Trivista 21 is an upsampling DAC, whatever format you feed it, it internally upsamples to 96KHz or 192KHz depending on the position of the back-panel switch. Upsampling is performed by the CS8420 chip which is a 1990’s signal processor running a rudimentary algorithm in integer that is sure to introduce all kinds of errors.
My approach is to minimise the CS8420’s processing by setting the back panel switch to 96KHz and to feed the Trivista only 96KHz audio. By feeding 96KHz audio into the CS8420 chip and extracting 96KHz audio from it, the (nasty) effects of the chip is minimised. Roon’s DSP function performs sample rate conversion on the fly and also allows A/B comparison quickly and easily. Both 44KHz and 88KHz material sound much better when upsampled by Roon to 96KHz before being fed to the Trivista. The reason must be that Roon’s upsampling is better than the CS8420. This is hardly surprising: modern software with sophisticated algorithms running at 64 bit floating point precision on a powerful computer is going to be far superior to a 20 year old DSP. We are talking primarily 44KHz here: if the source material is already 96KHz then there should be no difference.
It is not possible to repeat the same trick at 192KHz because Trivista does not accept 192KHz audio. I expect it’s possible to bypass the CS8420 entirely by a hardware mod and if you have done it maybe you can share your findings. The 96KHz setting gives the solid, in-your-living-room kind of sound that I like. According to the data sheet, the DSD1792 DAC chip’s distortion at 96KHz is 0.0008% but at 192KHz it is 0.0015% so it is not clear to me whether the higher sampling rate will be better. I do have another 1792 based DAC that does give a better sound at 192KHz in my system.
If your source is Roon or other software that can perform on the fly sample rate conversion, go ahead and try it – I think you will like the result.
My first encounter with reel to reel tape recorders came in the late 1990s when I bought a Nagra IV-S with QGB 10.5″ reel adapter. I had already been interested in high fidelity audio for 15 years by then, but I was introduced to the world of audio recording by some friends after my return to Hong Kong in 1997. I disliked the quality of Redbook digital, and high definition digital was still new at that time and the equipment was expensive. Analogue recorders, on the other hand, were very cheap, and I was able to buy a large lot of Ampex studio recording tapes at a close out price. Used aluminum reels could be bought for US$1 each in surplus stores in Hollywood ! Over the following 7 to 8 years, we made dozens of recordings, mostly orchestral and some chamber music. After moving to a larger apartment 12 years ago, I had more space and wanted a master recorder for editing and playback. The IV-S is good for neither task; its playback electronics is more of an afterthought, and it does not allow for precise positioning of the tape for editing. The Nagra T-Audio recorder was initially developed as a scientific instrument, and later adapted for the television and film industry. Due to its substantial cost when it was introduced, it was too expensive for most music studios. The listed price in 1983 was £26,000, enough to buy a modest house in London ! Luckily, by the mid-2000s, analogue has fallen out of favor, and I was able to pick one up, fully refurbished with new heads, from Nagra for 8000 CHF. Still a substantial sum, and about twice the going rate for a Studer A80 at the time. However, I was attracted to its small footprint and the amazingly precise mechanical function, which makes tape editing very easy.
At around the time I bought the machine, The Tape Project in the US started to release commercial recordings on 15ips open reel tape. In the past, we were only able to get professionally recorded materials on bootlegged copies of master tapes, usually production masters, distribution masters or safety masters. One had to know the ex-engineers who possessed these tapes, and convince them to make copies for you (for a price). The Tape Project brought this into the mainstream, with fully licensed copies made from session masters provided by record labels. Soon, others joined in the fray, with some small record companies releasing their old analogue recordings (Opus 3, Foné), others licensing famous recordings of yesteryear (The Tape Project, Analogue Productions), and companies actually recording new materials for release on tape (Yarlung Records, Master Tape Sound Lab, Chasing the Dragon). The Tape Project managed to license two famous Decca recordings, thanks to Mr. Winston Ma of First Impression Music. The Suite Espanola in particular is a recording masterpiece by Kenneth Wilkinson, and it is well worth getting into reel to reel playback for this recording alone. Apparently, it is extremely difficult to convince Universal Music, the owner of the Decca catalogue, to license material for tape release, which Mr. Ma managed to do. Up to this day, these are the only two Decca recordings released in tape format that I am aware of. To be honest, it is much easier to pirate a high definition download than an analogue tape, and audiophiles would always want a first generation tape if given a choice, so their fear is misplaced. Happily, Decca also made recordings for RCA during the 1960s, and some of these wonderful recordings have been made available by Analogue Productions.
Dan Schmalle (a.k.a Dr. Bottlehead), one of the founders of The Tape Project, also uses a T-Audio. He advocates using an external head preamplifier to bypass the player’s native repro electronics. I was fascinated by this possibility and decided to experiment. I contacted Tim de Paravicini in the UK, whom I have known since the mid-80s, since he is well known for modifying tape recorders and supplying the electronics for some of the best recording and mastering studios in the world. Doc B’s approach is to disconnect the cables that run from the head to the repro electronics at the back of the machine and solder these to connectors installed on the spare holes on the back panel (the machine has two spare holes for the optional 30ips copy output). He runs another set of cables from these connectors to his head preamp. Tim thought this was a bad idea, since he predicted that noise would become a problem. Tim’s opinion was that the sound quality of the native electronics is mediocre, but actually quite respectable for a studio machine. He could improve the sound by modifying the repro electronics, but I would need to remove the boards and send them to him. Being habitually paranoid, I was afraid something could go wrong and the boards could get lost or damaged in transit. I contacted Nagra, but they informed me they no longer had any spares. I looked everywhere on line (and I am still looking), and although I have found lots of logic boards, I have never come across the audio board available for sale. Without the boards, the machine would not work. I thought about hand carrying the boards to him when I visit my son at Cambridge, but he still needs to send them back to me, as I would not be hanging around for weeks waiting for him to finish the work. As an aside, my recording partner sent him the repro boards for a Revox tape machine, and after the mods, the sound was magically transformed.
