Jun 022017

It is my great pleasure to bring you the third post for the article that Yazaki-san is writing for us, “My Adventure With My Old Marantz Model 7”.

Many thanks to Yazaki-san for taking time to write this article and share his wisdom with us, it is very much appreciated by me, as well as all of us here at Jeff’s Place!

Yazaki-san’s vintage Marantz 7k preamplifier.

In Part 1 Yazaki-san told us about his near forty year passion of pursuing ‘real sound’ with his vintage Marantz Model 7 preamplifier, and the resulting enjoyment & satisfaction that pursuit has brought to his life.

Yazaki-san shared with us his thoughts about several brilliant design aspects of the Marantz Model 7, and how its design allows it to still hold its own against, or even surpass, anything made today in musical ‘real sound’ terms.

Yazaki-san also told us about his approach for fine-tuning the Model 7’s performance for ‘real sound’ through a careful selection of internal parts like capacitors and resistors, and finally, hinted at the modifications he was going to tell us about in Part 2 of his article that would take the Model 7’s performance to an even higher level.

In Part 2 Yazaki-san told us about his way of modification with his Marantz Model 7, and how important the quality of the power supply is to the overall performance, because the current from the power supply turns into the signal current.

Yazaki-san described for us how the switch to the Ultra-Fast & Soft Recovery STTH6112TV2 for +B rectification improved the speed of the current from the power supply, and lowered the noise.

Yazaki-san also described for us how he likes to install a hermetically sealed oil-filled capacitor into the subsequent stage of the rectifier tube or diode, connected in parallel with the main electrolytic capacitor for rectification, and how that its addition provides a more responsive and organic sound.

Now, without further ado, here’s Part 3 of Yazaki-san’s article about his life with the Marantz Model 7k – enjoy!


My Adventure with My Old Marantz Model 7k

Part 3: My Way of Modification

Dedicated to Saul B Marantz and Sidney Smith for their true masterpiece!


Shirokazu Yazaki

Addition of Decoupling for Cathode Follower

I happened to experience a surprise just one year ago when I completed the restoration and modification of Handoko-san’s original Model 7, when I made the addition of a decoupling capacitor for the +B voltage line (280 VDC for V3 and V6) of the cathode follower, which brought out an extremely massive and deep sound.

Surely, I wanted to realize the same type of sound with my Model 7 k, and I started to search for the reasons of that result.

At last, I reached the conclusion that the one major difference of the modifications between my Model 7k and Handoko-san’s original Model 7 that I restored & modified, was the addition of the decoupling capacitor for the cathode follower.

Actually, I added my favorite Mallory 30μF/450VDC ( C4 ) just before the 280 VDC line for the cathode follower tubes, and was able to experience the same renewed sound with remarkable forcefulness also through Model 7 k, which I had never heard before.

Please look at the figure, “Comparison of Tube Preamplifiers” by Kato-san and “Tube type” in it.

You can see the tube types of phono EQ circuit for each preamplifier, Marantz Model 7, McIntosh C22, MX-110, and Audio Research PS-6C, consists of exactly the same tube, the 12AX7 (ECC83) for V1 & V2, but the Model 7 and C22 use a 12AX7 for V3 as the cathode follower.

However, the MX-110 was designed to use the 6D10 for V3, and a 6DJ8 was adopted in the PS-6C at V3, to use as the cathode follower.

The values that express the amplification degree of these tubes are, for the 12AX7; μ=100, the 6D10; μ=57, and the 6DJ8; μ=33.

Therefore it could be easily understandable that 12AX7 has the highest gain of these tubes, and it apparently fits well for the phono EQ circuit, which requires a high gain tube, but would it also be a good fit to use as a cathode follower?

I understand well that the cathode follower circuit is designed for reducing the output impedance, and the lower output impedance of a preamplifier would be desirable.

The relation with the output impedance and the input impedance of the power amplifier would be a relative matter, and because the usual tube power amplifier’s input impedance would be higher than 100k ohm, that the Model 7’s output impedance could be sufficiently low level, and so practically we could see no problem about the design of Model 7, I suppose.

