CHATER AMP NOTE
A FEATURE OF THE MOSFET Power Amplifier which might cause user difficulty (see ‘A 40 Watt MOSFET Power Amplifier,’ TAA 2-3/88) is mentioned in the original article, where I stated the circuit should be protected from large input signals when the speaker load is disconnected.
This is seldom a severe restriction either on the test bench or in music use. How ever, a case might arise which violates this requirement as when a speaker fuse blows on overload. Another instance: if the system is temporarily disassembled for test purposes, and the audio level accidentally overloads the amplifier during speaker reconnections. If such an event were to occur, the bias servo in the amplifier might lose lock, as described in the article. This could result in a blown fuse, or instead in a blown transistor. To re move the restriction that a load always be present across the amplifier output terminals, you can change the value of the 100 ohm, 5W resistor R43 to 50 ohm, 5W. This lower value will prevent loss of the servo lock. If you expect to run the amplifier at continuous high-power output levels, the heat build-up inR43 could damage the circuit board. Remove R43 and mount a 10W version on the chassis instead. This rearrangement should provide safety against this failure mode and would allow less restricted use of the system.
W. T. CHATER; Rancho Palos Verdes, CA 90274
ADIEU TO BOAK'S C1?
AFTER READING Jim Boak's reply to Hal Weinberger (TAA 3/88, p. 57), I was rather disappointed. I put in quite some time trying to find an answer to the problems Mr. Fitzgerald (TAA 1/88, p. 54) experienced with his Boak regulators. Mr. Boak apparently didn't find it necessary to do the same. In the process I think I found a potential improvement in his supply. Mr. Boak made some general statements, in effect saying to disregard my comments. I do think TAA's letters section is a forum where authors and readers alike can extend their knowledge and understanding through positive criticism and discussion, not by saying, ‘Don't listen to him! ’ I will clarify my comments, and if Mr. Boak thinks the transient response of his supply is dominated by other factors that make C1’s effects insignificant, let him ex plain. I think my comments have merit, so let's try to improve an already good design.
My comments are related to Boak's Fig. 1 (TAA 1/80, p. 7), reproduced here:
... part of the output current is supplied by the LM340 through R2, and part in TR1 through R1. Because the voltages across R1 and R2 are the same (D1 compensates for TR1's Vp), the currents are divided by the ratio of R1 to R2. TR1 is really a constant current source set by the voltage across R2. The (desired) low output impedance is supplied by the LM340 and C2. Any transient current demand, therefore, initially must be met by the LM340 and C2.
Ideally, an extra current demand by the LM340 would increase the voltage across R2, increasing the voltage across R1 and making TR1 supply the bulk of the extra current. But C1 will first have to discharge to let the voltage across R2 rise, and therefore delays the point when TR1 starts to assist the LM340. Simply deleting or decreasing C1 thus improves the transient response. (A similar argument for the input noise rejection: a raw supply noise pulse increases the voltage across R2 because of the ‘battery’ action of C1.
Therefore, the voltage across R1 in creases, increasing TR1's output current, leaving the LM340 tc correct the output voltage. Deleting or decreasing C1 will improve the input noise rejection.) I realize that C1 was put in for some reason, perhaps to prevent the LM340 from oscillating. But I think it is worth while to see if there is a way to get rid of C1, or at least to minimize its value, to wring an extra drop of performance out of the supply.
J. M. DIDDEN; Uchte, West Germany D-3079
Contributing Editor, James Boak replies:
Thank you for your letter and for your interest in my regulator circuit. I hope you will take my comments in the con text they are intended. I know from your work in TAA that you are familiar with electronic circuits at a sophisticated level, and therefore we can discuss the design philosophy from a perspective of mutual understanding. Perhaps I was a little terse in my letter to Mr. Weinberger, not because of any lack of respect for your perspective; rather I wanted to lay the facts out in the most straightforward manner.
I fully concur with the bulk of your analysis of Mr. Fitzgerald's problems; it is primarily your final comments on transient response with which I took issue.
I have had numerous regulator failures caused by violation of the SOA parameters, most usually when someone used the regulator in a high-voltage circuit.
The addition of low ESR, high-capacity storage on the regulator output of course exacerbates the problem.
Most of the failures I have seen resulted from components being overstressed by excessive voltage, or from an undetected failure of mis-wired protection circuitry.
We have successfully used the basic regulator in supplies providing hundreds of volts, but not without a careful analysis of the transient loading problems.