I just happened to have a spare phono preamp sitting on my shelf and I don’t have to invest in a new preamp, so it would be no great loss if I experiment with Doc B’s approach and fail. This preamp is Allen Wright’s RTP-3C. Or I should say a prototype of the RTP-3C. Allen was an Australian audio electronics guru who started his career at Tektronix designing the amplifier for oscilloscopes. Those were the days when the electronics were tube based, and needed to have a wide bandwidth in the megahertz range and with extremely low distortion. He subsequently built upon this basic circuit to design high performance audio amplifiers. He consulted for various manufacturers and did custom design work, and towards the end of his life, he started to manufacture his own equipment. He thought it was a good idea to publish his designs and to get feedback from enthusiasts. He recruited some “beta-testers”, me being one of the ten, and we were sent parts and instructions to build our prototypes. Everything was hard wired using the dark art of point to point wiring. It was an extremely complicated design, with twelve 6922/E88CC tubes. The phono stage only has one amplification stage, and the line stage only one amplification and one buffer stage. The design is fully balanced/differential without any feedback, including the passive RIAA equalization. The phono stage has 0.1mV sensitivity without resorting to transformers, and the bandwidth of the preamplifier is 1 MHz. It has a choke filtered power supply in a separate chassis, and the “Superreg” shunt regulators. It took us two years to build and optimize the final design. The cost was only 10% of retail, but the experience was priceless. After using the preamp for several years, I bought a factory built unit (with PCBs) in RTP-3D guise (the A and B iterations were never commercialized), and my original prototype has been sitting on my shelf ever since.
I plugged in the preamplifier for soak testing but after a while, two tubes blew. I replaced the tubes but they blew again. I finally figured out that the insulation pads of two of the LM317 regulators for the tube heaters have deteriorated, and their casings were shorting to the ground plane. I went ahead and replaced all the regulators just in case, and the preamplifier has been working fine thereafter. The preamp has RIAA equalization for phono cartridges, but for tape, we need different EQ curves. I mostly use Nagramaster EQ for my live recordings, and most commercial releases use CCIR EQ. As I don’t have 7.5ips tapes or use other types of EQ, these are the only two I need. The CCIR EQ has a treble cut at 4500 Hz (a time constant of 35μs) during recording, but does not define a bass boost. The Nagramaster EQ defines a bass boost at 50 Hz (Tc = 3180μs) and a treble cut at 11,800 Hz (Tc = 13.5μs). The output of the repro head increases with frequency at a slope of 6dB/octave, which means the output doubles for every doubling of frequency. This is due to the physical property of magnetic tape playback. Therefore, the EQ for playback needs to have a low pass filter at the bass region starting after the bass boost, and a high pass filter to correct the treble cut. In the RTP-3C, the tube phono stage is a cascode, which for all intents and purposes has infinite source impedance, thus negating the need for feedback. The EQ calculations therefore only depends on the other resistive elements of the circuit. It was a simple matter of removing the components of the RIAA EQ and replacing them with a variable resistor. The resistance is adjusted until the output decreases by exactly 50%. The resistance across the resistor is therefore the resistance of the circuit. It is then a matter of dividing the bass time constant by the resistance to find the correct value of the capacitor to form the low pass filter. Although the CIRR curve does not specify a corner frequency for the low pass, the curve has to start from somewhere, and most recorders probably use a frequency of 25 to 30 Hz. There is also a “head bump” where the wavelength coincides with the width of the head gap causing resonance, which is usually at around 30 to 40 Hz with 1/4″ two track heads playing at 15ips. Experimentation is therefore needed to find the best value. For the high pass filter, it is a matter of dividing the time constant by the value of the capacitor we have chosen to find the correct resistor value. This is used as a ballpark figure to start our experimentation. By trial and error, I found that 0.34μF gave me the best looking low end response on CCIR, and 0.12μF for Nagramaster. To find the best resistor value for the high pass, I put in a 200R trimmer and adjusted until the frequency response is the flattest. To switch between the two EQs, I installed a DPDT latching relay and directly soldered the components onto the pins. I would love to use film and foil caps but there was just not enough space. Therefore, I used Kemet MKP caps and Halco precision metal film resistors. Having defined the value of the EQ components is only half the battle. There is still the small matter of head loading. The inductance of the head, the capacitance of the cables plus the input stage, and the input resistance form a resonance circuit. If the resonance frequency is within the audio band, it will affect the frequency response. If used judiciously, it could be used to boost the sagging high frequency response. The recorder uses coaxial cables to connect the head to the repro electronics, and the cables have pretty high capacitance. When I measured the inductance of the playback head, I was astonished to find that it measured 670mH on the left and 690mH on the right. Most modern repro heads have inductance in the 100mH range, so this is rather high. This also means I don’t have a lot of room to add capacitance before the resonance frequency enters the audible range. By using the internal cables as suggested by Doc B (but extending them rather than soldering onto a connector), the resonance frequency ends up to be around 4kHz, and I needed to use a fairly low input resistance (23.5K per phase) to damp down the resonance. But this results in a 6dB drop at 16kHz, and 9dB down at 20kHz. Not a disaster, but could be better. The internal electronics of the T-Audio is pretty complicated and has all sorts of circuits to compensate for this, but this is anathema to my audiophile sensibilities. I left it at that for a while, and the sound is actually very good, without any obvious loss of high frequencies. However, the native playback of the machine does sound brighter and seems more dynamic, albeit more “electronic” and less organic than the tube preamp. The one advantage of the RTP-3C is that it is a fully balanced design. Therefore, I can connect the head to the preamp in a balanced configuration, with the benefit of at least 60dB of common mode noise rejection if the components are well matched. I therefore twisted a pair of teflon coated pure silver wires together and soldered them directly onto the head block. With only 75cm of cable, the capacitance of this set up is 27pF. I played around with resistor and capacitor loading at the input, and managed to get a flat response to 16kHz and -1.2dB at 20kHz by loading with 15pF and 240K each phase. I called it quits at this point, being sick of the cycles of desoldering, resoldering, measuring and repeat. The frequency response using an MRL test tape at this point is +2.5dB at 32Hz, +1.2dB at 63Hz, flat to 16kHz and -1.2dB at 20kHz. The rest of the response is +/- <0.5dB.