Meanwhile, Kato-san described with his words how in the PS-6C that V3 is replaced by a 6DJ8 tube, and as a result the output impedance is much lower compared to a 12AX7.

Generally speaking, the lower output impedance tubes would be better for driving later tubes forcefully, and I agree with that. In such a point of view, the sensitive tube, 12AX7 would not to be the best tube for using as the cathode follower circuit, and so the engineers at Audio Research obviously made a decision to use a 6DJ8 instead of 12AX7 for V3.

Also, according to one popular theory in Japan, because the 12AX7 is not a powerful tube, and furthermore, because of its higher impedance, using a 12AX7 as a cathode follower is questionable, I have often heard.

But I have found that such a concern for 12AX7 was proved unfounded through the results of Handoko-san’s Model 7’s modification.

Yes, 12AX7 transformed itself to be like a more powerful tube! Can you imagine?

I found out the fact that when we need more forceful sound through the cathode follower circuit, installing a little bit excessive value capacitor for the decoupling of the local power supply accomplishes this quite well.

I also installed oil-filled capacitors by Arizona Capacitors into + B, 245 VDC line for V1, V2 ( C6 ), and also +B 265VDC line for V4, V5 ( C5 ).

These additions of decoupling capacitors bring out more sensitive, but also stable sound, from Handoko-san’s Model 7, and also my old Model 7 k.

How to Bring Out the Potential of π Filters

Please take a look at the circuit diagram for the modifications below.

You can see 3 π filters before the +B, 280 VDC line, for the cathode follower, V3 and V6 tubes, and a total of 5 π filters are installed before 245 VDC line, for V1 and V2 tubes, for the phono EQ.

Basically, a π filter consists of a pair of capacitors and an inductor for reducing the noise and/or ripple from the rectification (mainly), and it works well under the condition of comparatively stable current flowing into the load through this relatively simple circuit.

But in the Model 7, 1/2W Allen Bradley carbon composition resistors are used for R105, 106, 107, 108, and 109, instead of using an inductor.

I have had a question for long time, why the designer adopted the carbon composition resistors for π filters, because the one major merit of carbon composition resistors is that they are fully free from inductance.

Of course, the current is only about 10 mA, so that a definite amount of inductance would not be needed, I suppose.

Well, it was just three years ago now, that I found the resistor that I had long searched for.

At that time, I longed to have an exceptional sounding resistor for improving the sound of our Class-D amplifier. I had already narrowed the target to a wire-wound resistor, when I came across the Ohmite 200 series, the enamel coated Brown Devil.

I used this type of resistor when I built up the monaural DA30 SET’s for my friend Ookubo-san, 5 years ago now.

I recalled the sturdily-built sound of Ookubo-san’s SET’s at that time, and I did trials with the Ohmite Brown Devil resistor for our SPEC products. The result was beyond my imagination!

I made a hearing comparison of Brown Devil resistors for with values of 20W, 12W, 8W, and 5.25W, and they all had in common such rich timbre in the mid-to-low range, and I was impressed with the sturdily-built sound, just as I remembered from Ookubo-san’s monaural SET’s before.

The tonal balance was different for each one. The 20W Brown Devil had a very heavy tone, and 5.25W Brown Devil’s tonal character stood out in the mid-to-high range. I got the most desirable impressions from the 8W and 12W Brown Devil’s as well.

Well, when I measured the inductance value of of each Brown Devil, I understand the reason why.

The inductance value of 20W Brown Devil was the highest, and 5.25W Brown Devil had the lowest inductance value.

Anyway, I recognized the Brown Devil has some degree inductance, and I decided to change the Allen Bradley carbon composition resistors to Brown Devil wirewound resistors for R105, 106 and 109 of my Model 7 k.

But I could not get a 12W 2.2k Ohm Brown Devil for R107, or a 4.7K Ohm Brown Devil for R108, so I selected an Ohmite 20 type wire-wound resistor with an enamel coating, instead of the Brown Devil’s.

The results turned out marvelous enough, and I have enjoyed the clear but powerful sound with my Model 7 k after making the change.