Your analysis ( ‘Letters,’ TAA 1/88) of the preregulator circuit has a minor flaw; the two Ve diode drops of the pass element TR1 and its driver transistor subtract from the zener voltage, yielding an emitter voltage of about 25V DC, rather than the 30V you mention. The preregulator is designed to maintain 10V or so across the circuit of Fig. 1 of the 1/80 article (with a 15V IC1); in addition to placing the regulator near its ideal operating point for dissipation and cur rent capacity, this value also maintains a good 50 A under steady-state and routine transient conditions, and is still reasonably safe at 25V under short-circuit. The preregulator also offers about 40dB of input noise rejection, reducing the circuit's need for transient capability in this regard. As you correctly deter mined, this portion of the circuit is an unlikely source of trouble.
The transient response of the secondary regulator is indeed limited by C1.
This limitation is intentional, and has a dual function. The preregulator and the secondary regulator can work against each other under transient conditions; the LM340 is after all trying to maintain a constant voltage, but as you point out, the related pass element (TR1 of Fig. 1 of the 1/80 article) is driven by the differential in currents in R1 and R2. C1 slows the response of TR1 to assure stability. A secondary purpose is to stabilize the input of IC1; as you probably know, these ICs have a tendency to take off at about 10MHz if not tied down at this point.
The exact capacitor value is not critical; the circuit would probably be stable with a smaller value. During my initial development of this circuit I tested more than 20 different regulator types and discovered a wide variation in stability, noise rejection, and current capability; I chose values five times greater than those that worked with the worst regulators, to reduce the chance of a ‘parts bin’ assembly causing problems.
The source can charge C1 through both R1 and R2; however, since the R1 contribution is divided by the H_fe. of TR1 (at least 10 under the worst loading allowed in this circuit), the cutoff frequency is dominated by the combination of R2 and C1. Using the standard formula for F. and values of 0.5 ohm and 1uF we get a 3dB frequency of about 300kHz; using values of 1 ohm and 1uF we get about 150kHz. Thus the transient response to a demand for current is not likely to be limited by this portion of the circuit in audio applications, even with no output capacitor. The addition of downstream energy storage dramatically increases this value.
The discharge of the cap C1 is, as you describe, limited by the ability of the LM340 to pull current out, as TR1 is not in the discharge path to any significant extent. However, with a nominal 1A draw, a 1uF cap will discharge in a period corresponding to a 3dB frequency of about 300kHz.
I think, in common with every designer, that I have a tendency to develop a limited outlook about my own work. However, as a practicing engineer, I am confronted every day by engineers on my staff, or by my colleagues, with questions about why work was done a certain way.
I cannot begin to count the number of times I have made an obvious error; I am constantly reminded of the limitations of my own ability. Walt Jung helped me to correct several design limitations in this very circuit. Certainly there are ways to improve this or any circuit, and it is through the comments and investigations of contributors such as yourself that such improvements take place. Far from saying ‘Don't listen to him,’ I would encourage every reader to take the time to dig as deeply as you have.
On the other hand, based on this analysis, I do not believe you have discovered a fruitful area for design modification. I have bench tested the frequency response of the supply with many different topologies, including a wide range of values for C1. The changes in output characteristics seemed insignificant to me; however, I would be de lighted to have you construct a regulator with the modifications you suggest and prove me wrong. I only wish it were so simple to gain substantial improvements in basic circuits.
I hope this will clarify my belief that the transient response of this portion of the circuit is not a limiting factor. I spent many hundreds of hours on the bench designing this supply, and have invested additional hundreds of hours testing various modifications and answering readers' questions. There was nothing offhand about my response. I hope you and other TAA readers will continue to provide the stimulating correspondence that helps us all provide better designs.
LONG LIVE POOGE
Stereophile for August 1988 included two articles concerning a follow-up review of a NAD 2600A 150WPC stereo amplifier after NAD did some mods to improve the sound. They replaced ceramic capacitors with polypropylene film types, increased values of electrolytics in the low level stages on the power supply rails and film caps to bypass the electrolytics. Could it be NAD has read some of Walt Jung's POOGE articles in TAA? It's nice to see at least one manufacturer wake up to changes that Audio Amateur has championed for years.
I've just finished replacing the cross over caps (or bypassing the larger values) in my Infinity RS2s with polypropylene capacitors I've purchased from Old Colony. Boy, what a difference! For a small investment, I've gotten a ‘new ’ set of speakers that are head and shoulders above the originals. I've also modified my Rotel RX2002 integrated receiver using POOGE techniques as well. Even casual listening highlights the improvements.
Long live POOGE! Keep the good articles coming.