So how is the sound ? Happily, Tim’s prediction did not come true. There is no noise whatsoever, only a soft tube rush if I put my ear against the mouth of the midrange horn. Compared to the native electronics, the sound is much more musical, with a greater sense of presence and palpability. The upper frequencies of the native playback seem harsh by comparison. The sense of space is eerie; the entry of the trumpets at the beginning of Leibowitz’s RCA recording of Pictures At An Exhibition (Analogue Productions) immediately showed the difference. The tube preamp gives the impression of distance and space, with the strings coming in in front of the trumpet, whereas the native playback presents a much flatter, two-dimensional view, with the trumpet placed more forward and without the sense of space surrounding the player. The timber of the instruments is more colorful and realistic, whereas the native playback tends to homogenize the sound a bit. A studio master copy of a Kondrashin recording of Shostakovich 10th symphony on Melodiya shows good layering of the instruments, the sound of the military drums at a distance giving a tremendous sense of foreboding. The scale of the sound can be overwhelming at times for those not used to hearing master tapes, as LPs and even commercial digital releases often have compressors and limiters applied. Bill Evan’s Waltz for Debbie released by The Tape Project gives a wonderful life-like presentation of the ensemble, the scintillating cymbals with rich overtones and excellent presence of the piano. Vinyl would not give such a solid impact to the piano notes, not even on my souped up Garrard 301. With high resolution digital downloads now widely available, are there any reasons to go back to tape ? Is this pure anachronism, or another fantasy created by analogue diehards ? Admittedly, digital audio has made tremendous improvements since the days of perfect sound forever. They do sound ever more perfect than before, but this improvement only applies to newly recorded materials. Older digital recordings made with Redbook CD specifications would always retain their deficiencies, since up-sampling cannot rectify the problems created by anti-alias filtering, for example. The high frequencies lost to filtering, and the phase shift caused by the brick wall filter could never be undone. Ironically, analogue recordings transferred to high resolution formats using modern equipment can sound very good indeed. This depends on the condition of the master tapes as well as the skill of the mastering engineer and the care taken. In my experience, master tapes transferred without any compression to high bit rate DSD can sound extremely close to the analogue source. The availability of such transfers is still limited, and good DACs capable of fully realizing the quality of these high resolution formats are expensive. The best digital recordings are no doubt modern recordings made in high resolution digital formats. The sad fact is, the major labels nowadays are no longer investing money into making high quality recordings. Most labels no longer have recording teams and contract out the job to the lowest bidder. I doubt we will ever see the likes of Kenneth Wilkinson, Gordon Perry, Lewis Layton and Bob Fine again. And Kingsway Hall is but a distant memory……… And there is a ritual associated with analogue playback that endears enthusiasts. The periodic head demagnetization, head cleaning, threading the tape, splicing with a razor, cleaning the LPs, maniacally adjusting the VTA……… It keeps us busy. So, the argument is not so much digital vs. analogue, but finding the best way to maximize the pleasure from recordings made in the different era.
My first open reel tape recorder was an Otari MX-5050 that a friend gave me in the 1990s. It was a solid machine that could record and play both 2 and 4 track 1/4″ tapes. However, after I started doing live concert recordings, I needed something better since the Otari was not exactly portable and the quality of the recordings made on this machine was not great. The Nagra IV-S had been a standard for location sound recording since the early 1970s until digital came along. These machines were built like tanks and still sound marvelous today. I bought mine from a studio equipment store in the UK. It was an ex-BBC unit that came with the QGB 10.5″ reel adapter. It had served me well for more than a decade, making many recordings of mainly classical concerts. Last year, I learned that the engineer who was in charge of the analogue recorders at Nagra, M. Herbert Bartels, had just retired. I therefore called Nagra (now called Audio Technology Switzerland) to ask if I could send my IV-S in for a major service. It was about time since I had been using it for 12 years without any professional servicing. M. Bartels was very graceful in agreeing to return to work part-time for this. After I sent them the recorder, an idea came to me. The IV-S is a fantastic analogue recorder, but the playback function is a bit of an afterthought given its main function as a location recorder. As the analogue tapes I made were for archiving, with the multitrack digital files being the source for mastering, playback function was not important other than for my own amusement. Nagra of course had their famous studio machine, the T-Audio, since the early 80s for mastering function. This was a marvel of Swiss engineering, and apparently cost £23,000 when it debuted in 1983. For that amount, one could buy a luxury car in those days or even a small flat in London ! This was simply too expensive for most music studios, especially for a two-track machine, but it became the darling of movie studios with their much larger budgets. It also gained traction in the scientific community as a data recorder. So I figured I could ask if they had one available, and if the price was not crazy, I would consider buying one. It turned out that they did have one available, and M. Bartels could work it over and change all the necessary parts including the heads, rollers and belts. The price of 8500 CHF seemed very fair to me, especially since the man who wrote the original service manual would be giving it a once over.