For Handoko-san’s original Model 7, I used mainly Brown Devil 8W resistors because the 12W version is big and long, and makes installation too difficult.

The Brown Devil is a primitive, simple, and very old type of resistor, and has some inductance value, and so in a very real sense, the Brown Devil should be far from the ideal resistor.

But for some analog circuits, in which flows some amount of current, especially as a plate resistor for a driver tube, or a cathode resistor for a power tube, in a self-bias circuit, the Brown Devil can work very well because of its high reliability, and it brings out very real and musical sound, I have experienced.

I suppose that the inductive aspect of the Brown Devil rejects or does not transmit the harsh part of the spectrum in the current.

So I would like to highly recommend that you try the Brown Devil’s not only in your Model 7, but also in your favorite tube power amplifier.

Well, my way of the modification for power supply will come to a close for now.

It might seem complicated for you to try all the modifications I have written about at one time, but you could start first with just changing the rectifier to STTH6112, and installing the oil-filled capacitor for C1, and through these modifications, you could get a feel for how a modified Marantz Model 7 is such a “real” music reproducer even today.

To be continued.


Many thanks to Yazaki-san for sharing his life with the Marantz Model 7 with us!

There’s a lot to think about and absorb from what Yazaki-san has said in Part 3, just as in Part 2.

I’m in the process of digesting it all, while at the same time contemplating some adventures based on his recommendations!

I’m already looking forward to reading the next part of Yazaki-san’s article My Adventure with My Old Marantz Model 7k, and as Yazaki-san is known to say, please wait for it!

As always, thanks for stopping by, and may the tone be with you!

 Posted by at 4:10 pm

  26 Responses to “A Guest Article from Mr. Shirokazu Yazaki: “My Adventure With My Old Marantz Model 7” Part 3.”

  1. It is not the first time I have thought about the curious use of a half wave rectifier in this audio component. It certainly could not have arisen from cost concerns. They must have found a good reason to work around the compromises of this approach. You have to use a higher voltage transformer or if what was once a choke input supply will now require an input capacitor.

    I have a SALAS phono stage which is a DIY FET design.

    I am going to try the half wave with that incredible rectifier and hear if there is a difference here. The supply will be CLCLC and then to Salas’s onboard regulator.

    Easy enough to do.

    Mr. Yazaki’s reports are a gift. There is much to learn and use. A very generous man.

    • Thank you for your comment, Rick, appreciated!

      Yazaki-san is a true gentleman, and a very generous man.

      I truly appreciate Yazaki-san taking time to write down his audio wisdom for us, and each article has so many insights!



  2. Well, there does seem to be something to the half wave rectifier.

    Luckily my base power supply was very robust since this approach is not efficient. What had been and LCLC filter (5H/1500uF/5H/1500uf – the caps are giant PPs) now requires 14 uF as an input cap to have the same voltage at the input of the onboard regulator. The voltage drop from unloaded to loaded is massive! Does require some careful tweaking of the input cap and the bleeder resistor to get it right.

    The sound has the ease one expects when using the STTH6112TV2 but there is more than just that. There is something like digital noise removed from the sound. Of course, it would have made for a better comparison if I had been using this rectifier as a bridge. I was using a rectifier recommended by Mark Johnson of QUASIMODO transformer secondary snubber fame. It is obviously not in the same league as the STT… . The high frequencies are just there without any fanfare. Like the lesson I learned when I heard the HILL Plasmatronics – really fine high frequency reproduction, at first listen, sounds like they are missing since we have grown used to the resonances and grittiness.

    I was too cheap to use the STTH6112TV2s in this power supply, initially. I have used them in all of my amplifiers so I know how important they are. I was concerned the current draw was not enough for them to be at their best. I figure in half wave mode they are getting a good workout.

    Only had one on hand so I must order another for the other channel. Which I will.

    One can hear a soft clicking sound coming from the single rectifier. I have not snubbed the secondary of the power transformer but will later today. Will be interesting to hear if this has any affect on the clicking noise and more importantly the sound of the phono stage.

    So for those who want to take a step into what the Lord Protectors of Audio Electronics at DIYAudio would probably require a hospital visit to recover from you might be surprised at what happens. The folks at MARANTZ knew what they were doing.

