getting the most out of violin recordings

[leok]

Additions:

SET output transformers are wound for 4 Ohm load.

RF-7s and Chorus-IIs are on spikes

Air core inductors replaced ferrite core devices in the RF-7 crossover

----------------------------------------

Notes on getting the most out of violin recordings.

Playback volume is never louder than what I would hear in an acoustic setting 10' to 20' from the instrument. A violin can be very loud and at times piercing. If volume is louder than life it's probably going to hurt and it's not a realistic measure of the system's ability to reproduce the sound. I regularly attend acoustic events involving solo, several, and/or "massed" violin(s) and what I hear at those events is what I try to reproduce on my systems.

There is a tremendous variety in violin sound, from instrument to instrument and player to player. For a rich, woody tone I like Andrew Manze (especially his 1690 Amati on the Pandolfi Sonatas), Gil Shaham on "Schubert for Two," and Jascha Heifetz on the Sibelius/Prokofiev/Glazunov Living Stereo SACD. Most other recordings don't sound as rich, but in real life, they don't either. I've heard Manze at that 10' to 20' distance and his recordings are pretty accurate.

What I've done to get it "right" on my systems (somewhat in order of impact):

Amplifier: Has to be able to produce milliwatts with low distortion. Low distortion at low power has to be the focus of the amplifier's design. I've had luck with the amps listed below. This may involve a compromise. I don't like the way my no-feedback tube amps accent the Chorus-II midrange, but they're great with violin. If I want smooth frequency response I use the Crown.

Speaker: Use high quality caps in the crossover and Voltage condition them before installation (Voltage conditioning is applying, in both polarities, DC at 100V or 2/3 the cap's rated Voltage, whichever is lower). I prefer my RF-7s: I find the high end much cleaner.

Power: Use a decent Power conditioner .. one that filters out rf. I use a Belkin F5C980.

Source: It has to be clean. As a rule, garden variety op-amp buffers won't work. All my sources are class-A (op-amps or discrete). I did upgrade the power supply cap in my Philips 963SA and that was very effective. I don't know if that would have been as necessary if I had been using the power conditioner at the time. The sound is obviously never going to be better than the source so you have to pay attention to what you use. In my Sonic Frontiers "TransDAC" I've upgraded the op-amp buffers to the latest from Burr-Brown (owned by TI).

Tweaks:--------------------------

In the RF-7s I slightly reduced the resistance in the horn midband shaping circuit.

RF-7 break-in. I used a 20Hz - 20KHz sweep (reduction in power with frequency) track 18 on the Boston Audio Society "Test CD-1" and let it repeat all day, 2 days in a row (about 20 hours total). Using a sine wave doesn't require much power to get substantial driver excursion. I think the cones were moving about 1/4" at the lowest frequencies. Be careful with this, but it is effective. I had to use earplugs when setting it up and even going into the room to turn it off.

Crown D-45: stock, and remarkably close to the SET.

SET (my favorite for violin): Started with a Moondog. Removed 1st stage cathode cap. Now using Hovland musicaps for interstage. change output to parallel-feed removed the 2A3 cathode cap and used a big enough (Solen) "ultrapath" cap to go down to 10Hz. Added high value resistor (can't remember the value) across the output transformer primary to damp ringing. added grid-stopper resistors on all tube stages.

PP: This amp pretty much reflects the SET approach but in a pp config. Gain/phase inverter is a differential config. with true current source. Magnequest transformer. The transformer is usually used with EL84 tubes, but since I'm using the KT66 (or actually 350Bs right now) in triode, no feedback mode I prefered the smaller size, higher input impedance (lower output impedance).

Tripath PWM: The P6D's major tweak is removal of the input cap and general use of high quality parts. (Don't remove the input cap on other pwm amps, some additional power supply changes are necessary to make this work.)

Minimal active stages: I don't use a pre-amp. I use a resistor ladder type attenuator. 1/2M decent quality cables. However all my amps develop full power with 1Vrms in. Some amps need more drive and a pre-amp is necessary. Just don't have any more active circuitry than absolutely necessary.

No tone controls.

Signal Cables: I prefer cables that use two conductors (something approximating Litz construction) for signal and reference with the shield connected at only one end. It's a more differential treatment of the signal.

Speaker cables: Again, I prefer something approximating Litz. I'm using Audioquest "GBC Delux SST" from HCM Audio.

That's all I can think of for now. If I think of anything else, I'll edit this. By the way, while writing this I've been listening (through RF-7s, SET, Philips) to the Shaham and Manze disks mentioned above. Beautiful, natural violin sound and absolutely no listener fatigue. It can be done.

Leo

[maxg]

One thing we agree on Leo - right out of the box - is that violin can be very hard to get right. I generally try to aim for just slightly shy of being piercing - which sounds a bit bizzare but seems to work OK. A real life violin can scream when it wants to (and even when it doesn't in the hands of an amateur) and the job of a system ought to be to reproduce that sound faithfully. In fact I am not sure that would work very well for the majority of listeners. If the system molifies the top end just a little that is generally no bad thing for ease of listening.