After I received the machines back, both in brand new, original shipping cartons, I immediately made comparisons. While the IV-S playback is competitive with an excellent turntable set-up, the T-Audio is in a different class altogether. The sound is huge, with more weight, more solidity and more power. It is as if the energy of the music has been cranked up by an order of magnitude. However, while it is widely acknowledged in the pro audio community that the T-Audio is one of the best transports available, its playback electronics are not up to the same level of performance. Soon, audiophilia nervosa compelled me to find ways to improve the playback performance. The major criticism I have with the machine is that it sounds a bit electronic, and what is missing is the organic flow of the music and the natural tonality of the instruments. This would not surprise anyone who has looked at the schematic of the repro board. The circuit design is heavily reliant on op amps, mainly LF353N. Whereas one can design pretty respectable audio circuits with modern op amps nowadays, this was not the case in the early 1980s. After doing a bit of research, I found a trove of information on how to wire the repro head of tape machines to outboard electronics. Dr. Bottlehead even has a T-Audio thus modified to work with his tube repro electronics. I therefore called Tim de Paravicini, whom I have known since the 1980s. He was not keen on this idea, since he thought noise could be a problem. He suggested that I send him the repro board and he would work his magic on it. I called Nagra to see if they have spare repro boards, just in case things go wrong. The answer was no. Therefore, if something goes wrong, if the board gets lost in the post, I am toast. The transport simply won’t work without the board, even if I use outboard electronics. Unless I bring the board over to Tim and wait while he modifies it, there will always be a risk. I also don’t want to do irreversible modifications on the machine. Well, I guess I would have to do some experimentation myself.
First, I had to wire the repro head out. The original coax cables from the heads were connected to the vertically placed main circuit board and easily accessible after opening the back cover, with the various daughter boards inserted horizontally from the front. The two cables for the repro head were unsoldered, and I soldered a pair of twisted, teflon-sheathed solid silver wires onto each cable (Fig. 3). I then connected a ground cable to the chassis ground. The signal cables were terminated with Lemo plugs in a single-ended fashion.
I had an Allen Wright RTP-3C phono preamp that I painstakingly built over the course of two years sitting idle on the shelf, as I was using the RTP-3D as my preamp. Built entirely by point-to-point wiring, it was easy to modify. All I had to do was to modify the RIAA section for tape equalization instead. Since the design used passive RIAA, it made life a lot easier. As I only use Nagramaster and CCIR EQ, there was no need for NAB. I determined the input impedance of the stage following the EQ section using a trim pot, and calculated the values of the capacitors and resisters. I used latching relays to switch between the different EQs so that connections are kept as short as possible. As I did not have a test tape for Nagramaster, I recorded the test tones using my IV-S, reasoning that all the Nagramaster tapes I have were recorded using this machine anyway. I then adjusted the resistor values using trim pots until I could get a satisfactory frequency response curve for both EQs. The Nagramaster EQ gave a very extended response, with a slight +2dB bump at around 15 KHz, 0dB at 18KHz and then steeply drops off thereafter. I managed to get the bass response to -3dB at around 30Hz. For CCIR, -3dB was at around 18KHz. The sound of the playback was much improved. The hard edge was gone and the tone of instruments was much more organic.
Good things don’t last forever, and in this case, with the 15-year-old tube preamp having been sitting idle on the shelf for about 8 years, something was bound to go wrong. One of the regulator boards went up in a puff of smoke after about 6 months of use. It was a shunt regulator and ran hot normally, as I liked to keep a fairly high shunt current for better dynamics. One of the resistors has turned into charcoal, damaging the board at the same time. I tried repairing it to no avail. As my dear friend Allen had passed away some years before, I did not know whether I could get a replacement board. Fortunately, Mrs. Wright has continued to run the business and I managed to get another blank circuit board from her. After installing all the components, and setting the correct voltage and shunt current, everything checked out. I had just received the latest installment of the Analogue Production tapes and I was eagerly looking forward to hearing them. I sat down to listen, and halfway through the first tape, the new board went up in a puff of smoke again ! A new board with new components lasting only 15 minutes ? It must be the amplifier circuit that had a fault and somehow shorting out the regulator. The fault was probably intermittent, and therefore everything checked out when I first installed the new regulator. As all the tubes were new (less than 6 months) and soak tested beforehand, it was likely due to some other components and I probably have to rebuild the whole channel.
However much I like my soldering iron, I was getting fed up. Truth be told, I still get pretty nervous when testing live equipment with 400V DC rails after all these years; another good reason to go solid state. In fact, I had been reading a lot of good things about Charles King’s King-Cello tape preamp. This preamp is based on the legendary Cello Audio Suite tape preamp circuit, updated with modern components. It is built to order with a lot of customization possible. I wrote to Charles and told him what I needed. He had quite a few orders to fulfill, as these preamps are hand-made one by one, but he got to mine in November. We decided to have precision pots installed for EQ adjustment, so that I can dial in the precise corner frequencies. The only caveat is that the preamp is single-ended. My whole system is balanced differential, and I was a bit worried about noise with a single-ended connection. Given the arrangement of my set-up, my main preamp is about six feet away from the tape preamp, as I have to accommodate my turntable as well. I generally prefer unshielded cables to reduce capacitance, and the better than 60dB of noise rejection in a balanced connection is highly welcomed in a system as sensitive as mine.