LYLE W. MCWILLIAMS; Peculiar, MO 64078
POOGE PRELUDE
THE RECENT TWO-PART article on modifying the Philips/Magnavox series of CD players (POOGE-4, 1-2/88) was very in formative. However, I would like to add to the article's bibliography a previous piece of work in the same area. In the Nov. 87-Jan. 88 issues of Hi-Fi News and Record Review, Ben Duncan explored the same series of CD players and explained modifications in many of the same areas as Jung and Childress. Mr. Duncan's approach is well thought out, requiring the design of a new analog board to implement his changes, whereas Jung and Childress worked within the existing circuit board. His writing style is unusual in technical articles. As a previous piece of work, and as a valuable companion to the Audio Amateur articles, I think your readers should be aware of the Hi-Fi News features. (See p. 56 for HFN&RR address.)
JOHN A. OTT; Seattle, WA 98103 USA.
AURORA VOLTAGES
I HAVE A QUESTION about the Aurora article (TAA 3/87, pp. 13-18). The parts list seems to be missing the voltage rating for several capacitors-specifically C102, 104, 105, 203, 205, 109, and 110; also, 304, 305 in the power supply.
WILLIAM WAGAMAN Mertztown, PA 19539
Pat Amer replies:
Voltage specs for the capacitors listed were not specified because the voltages those capacitors will ' ‘see ’ are so low that virtually all readily available capacitors will have voltage specs well above what is needed. Use your favorite brand, or whatever is in your parts box.
BIAS LEVELS AND LINE VOLTAGE
I WOULD LIKE TO comment about my experience with tube power amplifiers. I own (and use) a McIntosh 7C240 and a pair of Marantz 9s. If the former always worked flawlessly, the Marantz blew their 3A slow-blow fuses regularly and occasionally some EL-34s. I first suspected the quality and ruggedness of the output tubes, but the true answer proved to be elsewhere.
After having checked the bias circuit thoroughly, I also measured the voltage of the main power rail and found a 485V level, against the 427V shown on the schematics. Even if the latter is not a true quiescent figure, the difference cannot be accepted as only due to a voltage drop on the power rail between quiescent and running conditions. The 485V originated in a 128V AC line voltage, and the cure was obviously to drop the AC voltage to the 117V.
As the quiescent DC voltage on the power rail was measured at 440V for a 117V AC, the quiescent power dissipated in each EL-34 is about 24W for a maximum rating near 27W in VL or triode connection (this value is a guessed one for EL-34/6CA7, but is indicated for the similar KT46). With a 485V power rail, the dissipation rises to a value close to the maximum 27W.
The Marantz 9 design drives the output tubes in quiescent conditions very close to the design maximum values, which are reached allowing a 10% variation in AC voltage (music is usually listened to in the evening or during a weekend when the AC line is higher than the nominal figure).
Two pieces of advice for Marantz users:
1. Preferably feed your amps with a variable transformer to adjust the AC voltage to a maximum of 117V.
2. Choose good valves. The 6CA7/EL-34 brands are not equal.
SELECTRON 6CA7s proved to be very sensitive to quiescent conditions.
MULLARD EL-34s showed dark-red areas on their plate, but never failed.
PATRICK HERRMANN F78000 Versailles, France
The Editor replies:
Output tubes in the EL-34, KT-88 families are very sensitive to the correct relationship between their DC plate voltages and the bias. When new tubes (or valves) are installed, it is easy to set the new bias-levels incorrectly unless the amateur also checks the line voltage. Unless bias is checked for correct level while the line voltage is being held to its nominal level (usually 117V), severe damage, poor performance or both can result.
Use a variable transformer to set the amp’s supply voltage to the proper level while setting the bias level. The amplifier can probably tolerate wider variations in voltage level as the correct relationship is established between bias and DC plate voltages than if this relationship is ignored.
Reader Herrmann may find Jim Boak's 1978article (TAA, issue 1, pp. 32-37) on a regulator for the Dynaco Mark Ill power amp useful, with some modifications, for controlling the voltages in his Marantz 9.
PASS A40 MINIMALISM
I HAVE TOYED WITH building an attenuator to completely eliminate my pre amp, since I rarely use LPs. Secondly, I am currently building a six-way multi amplified, segmented ESL. Just thinking of it gives me goose bumps. Neil Shattles (Shadow active crossover designer) described what I could build in a recent letter, but his suggested crossover is far too complex for my liking.
Perhaps Mr. Didden could help? I thoroughly enjoyed his less-is-better article (TAA 2/88). With CD, we finally have a superb line level medium, which does not necessarily require preamplification. I am not a straight-line fanatic, but less active circuitry, tube or transistor, must be better.
I don't expect him to design a passive crossover for me, but perhaps he could provide the component values he currently uses for the Pass A40, as shown in Fig. 4 of his article. Using the suggested passive crossover with synthesized bandpass would significantly simplify my design. I could use three, two-way crossovers and a passive crossover, as de scribed, to accomplish my goal much more easily than cascading active crossovers. Details of Mr. Didden's inverting circuit would be very helpful.
