a buffer after it... these are poor designs especially when using small signal tubes, but they exist!

If it is the speaker/amp interface that's the issue nobody has said why, besides "gnarly" crossovers, but IDK what that means, I've been to YG, seen the XOs and even have pics of them, they aren't atypical and I don't believe they behave much differently vs any other higher order xo. The impedance charts show the lowest impedance is still over 3.5 ohms and 87 dB isn't extraordinarily low sensitivity either.

 

[bazelio]

Thomas can do whatever he wants. A silver TVC that just gets buffered is pointless to me. And expensive. The ability to eliminate an entire power supply in the chain is what I find invaluable about the TVCs. To be honest, I consider the AVC (as opposed to TVC, in my case) to be a secret weapon that most people are missing out with active devices. The drive, detail, dynamics, openness, transparency, etc you can achieve is hard to match at any price.

Of course the passive transformer volume devices are difficult to productize. You never know what people are going to do with them. Like hang parallel loads off them, drive through them from weak OTL output stages. Etc. I just say, know what you're doing. They're very rewarding when you do.