Wow, it is worth to live in Boston and have a day of miserable rain.

I really like what I was getting with CD and I have a few very interesting results coming that should be even more exiting, at least one of them is more defiantly is very exciting – the another revision of my Bidat. I will post about it as time came by… However, today I have no Bidat and nothing can properly run my TL0transport. The Pacific does very well but it too cumbersome and Lavry has no proper upper bass harmonics with my TL0.

However, it was miserable rain and I was officially bored. I was looking at my Lavry DA-924 as asked it why you the bitch do not run well from my CD transport.  The upper bas and lover mid is kind of weak region of CEC TL0 and it was not right. But it was rain and the wondering hands were itching…. I figured that Mr. Lavry will be very pissed but to piss people is my specially… and I went for it.

It took 2.5 hours and the result… why the hell I did not do it before? I just got a phenomenal uppers bass from Lavry DAC + a few others very-very interesting effects. I will not finales my comment and will do them in month (I have good experience with some parts I put in the unit and I know how then change sound with time – in 30 days it will be over) but at this moment (and I know how those parts should sound juts installed) the result are spectacularly good (and not only in upper bass but in HF as well)

Ironically it was do very little. I opened the DAC and learned that Lavry use common standard recovery diodes in unit PS of the unit. The DAC used 8 x1N5401 and 2 x1N4002 diodes. They are 3A 100V. I put in 10 Schottky 31DQ09 90V 3.3A with no reverse recovery and replaced all power caps with Nichicon Muze KZ. (The only cap that I respect for SS)  How simple it is but how funny the result! The upperbass made sound so LARGE that it is actually not funny, evens though the small size never was Lavry problems. However, the most important I got successfully lost that mean HF Lavry’s extension that I always hated. Now Lavry sound almost like Bidat at HF….

Sure, the Muze KZ will sound all, over for the next week and in a certain degree month and I will see when it settles down. How good would it be if the engineers of Dan Lavry caliber sometime listen what they do and invest $3.45 into this unit to get THIS results.

Very exiting Sound so far.

I never hit such a huge jackpot as I just did with the Lavry DA-924. The most funny in that even that I did it absolutely blindly. I had some minor beef with DA-924 sound: upperbass/lower midrange and the slightly too much extended HF (very very clean HF however). I even experimented with HF passive filters after the DA-924 to slightly roll it off (as Pacific did). I did not dream that my minor parts swapping in the DAC PS will address both of the DA-924 in such a perfect way. However there is more to it.

DA-924 did very fine imaging and did OK all those “space” tricks. It was not the best DAC I have seen in term of space but it was still good. Well, if it was able do not screw up the “space”  of Sansui TU-1X running live broadcast then what else you can expect from a pieces of electronics – it does not get more “space” then that.  However, with slightly coming down in HF extension (always good for digital in my books) and obtaining the necessary weight in lower mid range the DA924 suddenly throw such a firework of “space” that it is shocked my believe in capacity of DIGITAL and this DAC particularly to the very foundation.

This Friday I left for work I setup my recording system to record the Friday matinee live broadcast from my local Symphony Hall. Conductor Emeritus Bernard Haitink led Boston Symphony with Bartók's Piano Concerto No.3, played by András Schiff. However my major objective to record was the second part – the Schubert's Symphony in C, The Great. I got home when WHRB just finished the live airing. I played the recording. The Schubert's was quite good. It was not the best I have heard but it was live and you can’t listen the Schubert work in studio sound – it just does not work for me with the “The Great”. BSO, screw up here and there but Haitink managed to get out of BSO the “large” sound that is VERY difficult for BSO. The BSO’s Symphony Hall is basically is a large garage and BSO hardly ever sounds “wide” and “large”. I know where microphones hanging in Symphony Hall and I know how BSO sounds during the live transmissions. Haitink did made BSO to sound “wide” and “large”. It was good Schubert's The Great… I did listen if and then later on on Friday I modified the DA-924 power supply.