    Needless to say filtering after a half wave rectifier is MUCH more important. I would think it best to use chokes instead of resistors but then I always do.

    Again my thanks to Mr. Yazaki and Mr. Day for a thought provoking article.

  3. Just looked and the STTH6112TV2 seems to have been discontinued.

    Luckily the other version STTH6112TV1 is still available at MOUSER.

  4. I can confirm Yazaki-san’s findings with the Ohmite Brown Devil vitreous enamel wire wound resistors in the power supply of my McIntosh MC240 tube amp. However, I chose to go with the 100kohm, 20W Brown Devil resistors for the R56 and RR60 replacements and the 10kohm, 12W Brown Devil resistor for R57. I also installed Sprague Atom TVA electrolytic caps as cathode bypass caps on the cathode bypass for the 12AX7 signal input tube. Those changes made a profound improvement in tonality for my McIntosh amp. Yazaki-san’s MC40 capacitor adventure for Jeff’s power amps really provided some great revelations on improving the tonality and soundstage of already wonderful vintage tube amps. Thank you again Yazaki-san!

    I recently moved from South Carolina to Minneapolis, MN with my family. Unfortunately, all of my gear will be in storage for the next 12 months until we move into our house. Until then, I will live my audio life vicariously through Jeff et al here at Jeff’s Place. That won’t stop me from browsing through the great vinyl stores like the Electric Fetus or visiting the Audiobook Research headquarters here in the Twin Cities.

    • Howdy Rich!

      Congrats on the move with your family, but I am so sorry to hear about your hi-fi gear being in storage! That’s rough!

      You’re always welcome here at Jeff’s Place, my friend!

      Kind regards,


  5. Dear Rick-san and Rich-san,

    Thank you very much for your comment!!

    I just sent the draft of Part 4 and the presentation sheets to Jeff-san,
    And the Part 4 is writtnen about my capacitor adventure especially
    of mica capacitor. I hope you enjoy !!

    Kind regards, Shirokazu Yazaki From Tokyo / Japan

  6. Dear Yazaki-san.

    Your articles are a masterclass of technical nuance in pursuit of lifelong study. Reading through this series inspired me to tinker more with tuning my audio equipment. I recently sought out an early McIntosh C20 preamplifier to partner with the EIFL Asano 2A3 amp Wakabayashi-san is graciously building for me. The input stages for the McIntosh C20 and the MX110Z are the same. As such, when I’ve developed sufficient soldering skills, (I’m decent but not great) I’ll be changing the input resistors and capacitors according to your suggestions you made for Jeff’s MX110 tuner preamplifier.

    By then, maybe I’ll be comfortable enough to modify the C1 and power supply capacitors 🙂

    Again, thank you for the inspiration.


    • Hi Grant,

      Just a note on my MX110Z: It is a very different animal compared to the Marantz Model 7 I suspect, and it’s hard to improve upon the overall balance that McIntosh achieved in the MX110Z.

      Steve Hoffman has told me that he has never heard a modified MX110Z sound as good as the stock unit, and I would tend to agree, as they are hard to improve on.

      The modifications I have written about ended up making my MX110Z very detailed sounding, but also too forward and harsh sounding, and I ended up taking out all the modifications I did to my MX110Z, and putting the stock components back in, with the exception of the prototype 0.22uF Duelund CAST tinned-copper capacitors that I put in the first stage cathode follower of the high level input, which are by far the best sounding capacitors I have heard in my MX110Z, and I’ve tried many at this point.

      As a refresher, in Modification B to my MX110Z I made changes to the first stage cathode follower of the high level input, by replacing two key pairs of 0.1uF capacitors at C93 & C95 and C94 & C96 with various pairs of 0.22 uF capacitors. The Duelund 0.22F CAST tinned-copper capacitors are the only capacitor in all my trials that I thought was truly complementary to the MX110Z’s voicing in my system.

      Of the three MX110Z’s that I modified, I had to restore all three to their original configuration with the exception of the 0.22 uF capacitors that replaced the C93 & C95 and C94 & C96 pairs.