I am, however, a little non-plussed by the list of adjustments needed to get there. Questions as follows:

1. These tweaks have been to correct the sound of the violin and just the violin?

2. What were the problems before the changes?

3. As you progressed through the changes how did you keep track of the improvements? In other words could it be that a latter change would actually have worked better without a previous change?

4. If these changes were, first and foremost, to correct the sound of the violin have there been any costs in terms of performance with other instruments? In other words - do drums (for example - any instrument that was fine before will do here) sound as good now as they did before?

Of course there is one more tweak I would have added to your list if you really want the best out of a violin recording - 1 12 inch black disk - but that's just me.[]

[LarryC]

Leo, you have carried the integration of fine musical sensibility and working away at equipment to a very high art! I agree with Max on the big challenge of getting a violin right, perhaps the most difficult of any instrument, although for me the cello has been very difficult as well. I, like you, have found that every element of the chain has to be good -- LP/CD player, preamp/amp, and the speaker system. I haven't done anything with my stock electronics or LP player, although I did have mods done to the CDP to make it closer to the LPP. In the speaker, the AK-4 networks made a major improvement in the coherency of violin sound from my K-horns. I haven't done anything to try to improve those networks, though, and I would guess your system would excel in comparison.

Larry

[leok]

Max and Larry,

The tweaks all positively influenced violin reproduction and were

always motivated by some problem with the sound. There were plenty of

ineffective changes which I removed when they didn't work. Obviously I

haven't included them in the list.

A few examples of problems and related fixes:

* Weak lower midrange on the RF-7s caused violins to sound thin. The

aggressive break in effectively addressed that. Since I don't listen at

loud levels I had never really broken the Rf-7s in.

* There were a few persistent system resonances that would cause

certain high loud violin notes to be way too loud and lacking in

timbral consistency. The resistor across the output transformer

addressed that problem. It was one of the hardest things to track down

and several other attempts had failed.

* RF filtering (power and power supply) removed a harsh an gritty

texture that would accompany louder bow sounds (when the bowing is

started or stopped for example). Sometimes there would be a pronounced

swishing sound accompanying changes in bowing direction that affected

violin and even more: cello.

All other instruments and vocals benefit from the changes, but getting

violin right was the main motivator. The bass in CSN&Y "Carry On"

just came to life after the RF-7 break in.

Both of you have better record playing equipment than I do. My

equipment is a Basis 1400, Graham 1.5 arm, Grado Gold, Creek OBH-8SE.

Violin sounds very good, but lacks the timbral accuracy of the CD and

SACD source. Also, when things get really loud and complicated there is

some obvious intermodulation that doesn't exist in the digital sources.

I'm aware that this doesn't have to be the case and one of these days I

should improve things .. probably cartridge and phono pre?

Leo

[LarryC]

All other instruments and vocals benefit from the changes, but getting violin right was the main motivator... My equipment is a Basis 1400, Graham 1.5 arm, Grado Gold, Creek OBH-8SE. Violin sounds very good, but lacks the timbral accuracy of the CD and SACD source. Also, when things get really loud and complicated there is some obvious intermodulation that doesn't exist in the digital sources. I'm aware that this doesn't have to be the case and one of these days I should improve things .. probably cartridge and phono pre?

You must have very good CD and SACD players, IMO!

That's how I've always operated, getting the toughest instruments right and the rest will follow. For my struggles with CD repro, it was solo strings for me as well, specifically cello in the Bach suites. Yo-Yo's CDs were a bear to make sound interesting, realistically delicate and subtle, and acceptably close to my TT.

I agree that changing the cart is the right focus. That table and arm are excellent and should be sufficient. As you know, I, Gary, Jazman, and others have gone to LOMC carts, which I think would make quite a difference. It sounds like that wouldn't outrun or much outrun your CD/SACD player, which would allow you to keep enjoying your CD's. I myself would defer to Jazman for specific recommendations, but do like my Transfiguration cartridge. I don't know the gain in that model of Creek, which I think is for MMs, but guess you'd have to have more like 55-61 db gain for many LOMC's. 3dz and Mdeneen seem to like their Ortofon Kontrapunkts.

Larry

[Daddy Dee]

Thank you Leo.

[Erik Mandaville]

Leo:

Some great information here.

I was wondering if you installed the grid stoppers before or after the grid-to-ground resistors. It might be interesting to try one way or the other if you haven't.

You've done lots of work on those amps; I bet they are excellent.

Thanks for sharing these details,

Erik

[leok]

Hi Erik,

By the way, Erik encouraged me to try the "ultrapath" cap to clean up

the high end detail. I liked it so much I went on to parafeed. Then it

was Eric again who suggested I try removing the 2A3 cathode cap.