I listened the “new” DA924++ driver from my CD transport and was very-very pleased with what I heard. Then, today, I decided to see what will be if feed the DA924++  with my FM broadcasts recorded files. The Schubert's The Great was already loaded in my WaveLab and I played it. Honestly I did not expect that is will be so stunningly different. It was not juts “wide” and “large” but it was intolerably spacious and enormously large.  It was so gigantic and colossal that the first I did was … stopped play, take my phase tester and re-check the polarity of the drivers. Of cause the drivers were fine but Macondo in it’s current position juts was not able to built a proper image of this size. I know that Nichikon KZ do very funky things with imaging for the first 30 days (that I expected) but not like that! The Symphony Hall’s garage got converted into open Concertgebouw stage and the sound of BSO became so much spread in huge space that it more sounded like Hector Berlioz conducted his infamous 1200-musician orchestra with 600 violinist and 250 cellists. The Schubert's work sound like from no-were, like the BSO just flew up and were playing in air – and it is all with absolutely defined imaging and very fine caring about the relation between the separate instruments and the sections. It was just simply amazing!

Well, if it goes like this down the road I will be forced to make some decisions about the DACs I use… Meanwhile I took some measures enabling Lavry DA-924 to run 24/7 (like my Bidat does)…

Dear Romy,

It may be worthwhile to check what transformer is installed in your DA924 because Lavry did change over to torroid transformers from the laminated core power supply transformer.  Also it is a good practice to periodically ship the DA924 to Lavry for recalibration and testing of voltage and frequency.

With kind regards, Peter Foster.

Thanks, Peter

This is an interesting and in a way controversial subject, let us to view it in details.

peter foster wrote:

It may be worthwhile to check what transformer is installed in your DA924 because Lavry did change over to torroid transformers from the laminated core power supply transformer.

When I got my unit, it was used; the unit was perfectly operational and sounded flawlessly. I noted still a very slight exuberance in HF and some minor deficiency in lower MF. The HF exuberance was not line with any other DAC I heard – it was absolutely free from any colorations. What is interesting about that exuberance also that it happen ONLY what you feed Lavry from “alien” sources. What you feed Lavry from Lavry own source (AD122) the exuberance did not show itself up and the entire result sound very neutral. Anyhow, I thought that the exuberance was due to some kind of internal miscalibration or fault – sine the unit was not new and I sent it to Lavry for recalibration and testing.  As I remember Lavry charged quite a lot of money to do so (somewhere around $800) and informed me that the unit was OK but still they were able to found that some measurements were not perfect and they we able to set them up according to what they consider they shell be. When I got my unit however I did not notice any change in sound at all but I got a piece with the fact that my DAC operates as it should. I do not remember already now but I believed Dan told me that they had a new transformers and new upgrades that make unit more thermo-stable, I think it was called MK3 modifications or something like that. It was some kinds of controversy about it as I requested them do not do it but they told me that since they have the unit then they insist to do the upgrade. I was not pleased but In the end I told them to do whatever they want just give me my unit back as I was needed to USE it. They did whatever they did, changed the front panel from DB Technologies to Lavry Engendering (it is reportedly a free procedure) and send me my DAC back. I do not know if they changed the transformers – I did not open the unit before but I presume that if they changed so much money than they did something else besides the calibration.  One more thing – I asked than to bypass the analog attenuators that Lavry has between the stages on the back panel and I would like to have as less contaminated signal path as possible.

peter foster wrote:

Also it is a good practice to periodically ship the DA924 to Lavry for recalibration and testing of voltage and frequency.

peter foster wrote:

You may have noticed that when you power it up, it makes you wait a few minutes as it performs a calibration. This is why it may be a good idea to cycle the power once a week."