Could Mr. Didden provide a schematic/parts list for his ‘three or four transistor ’ Class A preamp prototype? Is it necessary? Could the passive circuitry be used without it? Lastly, I am in need of an IBM AT compatible crossover CAD program. Old Colony can only provide Mr. Didden's program for Apple and Commodore computers. What is a suitable alternative? By the way, have you improved on the basic Pass A40? Have you tried the MOSFET mod? Also, if you need or would like any ESL related information, feel free to contact me.
I sincerely believe Mr. Didden's minimalist approach will be the new wave in high-end audio. Mod Squad's ‘Line Drive ’ unit supports this prediction. Thanks again for a fine article.
DAVID LANG Wauwatosa, WI 53213
J.M. Didden replies:
Eliminating the preamp for just a volume control is a very attractive idea. Choose a low value attenuator, like a 5k pot, on your CD player. This would give a low enough impedance to drive a Pass A40.
FIGURE 1: Didden's Pass A40 changes.
But you still must put a crossover some where. If you stay with a passive one, as I prefer, you are stuck with a traditional network between power amp and speakers. In my approach a preamp to drive the crossover is necessary, but I think I can trade off some power amp requirements.
For the crossover network, I do not yet have a usable circuit. I just finished the preamp, but other demands on my time seem to grow every day. If you are willing to spend some money you could consider Spectrum Software's analog simulator and a good design text like Williams' (see my article for references). Alternatively, we might be able to persuade some TAA reader to translate Old Colony's Crossover CAD for the IBM family (anyone listening out there?). As a starter, a Shareware program called Cheby lets you design passive Chebychev networks of almost infinite type and variety. I regularly scan Public Domain and Shareware catalogs, but analog design programs are very rare. [Three-way cross over CAD programs in IBM format are being worked on by several. -Ed.]
My changes on the Pass A40 are shown in the accompanying schematic. Re is just a gain equalization which is standard for a differential amp (see Walt Jung's Audio IC Op Amp Applications). The other mod eliminates the large value tantalum cap (C2 in the original schematic) and adds an offset null pot. The amp is very stable; I never have more than 10mV offset at turn-on, which later diminishes (adjust it after the amp has been running for an hour).
I have finished my preamp/driver, but despite its simplicity, you cannot just throw it together from a schematic and a parts list. Layout is very important for the cleanest sound and it uses two circuit boards; one carries the components and one, suspended about an inch above it, provides a star ground. Decoupling caps ‘stand ’ between the two boards. It also uses two separate power supplies per channel. As soon as I find time I'll write up the whole thing.
HELP FROM WA AND CT
THOSE READERS WHO have asked about service information for Accuphase equipment will find it at Accuphase by Madrigal, 2081 S. Main St., PO Box 781, Middletown, CT 06457, (203) 346-0896. Ser vice manuals cost $5 (a very fair price) and the staff is very helpful. By the way, I have come up with some changes for the Accuphase C-200 preamp and P-300 power amp. If any of your readers are interested, I would be glad to share them.
I would love to hear from any readers in the inland Northwest area of Washington and Idaho. I sell and service stereo equipment for a living and construct tube preamps as a hobby. In my line of work I have access to a complete bench of professional test equipment. If any nearby readers would like their hand-built projects checked, I would be glad to do this free of charge.
RICHARD FISHER; Rick's Stereo Shop 101 N. Union #204 Kennewick, WA 99336
SCHEMATICS FOR DYNACO STEREO 70 AMPS
I HAVE MATERIALS THAT might be of in terest to your readers who own Dynaco Stereo 70 amplifiers that do not have schematics. I have drawn up a board lay out and have numbered the components on the original schematic found in Dynaco's owner's manual.
I would be willing to send copies of the board layout and labeled schematic to your readers at a cost of $5 to cover time, materials and postage.
JAMES SMITH 518 Palo Alto Drive Vancouver, WA 98661
WASTED SPACE
I WOULD LIKE TO discuss a matter of editorial policy. I believe both Speaker Builder and Audio Amateur should and do strive to develop and expand the knowledge of amateurs and professionals alike. It therefore saddens me to see such things as your most recent ‘Showcase’ (TAA 1/88, page 42). The ‘Tempered Tone arm ’ by Mr. W. Wagaman serves very little the readers of your publication. Except that readers may be encouraged by the fact that the home builder may duplicate commercial products with good results, the story in no way helps them to do so.
Mr. Wagaman states ‘I will not divulge dimensions or viscosities.' This profits no one except to give W. Wagaman a pat on the back and a published story he can show to friends.
Well Tempered Labs has nothing to fear from audio amateurs, this is an old issue.