      So I urge caution, as I have found that the changes that work best in the Model 7 have not worked well in my MX110Z with its different voicing.

      Having said that, these capacitor and resistors adventures have taught me a lot, and Yazaki-san’s methods are very important to understand, and have given me a method for voicing my amplifiers and preamplifiers that is outstanding, but the overall character of each component and its effect on the overall system must be taken into consideration, and of course the tastes of the individual, or you could end up with unpleasant results.

      If you do experiment with your Mac I would limit what you do at first to changing out the the C93 & C95 and C94 & C96 pairs with a pair of 0.22uF capacitors.

      For resistors, I have not found anything that I thought improved upon the NOS Allen Bradley carbon comp’s musicality in my MX110Z, and anything else I’ve tried (quite a few different resistors now) resulted in a more detailed, but also a harsher, edgier, sound that I found unpleasant.

      Think of these types of modifications like wine pairing with a meal. With a nice filet mignon one wine will be a better match than another, and sea bass or chicken will require something altogether different.

      Or like spices when cooking, certain spices will make a particular dish ‘pop’ and come alive, and be totally unpleasant in another.

      Modifying components is very much like the two culinary examples, and when hot-rodding a component one size does not fit all.

      Consider trying Yazaki-san’s methods, as they are very good, but then take into account also the difference in components, because what is complementary in one may not be in another, as I have found.

      But try a few things and enjoy the adventure, it is a great learning experience, and a very powerful way to understand how to voice your components in the way you want them to sound to complement the rest of your system.

      Kind regards,


      • Hi Jeff.

        Wow. Thanks so much for the rather detailed response. The C20 doesn’t get here for another week, but I don’t figure I’ll be crawling around inside it for at least several months (unless a cap or resistor fails in the mean time, of course) as I learn what I love and don’t love about it.

        And I wholeheartedly agree with you about making single changes at a time, learning how those changes effect the whole, then proceeding from there accordingly. Especially when it comes to voicing through modification.

        Just a few days ago I just bought a tidy pile of old mono amps, tuners, and tubes that I’ll be cleaning up and using for my sandbox. They were already on the verge of a one way trip to the dump since the guy was leaving for Boston in three days, so if I do a little permanent harm to one while learning, it’ll just be breaking an egg to make eventually make better omelettes. As a safety note, I’ve got a very experienced buddy to watch over my shoulder as I get started and learn how to properly prep a high voltage device for repair/mods.

        Also, while output stages are basically identical, it’s my understanding that, due to the phono compensator, the McIntosh C20 preamp has far more in common to the older C8 mono preamp compensator than to your later MX110Z tuner preamp. As such, I’m doing like you suggest and taking a heavy grain of salt with suggestions for capacitor/resistor types to used as mods/replacements.

        What struck me as the most sensible is Yazaki-san’s focus on where he chose to start, the input stage, and why. I see the same stage has remained your focus with the Duelund cap swap you linked to.

        As a thank you for that affirmation, I’d like to share something with you.

        A piece of gold fell into my lap yesterday. And that gold is a pristine example (and potentially prototype considering the source and that it lacks a serial number) of the McIntosh MCP-1 moving coil preamplifier with four optional impedance levels to match to the output your given cartridge. The MCP-1 is currently matched to my Zu DL-103 at 40ohms while Brahms is spinning on the Garrard, and totally outclassing the step-up transformers in my EAR 834P.

        If you can find one, it’s worth it.

  7. True impact out of Yazaki-san’s Marantz#7k meticulously engineered for 38 years.

    It has been so powerful that I have to admit that I need to reconsider the importance of preamplifiers. I happened to have the opportunity to be able to ear-witness Yazaki-san’s Marantz #7 k (later referred to as M7k) (line amp alone) as it was sent to my laboratory for its health check.
    This is the story.
    The impression of tonal quality: wide-range, deep low-end, brilliant highs and deep mid-range.
    I know that the sound quality is a subjective matter, but M7k line amp sounded as above, despite sophisticated circuits for tone-control (bypassed in flat position) and requires four capacitors for sound to go through to the output.
    Readers will find how Yazaki-san checked and changed some of the original parts as in the photos.