I do not know about that.  When I was asked them about 24/7 duty they did not told me about it. Also, as I told by Dan and as I understood (that might not be the same) Lavry fully recalibrates itself each 15 minutes. So, if they told about the needs of “cold reboot” then it is possible that during the startup they have some other type of calibrations that are not available during operation. I do not know if it is so and I need to run it with Lavry. I was under impression that the long start up (very annoying) is due to the unit waits until the temperature of internal heater got stabilized…

peter foster wrote:

Dear Romy, advice that I received from Lavry regarding heavy use (6 hours each day) of the DA924 was as follows: "If you really use it that much I would leave it ON but power cycle it one a week. If you don't plan on using if for about 3 days then you should turn it OFF.”

Hm, it does not sound convincing to me. I think it is foolishness to make the DAC in such a small enclosure and then invest time and efforts to fight with the unit overheating. I understand that when Lavry designed it is meant to be use in pro environment and the limitations of the standard pro rack space were critical for Dan. However, in my world I care less about the compliance with pro standards and I did intend to run my DAC 24/7, the way how my Bidat does. I have different listening habits my habits do not permit me to wait 15 min unit my DAC will start up and begin to sound right.  The sound of the NEW DA924++ (with modified PS) is so much more serious then the default DA924 that I might consider in future to keep the DA924++ as the ONLY DAC I use for the whole system, I do not know as now and will make this decision when DA924++ will be fully burned-in and what my new Bidat++ will be broken-in – it will take a month or so…

Meanwhile I did make the provision to enable to run my DA924++ 24/7 with less problems. As you see in the picture above I built a harness for my DA924++ and I run it VERTICALLY with the PS located up the unit. The right part of the unit has the regulars mounted to chassis and this is the only heat-critical part in the unit as I can judge. With vertically positioned DAC, PS atop and the ventilation holes act like a chimney the heat from PS does not go down to the DAC’ circuit board with digital and analog section. Furthermore I keep the right part (in my case of vertical poisoning it is the top part) of the DAC cover, right above PS slightly open and it very effectively removed all extra head from the PS to the air outside the unit. In the long run I would like to keep cover closed and I do not want dust go into the unit, so I bought last week a large heatsink that I am planning attach on the right panel of the init (in my case the top of the unit) where the regulators plate it. I think it will do the job, it not I would add a little fan on the heatsink… In my current setting the bottom of the unit (in my vertical positioning) heated no more then 45C-50C after many weeks of operation that I feel is perfectly reasonable temperature.

A week passed since I updated my Lavry-924, what a sound! If before the DA924 was just the most advanced and the most interestingly designed multibit with excellent sound then now it is becoming the most advanced and the most interestingly sounding DAC. It is not there yet, but some if the thing it does now are juts out of this world. My Larvy would be VERY surprised to hear it now…

In response to the email questions I got. The diodes are all Schottky now – that is no brainer. Also, I replaced all caps with power supply to Nichicon Muze KZ. The KZ is the top of the Muse line. They are low temperature (85C), low capacitance (up to 1000uF) and much larger then they might be. However they are my favorite electrolytic low voltage caps and decided to parallel them, sacrificing how it might look like, but there is no alternative to the KZ’s sound in my view. I was forces to add a heatsink to the boards with regulators – the unit positioned vertically in my rack with the heatsink atop – very effective cooling. I also put another power ensue module. Lavry used the filtered one PSOSXDS30, I use the identical but with the power filter. The common mode problems and HF nose should be killed in power lines not at the DAC’s entry. We always pay with dynamics and bass in each to those paralleled caps in the primary AC side.

With all magnificence of the new 924++ DAC it have entered the second week of the Nichicon KZ playbook. It was completely predictable and highly expected and now the DAC sound very funny.