The W.T.T. in fact bears an uncanny re semblance to a design by Ernie Lowinger published in Wireless World, Dec. 1977, pages 46-49 and 64.
I believe such ‘'showcases ’ should be omitted from your publication. Each and every page of TAA and SB should provide the reader with thought provoking and knowledge expanding material, not public praise for those who choose to keep secrets.
MORAY J. CAMPBELL; Calgary, Alberta, Canada T2B 1N5
The Editor replies:
The dearth of craftsmanship examples generally, and in audio particularly, is our reason for the ‘Showcase’ feature. We consider the items presented to satisfy reader Campbell's criteria for material.
We omitted detailed dimensional in formation out of regard for proprietary interests of Well Tempered Labs and still heard complaints from the company's head. Mr. Campbell's opinions regarding our and Mr. Wagaman’s motives are quite mistaken. -E.T.D.
A UNIQUE DAMPER
THE INCIDENT THAT prompted me to write this letter is one in which I feel TAA readers might be interested (maybe not, it may be ‘old hat ’ to such an inventive group), especially those who own one of the new Magnavox 16-bit CD players.
I purchased a Magnavox CD650 in November 1986 and was initially amazed at the improvement in detail, clarity and bass impact over my first CD player, a Sony D-5. Along with these sonic improvements, however, there was also an increase in stridency in high brass and violins. Another sonic artifact worth mentioning is that although there was also a huge improvement in soundstage positioning of instruments with the Magnavox, lower pitched instruments with a lot of harmonics such as French horns and cellos still gave the impression that they were running back and forth across the stage (I know this is partly due to the unequal response in my 17-year-old Heathkit AR-1500).
After checking out all the small companies who specialize in Magnavox modifications for up to more than twice the initial cost of the player and deciding this approach was madness, I decided to leave the electronics alone until TAA publishes some ‘how to ’ articles on Magnavox 16 bit modifications. My next approach was to see what I could do about the mechanical aspects of the player. I tried three ‘tiptoes ’ under the player. This had no effect I could discern. A Mod Squad disc damper had enough effect on the stridency aspect of the sound to continue using it, but barely, and only on discs with a high degree of transparency. I could hear no effect on opaque discs (RCA, Denon, older Telarcs). I then tried four ‘Sorbothane’ feet under the player. This yielded another slight decrease in high frequency ‘edginess,’ especially when used in combination with the disc damper. The real improvement came when I placed a big 14 ’ x 8 ’ x 4’ dictionary on top of the player (still sitting on the ‘sorbothane’ feet). I couldn't believe my ears until I had my wife remove and replace the dictionary several times while I listened. Not only was most of the stridency gone, but the instruments remained more localized (at least the solo cello stayed in the cello section). I have since removed the approximately 10 pound dictionary and replaced it with a couple of hefty textbooks of small enough size to leave the ventilation holes in the back of the player unobstructed.
If I were reading this claim, I'm not sure I would believe it either, but when one considers a disc spinning at such a high rate in such a lightweight plastic player, there must be some pretty severe vibrations present which affect the sound.
Over time, having had more of a chance to listen to the player, I noticed even more dramatic changes in the sound quality and am now wondering if the additional weight (evenly distributed to prevent ‘concaving’ of the flimsy aluminum cover) might be causing mechanical dam age to the player.
I will describe some of the changes, hopefully adhering to the currently accepted lexicon of audiophile terminology.
London, Philips, Chandos and CBS/ Sony compact discs which previously sounded so overly bright and strident that I had to turn down the treble control on my Heathkit as much as a quarter turn from flat, now sound beautifully open and airy, without a trace of stridency with the tone controls flat. For example, the Charles Dutoit Montreal Symphony recording of the Von Suppe overtures, which was harsh and strident, without a trace of the recording locale ambience I have been reading about in reviews, now has an open airy sound with a startling amount of ambience.
Older AAD or ADD remastered recordings that previously required a hefty de crease in treble to bring tape hiss down to tolerable levels now can be played with much less decrease in treble, or in some cases, such as the Dorati Philharmonica Hungarica 1956 Mercury recording of Resphigi's ‘Ancient Airs and Dances,' with the tone controls flat.
However, CDs I could previously play with tone controls flat, without the harshness or stridency described above, including most Telarcs and Denons, now sound rather muted or veiled without a slight treble boost. I know that Jack Renner of Telarc prefers Philips type players and perhaps his CDs are equalized to sound great on these. Even so, this dulling effect worries me.
Based on the brief descriptions I have given, is this enough information for other readers to judge whether I'm causing damage to my CD player's disc drive mechanism? Would the reduction in the need for error corrections due to mechanical vibrations be sufficient to account for the startling overall sound improvement? I would be interested in hearing whether anyone else has experienced this damping effect on Magnavox or other lightweight players.