    For curiosity, I conducted some ordinary health check.
    M7k showed very low noise level for an entire measuring range (10 to 200kHz) with no line (50Hz) related friends showing a good power supply filtering not to mention the THDs and IMD. No specific problems were found out of the data, of course. M7k line amp gain setting was about 12.5dB bypassing original gain VRs by good quality TF resistors.
    The preamp I use is my own design using only one tube with a gain of approx. 4 (non-NFB), almost identical to that of M7k. It is because the loudspeakers are so efficient about 97dBSPL/W/m (TAD unit base 2 way “Exclusive 2401W”) that their input power rarely exceed one watt at normal hearing SPL.
    What we hear is the sound of capacitors, as one of the writers in MJ magazine to show readers his superiority in circuit design vis‐à‐vis old tube amplifiers. If you want to feel musics as recorded on site, please forget this kind of philosophical discussion. Musics are full of harmonics. Therefore if your amplifies are reducing those harmonics, they may sound poorly, no matter how well they are publicized.
    The most critical judgment should be done by your own ears alone. Other factors are secondary. When I listened to M7k before I send it back, I felt musics before my loudspeaker setting. Mute trumpets and accordeon are swinging. Hearing is believing. If you can’t get or feel what you want, it may need your second thought. Then the main culprit may not be your main amplifiers rather it can be your preamp. In other words, control amplifiers literally “drive” your sound. Congratulations Yazaki-san!

    • Dear Kato-san,

      Thank you for your excellent comment about Yazaki-san’s modifications to the Marantz Model 7k!

      Kind regards,


  8. I use Miles 12 Watt 7.5k Ohm driver tube (6F6) cathode resistor in my 300B amplifier.
    Is it possible to use Ohmite Brown Devil 7.5k Ohm instead of Miles as cathode resistor ?
    The power dispersion on this cathode resistor in an idle state is about 3.3-3.5 Watt.
    I see 20 and 10 Watt 7.5k Ohm Brown Devil resistors for sale on eBay.
    Which resistor power rating will be enough?


  9. Hi Jeff,
    Thank you. I will write him.

  10. Hi Jef!I from Vietnam.I also have one Marrant 7C but it’s power capacitor seem reduced values.I try to replace them but only 2 of them success.for the 20-20-20uf-350v I can not find once suitable to it. I replace by Authentic capacitor.could u please show me what once can be replace for it?

    • Hi Đuc,

      I’m not really familiar enough with the Marantz 7C to make a useful recommendation, sorry about that. I’ll ask Yazaki-san if he can make a recommendation for you.

      More soon!

      Kind regards,


    • Dear Duc-san,

      I’d like to search for its replaceable capacitor and so please wait for a while.

      Best regards, Shirokazu Yazaki

    • Hi Đuc,

      I don’t own a 7c, but from what I can see on the pictures (on the web), you’ll need something that will fit in the same slots/spot.
      This might be an option:
      ‘Ask Jan first’:
      The KTL35 is a 20-20-20 550V, but do check the dimensions to make sure it’ll fit (or ask Jan….).

      Best regards,

    • Dear Duc-san,

      It’s too rare to find out the completely compatible capacitor of 1 inch dia.
      But I come across an idea and so if you could, I’m willingly to share it to you.
      To tell the truth, I’m facing on the same problem to find out the reasonable
      way about the modification of my friend’s Model 7 just right now.

      Please feel free to contact me below —-,


      Regards, Shirokazu Yazaki From Tokyo,

      • Dear Yazaki-san,

        Thank you for reaching out to Duc-san to help find the right capacitor choice for his Model 7.

        Kind regards,


  11. Hi!I see som sprague capacitors have dimension same original but it just double capacitor. I’ve just also order new model diode to replace replace for the original.because I relize it have problems with supply the vol DC, the original schematic is 328v DC but my just have 315v. Now i have trouble to mount it, i see the AC just have one line but the new diod need to conect 2 line of AV suply.please help me how to conect the new diode to replace the original.

    Best regards!

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>