Even before the preamp arrived, I wanted to prepare for the worst. I rummaged through my parts boxes and found a pair of line input transformers salvaged from a Neve mixing desk. These are quite excellent transformers, and since most music recordings during the golden age were mixed using these desks, they can’t be too bad. I therefore hooked them up using a plastic food storage box as chassis to give me a balanced output to drive long cables. When the tape preamp arrived, I quickly connected it up to the system and my worst fear was confirmed. There was a hum. I connected the output through the balancing transformers but the hum remained, so the ground loop was not between the tape preamp and the main preamp. I disconnected the earth at the power plugs of the tape machine and the tape preamp alternately to no avail. I asked Charles for advice and he recommended that I connect the headblock ground directly to the preamp chassis. I located the ground wire coming out of the headblock, followed it to its connection at the chassis ground, lifted this and extended it with an extra length of wire (Fig. 2) to connect to the tape preamp chassis ground. This significantly reduced the noise. I then star grounded the chassis grounds of the tape preamp, the tape machine and the main preamp to a CAD (Computer Audio Design) Ground Control device, and disconnected the earth connection of the tape machine and the tape preamp at the mains plug, so that everything is grounded through the main preamp only. Now, there is total silence even through the single-ended output connection.
The sound of this set up is definitely an improvement over the Nagra’s repro electronics. The scale and dynamics are preserved, but the electronic character is gone. Compared to the RTP-3C, the instruments seem to be more focused and there is more treble energy. The tube preamp did sound more “organic”, for lack of a better word. Listening to the Analogue Productions Power of the Orchestra Ultratape, the dynamics appear to be limitless. The build up of tension during long crescendos can be both exhilarating and foreboding, as one never knows whether the rest of the system could handle it, and the climaxes could make one’s hair stand on end. Aside from the other Ultratape releases, the only recording I have heard that is comparable would be Analogue Production’s own DSD release of the Dorati Firebird. I only hope Chad will make the Mercury Living Presence recordings available on tape format. Listening to my own session masters, my team’s original intention during the making of the recordings came through clearly. The King-Cello preamp adds or subtracts very little from the signal, resulting in a very neutral representation of what is on tape, which is what one wants for mastering purposes. I am therefore not surprised why so many professional mastering engineers are outfitting their decks with Charles’ electronics. With the quality of the sound and the ability to customize, it must be one of the great bargains in audio today.
Fig 1. The Nagra T Audio with the King/Cello tape preamp and the CAD Ground Control device. The four dials allow users to set the LF and HF equalization accurately.
Fig 2. Cable bundle from the headblock, showing the ground cable (red arrow) with the extension cable spliced in.
Fig 3. The red arrows indicate the connection points of the repro head cables to the main circuit board. The cables have been disconnected from the board and a pair of teflon-sheathed silver wires spliced in to connect to the tape preamp inputs.
Fig 4. The two signal cables, the headblock ground cable and the chassis ground cable are brought
In this final part of the series I will talk about the (in)famous TriVista tube buffer and why you should bypass it.
What! I hear you say. Bypass the tube buffer? But the tubes are the reason for Trivista 21’s existence, right? I thought so too. Let me take you through the effort and expenditure I took to reach that conclusion. Let’s talk circuits. In the DAC section above we already discovered the output from the third op amp in the filter circuit is ready for output to the preamp. But instead of being made available at the back panel this signal goes through a unity gain buffer that contains the 5703 triodes. The 5703 triodes are not wired as traditional (and good sounding) cathode followers. Instead each channel has a pair of 5703 which form a long tailed pair in a discrete op amp configuration that also includes 7 bipolar transistors. Below are the improvements I made:
a) Matching the input valves: Performance of any long tailed pair hinges on perfect matching of the input devices so I purchased twenty JAN NOS 5703 and hand matched 2 pairs using an AVO valve tester. I did find the original valves were not matched.
b) Matching the NPN and PNP transistors in the second stage and output stage: originals were not matched at all.
Despite all this effort at a formal A/B test with and without the valve buffer all the Society members preferred the bypassed output with its much better transparency, communication of musical performance and far greater width and depth. I would like to try the tubes in a better circuit configuration in future and for now it was decided to permanently bypass the tube stage and disconnect the HT transformer to further reduce noise.
Figure 7: Trivista 21 (re-engineered)
After formal A/B comparison in a reference system we have determined that the re-engineered Trivista 21 is fully competitive with current DACs from Chord, Naim, Merging, Lumin. The secret is to correct the mistakes and use the correct (not just expensive) components. One last problem: now that the tube stage is disconnected do we still call it a Trivista?
Welcome to the Asia Audio Society, our little virtual community of audio fanatics for sharing ideas and experiences in sound reproduction. I am one of the contributors/moderators of the site and I would like to take this opportunity to introduce myself.