I would also like to point out that one need not pay the outrageous prices for some of the CD accessories sold by some of the ‘high end’ audio salons. For example, I obtained four sets of four Sorbothane feet from Edmund Scientific Co. (101 E. Gloucester Pike, Barrington, NJ 08007) for $34 which is less than the price of four feet from any of the ‘high end ’ audio specialists. As far as weight, a ‘V.P.I. Brick ’ might dissipate heat better than a book but is it worth the price? On another subject, I am thinking seriously about replacing my old Heathkit AR-1500 with anew receiver (I don't have the room or money for separates), either an NAD 7175 or an ARX-10 with a little more spare power for com pact discs. It measured 75W at the rated distortion (0.1%) when I first built it in 1970 (at a McIntosh Clinic). Do you think it can be updated with metal film resistors, polypropylene capacitors, etc.? I am driving rather inefficient 1978 vintage second generation Advent speakers and I'll probably never part with them.
GERALD BURT; Moreland Hills, OH 44022
The Editor replies:
I think you will hear a large improvement if you will put modern polypropylenes into your Heath AR-1500. And the re placement of the resistors will be important too, especially in the preamp section.
Old Colony is selling a good line of resistors, lots of vendors offer excellent polypropylenes, including OCSL.
Neither TAA nor Speaker Builder have ever tried to evaluate the merits of the more esoteric accessories for recordings or speakers. We leave that to the review media best suited for comparative evaluations. If I want information of that sort, I look at Stereophile's reports.
HADAWAY'S DROPOUT DETECTOR
LIVE RECORDING DOESN'T get enough play in TAA. It's the best way to train your ears and gather first-rate recordings for equipment evaluation. The cheaper Sony processors are well within the budget of any serious listener, good mikes are available from $200 to $500 a pair, and video tape is dirt cheap (especially compared to $30 for a 3,600 ’ reel of audio tape). Regular live recording keeps audiophilia from becoming a spectator sport.
So I'm grateful for any article that deals with live recording (either directly or in directly) including Dave Hadaway's (TAA 2/87) piece on dropout detection. It has, however, two inaccuracies that need to be addressed.
The first is his statement that there is no significant difference in dropout severity among good grades of video tape. Ina literal sense, this is true. If you can play the master tape 50 or 100 times, and never get a mute, the actual dropout rate means nothing.
But anyone who records digitally is acutely aware of the medium’s fragility.
One ‘hard ’ mute (i.e., one due to a surface defect, and not a temporary one caused by dirt) is enough to ruin a recording. So for peace of mind (if nothing else), the cautious (or paranoid!) recordist wants the lowest dropout rate. And tapes differ significantly.
I usually record on Maxell HGX. It strikes the best balance (for me) between price and quality. I recently bought some Sony ES-HG, a comparable tape. I use both a Nakamichi DSP-100 (nee, Sony PCM-F1) and a dbx 700. The latter's error correction display can supply useful information about record and tape condition.
Maxell recordings light the display about once every ten seconds. Sony recordings light it about once a second.
Clearly, the Sony tape has a much worse dropout rate, although if you just listened to the recordings you'd never know it, since the errors are never severe enough to cause a hard mute.
Also, I always use L-750s despite Sony's warnings against ‘the thin stuff ’ which date from six or seven years back, when tapes had neither the high output nor the coating uniformity of present-day formulations. In the past five years, I've made about 20 live recordings on L-750 tape, and never had a hard mute. (And this is in 16-bit mode, which is not very dropout resistant.) L-750s provide the convenience of two concerts on one tape, at a total cost of less than $8! (3 hours at Beta III) My mentor, J. Gordon Holt, has reported similar success with L-750s and 16-bit recordings.
The second point is the author's inability to do assemble edits without mutes or glitches. I, too, had this problem, until I figured out my error. Actually, assemble edits are trivially easy. I recently per formed three of them in creating a running master from a concert tape, and got them all right the first time. To under stand how this is possible, you must know a little about video editing.
If you stop a VCR, and then start recording again, you almost always lose synch at the transition. The control track of the new recording doesn't properly overlap the old, so the picture jerks and tears while the recorder tries to regain lock.
To eliminate this, all camcorders (and the better table-top machines, of either format) include a scene transition stabilizer. Its operation is totally automatic.
You simply find the end of the previous recording, and press PAUSE, then RECORD. The VCR will make strange burping and clicking noises as it shuttles the tape back and forth, until the control track pulses are found and properly aligned. At the same time, the head drum frequency and phase locks itself to the in put video signal. When you release PAUSE, the VCR starts the new recording in exact synch with both the tail end of the old recording, and the applied video signal. Voila! No glitch.