The bass is fully developed already. The DA924 always was phenomenal (in fact beyond phenomenal) lower bass performer but it never had reach upper bass. Now the DA924’s new full bass is nothing short of another world. I it so reach and so acoustically slow that I do feel that it is the only DAC that equal in bass performance to my analog setup. The very lower bass is also got with DA924++ even more scarier. The best lover bass from DAC I ever heard in my room was from Spectral 2000 – that DAC had Bass!!! The DA924++ bass even more diabolic. I was listing today Brahms Violin Concerto with Bernstein leading Wiener Philharmoniker. It is where microphones were positioned in “opera” configuration and picked the Gidon Kremers and Bernstein’s foot movements. This performance is truly wonderful and I love it for years in addition this recoding is a good illustration of the “really deep bass”. If you want to hear how the entire Wiener stage response to the articulate Leonard Bernstein’s jumps then here is your chance. The DA924++ cares this bass with the force of submarine exposure - it sounds like somebody in the fifth floor of my building dropped a refrigerator….

However, according to the Nichicon KZ playbook the second week of KZ’s burning shell obliterates imaging… and it is exactly what my new DA924++ does now. Now the DAC has only extreme right and extreme left imaging with no connection between right and left. They are just two different worlds that refuse to talk to each other. This is normal and it should continue for the next 3 weeks…

Did I tell you that Pacific Microsonics DAC is a phenomenal 16-bit DAC? I did not lie and it is very good. Look however what the updated Lavry 924++ did with it.

They are very short and small (6Meg) files, mono, 88/24 with a few seconds of opening of the Beethoven Firth. It was played at CEC TL0 run across both DACs and then digitalized from output of my preamp via Lavry Gold AD

http://www.mediafire.com/?tlj1yjctzi1

http://www.mediafire.com/?2i9y0fboicc

You think you should not be able to hear the difference after this. Try it. There was nothing done to the files, they are as raw as could be….

Romy the Cat wrote:

Antonio, the .w64 are the same as .wav file only it allows windows do not have 2G limitation. If you “can’t” play the .w64 files then juts change the extension to .wav. The difference between those files does exists and it is in my view is a very large difference. In fact the difference in those files is very indicative about my observations about the ways how audio should sound. Rgs, Romy

Antonio J. wrote:

WaveLab won't play them either, which is very strange since they were "created" with that software. The only software playing those files, are they .w64 or .wav, is Sony's Sound Forge. Funnily enough I more "saw" the differences looking at the meter bars, than was able to hear them in this PC system. Rgrds

There is an observation that I would like to make as my updated DA924 getting in its final stage of burning in.

I feel that it is high likely that the lowest LF improvement that I reported above were not as huge as it might flow from my comments above. I mean even before the stock Lavry DA924 was phenomenal lowest LF performer. The updates that I made did push up upper/mid bass and make the phenomenal DA924’s lover bass more properly connected with the rest of the frequency region. I am sure it is all happen somewhere at the level of perceived harmonics and did not change any actual amplitude response and the DA924 was dead flat before and after.

What however was the revolutionary with my update is the “vigorousness” in the Lavry DA924’s highest frequency is completely gone. Let me to explain how it works. You have a source, let say a tape or a record. You run the source straight and compare it with the source running into AD direct coupled with DA (I use Lavry AD122). Before and after modification the sound during comparing the straight run with AD/DA let say identical. However, before modification it was identical ONLY at the absolutely equal volume level. As soon the AD/DA line has a fraction DB over the AD/DA line begins to sound brighter. After the modification the increase the AD/DA line for a few DB does not make the AD/DA line to sound different then straight feed. I also presume that the HF subsiding impacted my feeling about the “bass improvement”.

Now is the catch. I had 3 DA924 units in my listening room. One I had 9 years ago, a dealer brought one in.  Another one I had just before I bought my own unit, I borrowed it from a local guy to try it – this unit was listed VERY thoroughly. Both the first and second DA924 did not have that HF excessiveness, my own unit had. I sent my on wont in past to Lavry, pays a lot of money to confirm that my unit was to the specifications. So, I have assurance that it properly calibrated and does what is shell… Still, Lavry operation does not operate on the premises of sound but only on the premises of objective engineering. As good as they are the objectively auditable HF extension that my unit had might not have the procedures that Lavry is able to handle. What if my specific unit, of the party from where my unit came from, was made too bright sonically?   It is highly likely as my fist and second Lavry did not have that effect. I need to say that the most critical evolution of DA924 sound I made with the borrowed second unit, before I made the decision to buy DA924 to begin with.