I grew up during the 1960s in Hong Kong, and my first introduction to music came from my piano lessons. It was de rigueur in those days for kids to take up a musical instrument, and my mom probably chose the piano because we happened to have one at home, and the noise it generates is easier to put up with, as compared to say a violin or a trumpet. At the beginning, my enthusiasm was lacklustre to say the least, until I was sent away to an English boarding school and met my second teacher. He was a retired concert pianist who devoted his later years to nurturing the next generation. And for his sins, he ended up with me as his pupil. Nevertheless, his enthusiasm was quite infectious and I soon became fanatical about the instrument and music in general. In those days, having a boom box was a real luxury, but after the Sony Walkman became available, it was a revelation. In sixth form, a few of us in the A-level physics class formed an electronics club, and our physics teacher would teach us all the basic skills in soldering and putting things together. One Sunday afternoon, we went round to his house to help him work on his speakers. It turned out that he had these large, bronze coloured flat panels that looked like space heaters. He took the grill off one of them, put his hand in to disconnect a panel, and suddenly he was thrown back about 5 feet, landing on his backside with sparks flying. He did not discharge the things overnight as instructed, and all of us were duly impressed. From that point on, I lusted after a pair of Quad ESL57. The electronics training was very valuable, as proper technique and a good understanding of fundamentals is essential for our hobby, and so is the appreciation for safety, undischarged electrostatic panels notwithstanding. We learned to build power supplies, radios and even a robot controlled by a Motorola 6800 microprocessor.
I was able to secure summer jobs during my university days in Scotland and finally saved up enough to buy myself a music system. It comprised of a Systemdek II turntable (the “pressure cooker”), with Mission 774 tonearm and Audio Technica AT33 cartridge, which apparently is still available new 35 years later. I had a Mission Cyrus integrated amp and a pair of KEF Coda 3 speakers. All second hand, of course. I spent most of my spare cash buying LPs. I could not afford the flimsy new LPs, especially those miraculous digital recordings, which cost about ₤5 in those days. CDs just started appearing at that time, and they sounded so horrible to me even though they were supposed to have perfect sound, and they cost more than ₤10 each, so I opted to stay with imperfection. Sadly I had to settle for second hand LPs such as the narrow and wide band Deccas, postage stamp EMIs, Columbias SAXs and Lyritas etc. One would pay 50p to ₤1 for these. Fortunately, the Scots were frugal people and they took care of their possessions. LPs from those days still form the bulk of my collection. I used to salivate over my classmate’s Linn/Naim system. I remember him dragging me along to audition the LP12 at a dealership in Edinburgh. The salesman (a kid actually) brought out the turntable, had nowhere to put it and plonked it on top of the cardboard box it came in. The subchassis was bouncing literally sideways, and he made no effort to set it up whatsoever. He played a few tracks for us casually, probably thinking that these poor students were just wasting his time. The turntable didn’t sound right to me, but my friend had already made his decision long before we set foot in the store. He was brainwashed by his hero Ivor to think that even a poorly set up Linn Sondek was better than anything else out there.
After I started working, I was able to save up and buy something better. I sold my system to a friend and bought a Roksan Xerxes turntable with Artemis tonearm and Sumiko cartridge. I had a Musical Fidelity integrated driving a pair of Linn Tukan bookshelf speakers. These proved a considerable upgrade to the sound quality. I then got a job in the US and after moving there, I decided to buy some new amplification. The salesman at the secondhand shop convinced me that a tube preamp with a solid state power amp was the way to go, and I bought a Conrad-Johnson PV10 and an Aragon 2004. It was actually quite a nice combination, and sounded more musical than my previous integrated. I had a very busy job and soon got married, so there was no time to tinker. After 6 years, I returned home. Feeling more settled and with a more stable job, I became interested again in experimentation. It might be my early experience with the Quad ESL, or a romantic attachment to the golden years of high fidelity, I started looking into vintage gear. I studied circuits and learned all about vacuum tubes, transformers etc. I started looking through classified ads and secondhand shops. I bought several classic vintage amps including the Leak TL12.1, the Brook 12A, the Quad II, the Pye PF91, and the Telefunken V69a. I experimented with different passive components and tubes. Some I would restore and then sell, others I have kept. I also bought a pair of ESL57 while visiting my sister in Leicester. I came across an ad in the local paper and asked my sister to take me to see the seller. It turned out to be at a public housing estate and I managed to buy the pair for about 200 pounds. They needed a lot of work, but the exterior was in very good condition, which was exactly how I wanted it. Sending them home took more work, and I also ordered some new panels and EHT units from One Thing Audio. Over Easter holidays, I changed all the panels as well as the EHT unit and brought them up to spec. They sounded gorgeous with certain types of music, horrible with others. I tried driving them with different amplifiers. The Quad II was a bust. They sounded slow and anaemic. The TL12.1 were better; they sounded more transparent and lively, but bass was still lacking. The V69a were better still, giving the speakers more energy and better extension. The best match though was with my friend’s Mark Levinson ML2. With these amps, the speakers were transformed. Suddenly, the bass extension improved by at least one octave and sounded tuneful and solid. These speakers are fully capable of producing fairly deep bass, but most amplifiers cannot cope with the high impedance at these frequencies. It is probably a sacrilege to some people to drive the ESL57 with solid state amplification, but it works. Another idea came from Tim de Paravicini, which is to drive the panels directly with output tubes, which is not a bad idea considering that the output transformers are usually the most expensive components in a tube amp. I never got around to experimenting with this, since the memory of my physics teacher was still haunting me.