The scene transition stabilizer is the key to making ‘perfect ’ digital assemble edits. The VCR playing the source tape need not have this feature, but the dubbing machine must. Here's how it works.
When you reach the edit point, don't STOP the dubbing machine; instead, PAUSE it. This allows the scene transition stabilizer to maintain synch continuity. Wind the source tape to 10 or 15 seconds before the point where you want the dub to continue. Return the source recorder to PLAY, and wait a few seconds for the dubbing machine to lock itself to the source VCR. When you reach the edit point, release PAUSE on the dubbing VCR. That's it! Couldn't be simpler.
I've used this technique successfully in dubbing from a Sony SL-2000 to both an RCA VET-650and a Sony SL-HF900, with PCM and delta-mod recordings. The only possible problem is lack of standards for scene-transition stabilizers. Some VCRs (especially VHS) backwind the tape about half a second. If the new recording doesn't satisfactorily erase the old, the digital data may be sufficiently scrambled to cause a mute. The best way to learn this technique is to practice, and your experiments will quickly reveal anything special you need to take into account (not likely).
One other point. Dave Hadaway suggests that the digital-audio/ VCR format was designed to make editing difficult or impossible. This isn't likely. It is ‘in the nature of things ’ that one cannot edit a video signal in the middle of a frame! This ‘failing ’ can hardly be held against a marvelously-convenient format that allows one to record digitally on cheap tape and inexpensive VCRs. With pro quality videotape editors, one can easily edit to within a single frame. 1/30, of a second accuracy is more than good enough for simple ensemble editing, and might even allow musical passages to be replaced.
WILLIAM SOMMERWERCK ; Bellevue, WA 98007
CATHODE FOLLOWERS AND MORE
IN A REPLY LETTER to Mr. Adeff (TAA 4/86, pp. 56-8), Mr. Curcio provides three references for readers interested in dual supply (plus and minus) cathode followers. I have read the first two references cited and I recommend to TAA readers the following additional references:
1.J. Ross Macdonald, 'A Multi-Loop, Self-Balancing Power Amplifier, ’ IRE Transactions, Vol. AU-3 (July-August, 1955), pp. 92-107.
2. J. Ross Macdonald, ‘Some Augmented Cathode Follower Circuits, ’ IRE Transactions (May-June, 1957), pp. 63-70.
I have used direct-coupled augmented cathode-follower circuits (ref. 2, above) for power-tube grid drivers and electronic crossovers. My power-tube grid driver circuit uses two 12AX7 tubes and a single 6BX7GT to drive a pair of 8417s.
I am currently building my version of Mr. Curcio's preamp ‘Daniel ’ (TAA 2/85). In the meantime, I am quite pleased with a ‘Scratch-Built ’ Last PAS (TAA 4/82). A 12V, 4.5A Hr. Gel/Cell supplies voltage to the 7025 (12AX7) filaments: similar lead-acid batteries are readily available on the surplus market and battery chargers are easy to construct.
If you want pure DC powering your tube filaments, use a battery (of course, there is an associated inconvenience with batteries).
ERIC CARMICHEL Tucson, AZ 85714
ESOTERIC SOUND RE-EQUALIZER
I READ WITH KEEN interest the letter from Mr. Card and the reply/rebuttal from Gary Galo (TAA 2/88, p. 48). Since the Esoteric Sound Re-Equalizer is my project, I thought I should reply to make sure others are fully informed of it.
The Re-Equalizer, as Gary stated, uses an inverse RIAA filter followed by appropriate equalization for vintage records.
But unlike the old McIntosh C-8 preamp, you can use it with audiophile grade stereo preamps as well. It parallels the two channels of a stereo preamp/amp as well as processing the sound.
Actually, I started out with a C-8, but found it awkward to integrate into a stereo system. Also, its switching arrangement was very different (albeit with fine gradations) than all the other record compensators most of us are familiar with. A prototype for the Re-Equalizer employed a stage of inverse RIAA followed by a Fisher mono preamp. The concept worked much better than I expected. An interim design used conventional solid-state design; I sold only a few. The passive equalization design of re cent years clearly sounds better than the original-and that with 78s. Also being enamored of McIntosh, I chose the EQ selections of their vintage C-20 as the base and added to them.