Technically it made sense but Erik had to push me pretty hard to get me

to try it. Once that change is made output audio current stays

completely out of the power supply electrolytics. To do it, the

"ultrapath" cap has to be big enough to handle the lowest frequencies

the amp will deliver.

Anyway, the grid stopper: I thought I had it right on the grid

terminal. Is that what you mean? I'll check my notes. I didn't think

there was an option.

Larry,

Is the Graham 1.5 heavy enough to handle the lower compliance of a

moving coil cartridge? I am a complete novice when it comes to tt setup

so it's going to take me a while to come up to speed.

Leo

[Duke Spinner]

.

Crown D-45: stock, and remarkably close to the SET.

OMG ... LeoK ....

You said .....Crown ..!!!

Whas CRAIG gonna think ..???

[LarryC]

Larry,

Is the Graham 1.5 heavy enough to handle the lower compliance of a moving coil cartridge? I am a complete novice when it comes to tt setup so it's going to take me a while to come up to speed.

Leo

I started my Graham era with a 1.5, and ran it or its upgrades with a number of MC carts/models over time -- Benz to Transfiguration, a short stint with Clearaudio -- without a tracking problem. Subjectively, the Graham seems to have a low moment of inertia that's made for good MC carts, and it always felt very compatible in use. I'm not a maven on this subject, though, and hope others will chime in. I couldn't find a review of the Graham's tracking ability, and suggest you e-mail or PM Jazman.

Larry

[dubai2000]

Very interseting thread. I admit I have never approached music reproduction by concentrating mainly on one instrument. Perhaps that's because I don't play an instrument myself anymore - and when I did it was in my early teens. Plus I find it quite difficult to get a reliable benchmark - I mean like Leo wrote above: different violins have different timbres, so have the players and the spaces in which those instruments are played interact as well......but perhaps I just haven't given this approach enough thought.

When it comes to cartridges I no longer believe that it has to be a LOMC. I used to have a 'standard' SPU and an Ortofon Rohmann and although they might not have been set up perfectly (thinking mainly of VTA), I am convinced that my current MM carts are equal - no, in my opinion/in my vinyl chain - better performers......especially after the SME arms got rewired - a strongly recommended tweak! Now the difference between the Garrott and the MM III is more of flavour/character than of quality. Yes, the MM III remains the more refined, has a more silky top end, but the Garrott does not have to hide itself from carts costing a lot more. Admittedly it took some tweaking to get there, but now the results are pretty pleasing to my ears.

Wolfram

[leok]

Wolfram,

Can I use my Creek phono pre with the Garott P77? The load impedance

looked a bit low on the Needle Doctor website (47 Ohms) but maybe that

should have been 47KOhms.

Leo

[dubai2000]

Leo,

apart from the Holfi phono preamp I am also using the Garrott successfully on the phono input of the MX110 - thus I don't think it won't work on the Creek. According to the 'manual' that came with the cart input impedance is indeed 47kOhm.

I should add that before the SME arm got rewired I always thought of the Garrott as very musical but perhaps a tad on the warm side of neutral (and being less 'open' than the MM III) BUT that was wrong: it wasn't the cart that was less 'revealing', it was the sloppiness of the internal arm wires in the SME which caused that sound - which nevertheless appealed to me! Now the cart is more 'open'/revealing and first I thought that the warmth I had liked so much was gone, but that's not true. It's simply a more balanced sound now with less emphasis on the mids. In my experience the Garrott remains extremely musical and allows the listener to enjoy LPs without any listening fatigue.

One important aspect of the cart, though, is the ability to raise your tone arm. The manufacturer recommends that the pivot should be raised by 4-5mm. This is indeed crucial, so unfortunately that rules out arms without height adjustment.

Hope that info helps.

Wolfram

[sfellini]

Hi Leo,

Can you recommend specific Manze recordings?

I don't know this artist and could use a few pointers.

Do you know any of the details that Harmonia Mundi

uses to give his sound the richness to which you

refer?

Thanks,

Steve.

[leok]

Steve,

When I get home I'll get the Manze titles and numbers. The
"richness" I was refering to is the sound of the Amati violin (built in
1690) that he sometimes uses, and did use on the recording of Pandolfi
sonatas. The Manze Harmonia Mundi recordings themselves are some of the
most difficult I've encountered. For some reason they really illuminate
system flaws.

Leo

"Andrew Manze Portrtrait" HMX2907278 excellent sampler collection. Top recommendation.

Phantasticus "Romanesca" HMU907211 music by several 17th century Italian composers

Handel - Complete Violin Sonatas HMU907259 (very difficult recording, incredible music)

Pandolfi - Complete Violin Sonatas HMU907241 (one of the most beautiful sounding violins)

Corelli - Violin Sonatas HMU907298.99 (I like Corelli, If you want more v. sonatas, this is good)

Based on the sampler and a few of the others you'll know where you
want to go. All Manze's recordings (and performances) are great.