After a couple of years in Hong Kong, my turntable gave up the ghost. Probably due to the humidity, the subchassis warped. It was apparently a common problem with the early Roksan turntables. I was becoming intrigued by idler wheel turntables anyway, so I bought a Garrard 301. It had a grease bearing, and came with a slate plinth, an SME 3012 S2 tonearm and a fairly new Clearaudio cartridge. The whole package was ₤1000 from a secondhand shop in London, and as I did not like the cartridge, I sold it on Ebay. Over time, I have changed the main bearing and the tonearm bearing, rewired the tonearm and changed the arm base. The idea was to correct the weaknesses of these vintage components while preserving the characters made them great.
On the electronics front, in my quest to learn about tube circuits, I befriended Allen Wright of Vacuum State Electronics. Allen had his firmly held beliefs on circuit design, based on sound engineering principles and impervious to trends and fashion. I became his “beta tester” for his RTP-3 preamp. During this two year period of experimentation, when I built the preamp by point to point wiring, much was learnt about circuit topology, components, and the relationship between measurements and sound quality. We ended up with a superb sounding preamp that has won much critical acclaim. This was followed by the differential 300B power amps. Having worked with these amps, which use tubes for signal amplification and transistors as regulators and current sinks, one can appreciate that each kind of technology has its place. Each type of device has its advantages, which should be exploited to the full in a circuit. The criticism directed at tubes of being “soft”, “coloured” or warm sounding is not due to the inherent characteristics of these devices, but the way the circuits were designed. With proper implementation, tube amps can sound as dynamic, neutral, extended and speedy as their transistor counterparts.
I moved into my current home 10 years ago. As the flat needed to be completely renovated before I moved in anyway, I asked an acoustic architect friend to design the living room. He was more used to designing concert halls and music studios, but he obliged. The journey was another wonderful learning experience. He did a superb job, and it is an example of how one can marry good acoustics to an aesthetically pleasing living environment. Unfortunately, the living room has outgrown the Quads. The much larger space, coupled with the acoustic treatment, meant that the Quads were not able to produce enough sound pressure. I was sad to see them go, but this also presented me with an opportunity to experiment with horns, something I had always wanted to do. I have had some experience with different horn components, having listened to various vintage speakers, mainly Western Electric, Altec and JBL. After evaluating different drivers and horns, I decided to use the wonderful Electrovoice T350 tweeters, JBL 2450H mid-range compression drivers with 500Hz rectangular wood exponential horns and Altec 515C bass drivers in reflex cabs. Frequency divider duty is relegated to an Accuphase F25 analogue active crossover. A pair of Townshend ribbon supertweeters add some airiness at the top end. The horn drivers, with their 110dB sensitivity, are merciless in revealing any shortcomings and a very useful tool for reviewing components.
Today, I am still experimenting. With our hobby, the possibilities are endless. A high sensitivity horn system is like a microscope, and faults become very obvious. This could be frustrating and yet exhilarating when progress is made. I have experimented with making interconnects and speaker cables, which are finally getting to the point of being acceptable. There is still much to do, and I will share my experience here during this unending quest for perfection.
Now we get to to the heart of the project – the DAC section.
MF made some mistakes in the analog power supply and the op amp filters. Below I will show you how to correct these mistakes and unlock tremendous improvements. Trivista uses a Burr Brown DSD1792A 24/192 DAC in PCM mode. This chip was state of the art in 2003 and boasts 127dB dynamic range and 0.0004% (-108dB) THD+N, which is comfortably beyond the performance required for CD audio. DAC chips are all quite cheap (tens of USD) on their own but they need good supporting circuits to make them sound good. TI published a reference circuit in the 1792 data sheet and I will refer to parts of it as we proceed.
+5V DAC chip analog power supply. DAC chips need separate clean +5V for the analog power pins. So why did MF supply the DAC chip’s analog power pins VCC2L and VCC2R from the dirty digital 5V power rail? After analysis confirmed by listening tests I concluded this was indeed a mistake probably made when they designed the PCB. There actually is a clean +5V rail supplied by a 7805 regulator with choke filtering that takes power from the already clean +15V analog rail but it is only used for the DAC chip’s VCC1. Get out your soldering iron and connect the DAC chip’s VCC2L and VCC2R pins to the clean analog +5V rail. You have just doubled your Trivista 21’s sound quality. For further improvement replace the 7805 regulator with something better. In this critical application the difference between a fairly good LM340A and a state of the art (in 2017) LT3045 is clearly audible in terms of transparency, resolution and spaciousness. If you want perfection you can use 3 LT3045 to power VCC1, VCC2L and VCC2R but you would need to make a piggy back board. Finish off with a sprinkling of OSCON on the DAC chip power rails. Don’t use OSCON for the internal bias decoupling caps because they have high leakage current. You have now doubled the sound quality again.
+15 and -15V op amp power supply. Each rail is provided by one 78/79 regulator taking power from the +-24 volt rails. Each regulator supplies both left and right channels. Putting in separate regulators for the left and right op amps would have cost USD5 more in components so this kind of cost cutting by MF is surprising. Unfortunately, there’s not much that can be done without seriously cutting up the PCB so I have left as is and merely upgraded the regulators to LM340. At least there is room to put in decent electrolytics.