Gary found it difficult to believe that going through the processing described could be better than doing the proper equalization in the first place. I do not disagree, but Gary would have a difficult time playing stereo LPs on the McIntosh C-8, and most people would find it awkward to use a separate preamp (turn table?) with the audio system. I guess the C-8 could be plugged into an unused (with CDs?) 'aux' input of a stereo preamp with a Y connector (if you want both speakers/channels to get the sound), but then you must switch or unplug the turntable from one input to the other, and parallel the stereo cartridge, and you'd still be going through the extra gain stages of the stereo pre amp. By the way, the Re-Equalizer can also be used to accurately compensate tapes of vintage records recorded through RIAA compensated preamps.
The Re-Equalizer uses no frequency dependent feedback, employs passive equalization, and was thoroughly ‘POOGEd.’ With THD and IMD less than 0.01%, I think the limiting factor will usually be the preamp. It's just possible that when used with an Audio Research or Borbely preamp, it might be better than the C-8 by itself. I do not wish to get sloppy with reproducing vintage recordings, but I don't think it would significantly matter with these small levels of degradation. The software we are dealing with probably has problems an order of magnitude greater than either of our methods.
I'm not totally impressed by Carver, but their so-called ‘Autocorrelator’ is the best piece of modest-priced noise reduction gear I've heard. Fortunately, Phase Linear also publishes its schematic, which allows POOGEing. To hear some obviously bad junk, professional yet, listen to RCA ’s CD of Strauss waltzes with Reiner and the CSO. My old two-track tape plus the Phase Linear is far superior. The Burwen clearly pumps to my ears; the Phase Linear far less. The volume expansion of the Phase Linear is also well-designed in its original form.
On some material, it can provide a significant improvement. The engineer who ‘upgraded’ the same model's second version seems to have been totally con fused. The volume expansion controls where its indicator LED flashes--not a threshold! The original intent of the LED was to indicate the threshold at which peak un-limiting was adjusted. The old box version always requires replacing switches, along with most coupling capacitors. The addition of filter action LEDs helps immensely to set noise reduction. With various mods the Phase Linear can be turned into a useful and not bad sounding device.
MIKE STOSICH; Downers Grove, IL 60515, USA
Contributing Editor Galo replies:
First I would like to point out I was not criticizing the Re-Equalizer from Esoteric Sound. For those 78 collectors who, for whatever reasons, cannot use two turntables and a separate 78 preamp, this product should work very well.
I do prefer using a separate preamp for 78s, thus getting the EQ correct in the first place. I am pleased that Mike does not disagree with this. I do not find it at all difficult to integrate the C-8 into a stereo system. The dual mono outputs of my modified C-8 are fed to a spare high level input on my stereo preamp. If your preamp or integrated amp is short of in put space, the DB Systems switchbox, available from Old Colony (KM-10A or KM-10B) will solve the problem nicely.
If you are going to pass real hi-fi source material (such as a CD player) through the switchbox, get the KM-10B with gold-plated jacks and use high quality internal wiring, such as the center conductors from the Mogami Neglex 2535 cable.
Since I would avoid putting unnecessary switch contacts in the CD signal path, I'd use another input for the switchbox, such as tuner.
My C-8 modification article (TAA 1/85) does have provisions for paralleling the two channels of a stereo cartridge, with a switching arrangement allowing mono playback of both laterally and vertically cut recordings. For reasons I al ready stated, I also favor the use of a separate turntable for 78s. If you must use the same turntable for both LPs and 78s, do not use the same platter mat for both.
Even the ‘cleanest ’ 78s can leave abrasive matter on your mat which will, in turn, be deposited into the grooves of your LPs. Regarding the Dual CD-5000 mentioned by Mr. Card, my complete re view of this product appeared in Stereophile (Nov. 1988, p. 134).
CAVEAT CORRESPONDENTS
Things that go bump in our round file:
1. ‘I'm thinking of building a 16-in, 8-out console in my basement. What tape recorder should I buy? ’
2. ‘Is my Fisher Z-705 receiver worth up dating? Where should I begin?’
3. ‘Although I forgot to enclose a stamped, self-addressed envelope, please answer the following nine questions based on my experiences building your inverted RIAA kit. ’
4. ‘Please forward this (unstamped) letter to Ralph J. whose letter appeared in one of the 1970 issues--don’t remember which. ’
5. ‘I have a Milhous 10W integrated stereo amplifier and a Gesundheit turn table. Which of the following six cartridges would you recommend?'
6. Queries with no stamped, self addressed envelope enclosed.
7. Letters without return addresses on them whose envelopes have strayed away somewhere.
8. Illegible hand-written letters scrawled on odd scraps of paper. If you have no ac cess to a typewriter, please try to be sure our typesetter doesn't lose his eyesight and his mind in deciphering your writing.
(This is especially important if you want us to publish your classified ad.)
------------------
Also see:
A MULTI-TONE INTERMODULATION METER, PART 1
EDITORIAL Catch-Penny CD (2), By Gary Galo