After dealing with my tuner pick digital noise from my AD and DA, and particularly from Lavry DAC, I loaded Lavry with quite substantial amount of ferrite anti-RF filters:

http://www.romythecat.com/Forums/ShowPost.aspx?postID=7851

It includes the analog and digital cables. I was considering that if my tuner picks them up then it should be there…

I have to tell that the ferritenization of Lavry does work well even at the auditable level. Fully loaded with ferrite the DA925 sounds very-very slightly softer with slight less nervous-aggressiveness in very upper end – something that always was the Lavry’s issues and something that I always attributed to the seventh order brick-wall filter that Lavry uses.  A good discovery…. Still, I would like to have an access to the individual +/- cores of the Lavry wires (analog and digital) to make it even more “ferrited”. Also, if need to see if the ferritenization doe not screw the sound itself…

jp wrote:

In my experimentations, the da924 doesnt contribute much noise in my setup.  Rather in my setup its the lavry ad122 thats the trouble maker. When all digital is turned off I have a very quiet background.  Here are my observations: a) analog interconnects connected to ad122 and sansui, audible noise. b) analog interconnects connected to ad122 but not sansui, even more audible noise c) analog interconnects disconnected from ad122, little to no audible noise. Has the ferritization of the lavrys screwed the sound?

Still the key is to treat the cables, all cables that go into tuner and into the processors. Do not forget that you have a perfect loop: tuner goes to preamp, as in the same time tuner goes to A/D-DAW-D/A- preamp.  So, you have two grounds come to the same point and this is recipe for a loop disaster. You need to be watch how you connect it and always do not spare ferrite on PS cables (common and differential mode), digital interconnects (common mode), antenna cable (common mode), tuner analog output (common and differential mode). You will be able to decrease the digital noise inflicted to tuner VERY dramatically.

Romy the Cat wrote:

Still the key is to treat the cables, all cables that go into tuner and into the processors. Do not forget that you have a perfect loop: tuner goes to preamp, as in the same time tuner goes to A/D-DAW-D/A- preamp.  So, you have two grounds come to the same point and this is recipe for a loop disaster. You need to be watch how you connect it and always do not spare ferrite on PS cables (common and differential mode), digital interconnects (common mode), antenna cable (common mode), tuner analog output (common and differential mode). You will be able to decrease the digital noise inflicted to tuner VERY dramatically. The Cat

Spoke with Dima last night. He is working on a broken Lavry DA924 units, I guess Lavry own support is so good that people need to find alternative ways to fix them. With all criticism that Dima usually imposes to the design of digital devised he was absolutely thrilled and very impressed how Lavry made his DA924 multibit. We had a prolong conversation – did have leans in details how DA924 operates and my primary interests were the following:

1)      Option to use DA924 as master clock

2)      Option to use DA924 as salve clock

3)      Option to put clock doubler in DA924 and enable it to work at 4X

4)      Since it is a pure multibit  then an option to  bypass  the DA924’s 7th order filter at 2X

Lavry DA924 is running naturally at 2X speed and does not DSP for 2X sources. Upon farther assessments of all options to do DA924 we concluded that it is not truly worthy to make any changed to DA924 and whatever it is done in DA924 is done so nice that it very much worth to keep as is. I do not have any snick complains with DA924. I would like it to run 4X but Dima brought a lot of arguments that would not make the 4X sampling defiantly preferable. Ironically, listening Dima’s arguments I noted that his position is very much sound like the position of Dan Lavry, at least his arguments against the 4X were very similar. So, my DA924 will stay as it is and will remain my primary 2X DAC with any further modifications.