Op amp integrator and filter. MF uses the reference circuit and I agree 100%. Some hobbyists claim they can come up with a better circuit, if they are so good they can go and design the DAC chip as well. However here is where MF made another mistake on the PCB (but not on the circuit). For some reason MF uses half of a 5532 dual op amp in all six positions instead of the recommended single 5534. That in itself is not a problem and maybe they had a lot of 5532 in stock. But they failed to disable the unused half of the 5532’s in the correct way. They wired both inverting and non-inverting inputs to ground, which ensures the output is driven at open loop gain and will peg itself at the + or the – power rail at random. Thus the non-working half of each 5532 compromises the working half. Correction requires track cutting to reconnect the inverting input to the output to regain stability. While you are working on the op amps why not use something better? The data sheet specifies the 5534 loved by big Japanese electronics companies who are the main customers. But the same circuit appears in OPA1611/12 data sheet so TI engineers know better. I am using LME49710 and LME49720 but if you use AD797 I would love to hear from you. While you are in there replace those cheap resistors and capacitors in the filters with your favorite precision types. I went with 0.1% Dale metal films with 5 and 15 ppm/C and hand matched 1% Phillips NOS polystyrene. There’s not much space to put in anything bigger/better.
Output buffer added. The output of the third op amp is ready to feed into your preamp. In the Trivista 21 this signal is connected to the input of the tube stage and is not available at the back panel. I added an LME49720 unity gain buffer stage so I could bring this clean signal to some extra phono sockets on the back panel to facilitate A/B comparison with/without the tube stage.
Now let’s turn our attention to the main PCB.
Figure 3: Main PCB (original)
Decoupling capacitors – stick with ceramics but use low-loss non-microphonic C0G types
If you are reading this you are no stranger to capacitors. I will leave you to decide whether you are going to increase capacitance or stick with original values. I would like to offer the following suggestions.
a. High frequency decoupling: There are about 50 low quality 0.1uF disc ceramic capacitors providing power supply decoupling onto the ground plane. Replace these with 0.1uF ceramic C0G. Remember this is a digital board with a 50MHz clock so you need caps that are good to at least 1GHz. Save your plastic films for analog stuff.
b. Low frequency decoupling: There are about 25 Jamicon brand electrolytics ranging from 10uF to 1000uF. I would go for 105C rating since you don’t want to do this again in the next 10 years. I stayed with Nichicon UKA because some 3 terminal regulators are said to be unstable with OSCON on the output. If you use OSCON, please let me know how it went. Some electrolytics provide critical voltage reference to the DAC chip and I will talk about these in the appropriate section.
1) Main power supply – quality choke filtered PSU, leave as is
Figure 2: PSU PCB (Original)
Power supply fanatics will be delighted to find large chokes on both the low voltage and high voltage analog power supplies. These chokes are bigger even than the power transformers. There are 2 main reservoir capacitors for each rail and they are wired in C-L-C arrangement with the chokes for maximum rejection of ripple, hum and other noise. The 24V rails will feed the 15V regulators for the op amps and the 90V rails will go through additional filtering before reaching the tube section. A 10V rail feeds the 5V regulator for the digital section and a 15V rail feeds the 12V regulator for the tube heaters; these rails do not have choke filters and probably do not need it. Best not to mess around with this superb design and just replace the old Jamicon electrolytics with new 105C rated components. I used Nichicon LGU at 50% bigger capacity and you should use your own favorite.
Trivista 21 circuit has several mistakes!!
I could not find any circuit diagrams on the internet, so I had to draw them myself by studying the circuit board. In the process I found several mistakes and shortcuts that absolutely prevent the Trivista reaching anywhere near its full potential. I was initially baffled and tried hard to convince myself that they were intentional. But after exhaustive engineering analysis backed by listening tests I had to conclude that they were indeed mistakes. I will highlight these in the appropriate section and show how they can be corrected.
Let’s get started with something all of us can see and that is the build quality. Trivista 21 build quality is excellent. Look at those beautiful double sided fibre glass PCBs with wide tracks and extensive ground planes. The wiring is terminated on the PCBs via proper termination pins and all soldered wiring is stress relieved with heat shrink. The last time I saw something this nice may have been during my days in military electronics.
What follows is quite a long article as I take you through each section and explain what I discovered.
Can the 2003 Trivista 21 be modernised to compete with today’s (2017) reference DACs? Enthusiasts have tried putting in expensive capacitors and op amps and some even offer their services for a fee. I applaud their efforts and generosity in sharing their work on the internet. But I wanted to go much further. When you peer inside the Trivista 21 you realise it is an extremely serious piece of equipment that demands an equally serious engineering effort to modernize it. This is not going to be a case of just swapping capacitors – this is going to be a total re-engineering. We will have to reverse-engineer the circuit, identify and correct all the weaknesses and deploy our components budget where it matters. With some work and USD 500 of components the Trivista 21 can emphatically withstand A/B blind testing against the very best DACs available today (2017) at any price. Provided you also bypass the tube section entirely!
To be continued.
If the mere mention of valves and horns is enough to raise your pulse rate, this section is for you. Hosted by Dr. Adrian Wu, this section offers a modern take on the classic technology. Adrian’s own system is a self-built tri-amped horn system. That’s SIX valve amps! Oh, he also likes DSD.
The Playback Blog is for the technically minded audiophile who is prepared to go beyond commercially accepted standards in search of engineering perfection. Moderated by Ed Kwok, a UK qualified Chartered Engineer and patent holder with a military electronics background. If playback to you means spending hundreds of thousands on a system and having it set up by the dealer then stop reading now. However if you like meticulously engineered equipment based on proper application of scientific knowledge and you are happiest when you have a data sheet in one hand and a soldering iron in the other then read on!
The Recording Blog focuses on technique and equipment to make great live recordings of classical music performances. It is moderated by Glenn Fok. Glenn has an encyclopedic knowledge of concert hall sound gained from a lifetime of attending concerts all over the world. He has personally recorded many live orchestra and chamber performances. Glenn serves as a Member of the Board Of Governors of the Hong Kong Philharmonic Orchestra.