ART USB Dual Pre
Behringer B2 Pro
Denon DP-61F Hafler DH-110
Hafler DH-500 Hafler XL-280

  Klipsch KG1
Mogami/Canare RCAs
Nakamichi CM-300
Scarlett 2i2
Tannoy 603


  • Todays mailorder tubes are commonly pre-graded for noise, matched, and sometimes even burned in before being shipped...but things were much different a few decades ago. In the 1950-60's tubes were commonplace in TV and hi-fi equipment and it was normal to drive to the corner hardware store for replacements. Some US drugstores even had a self-service “U-Test-M" tube tester so people could bring in their own tubes and test them on the spot. You plugged your tube in the proper socket on the kiosk and pressed the 'test' button. The needle on the machine rated them as "Good", "Weak", or "Replace". While the manual states that the ST-70 came from the factory with matched tubes, there was no such thing at a drugstore in the 1960s. You bought what was sitting on the shelf.

    As we all know, the stock Dynaco ST-70 amp used four EL34/6CA7 power tubes...the rectifier tube was a GZ34/5AR4 and the two preamp tubes were 7199s. Sadly, production of the American 7199 ended a few decades ago but the Russian tube company Sovtek continued to produce them until 2007. NOS 7199s such as RCA, Sylvania, and mil-spec versions were still available online up through the 2000s but today all supplies are exhausted. I find it strange that with the large variety of tubes still being produced that no-one has re-introduced the well-known 7199. How come China has never made one? Luckily, as far back as 1994 drop-in boards using the substitute 6GH8 were available...people were smartly preparing for the 7199 demise even then. NOTE: the extinct JAN (Joint Army Navy) Philips 7199 was specially graded for military use and was available online until the 2000s, I should have bought a pair when I had the chance.

    When I began my ST-70 mod, I decided not to get sucked into the endless argument over whether the original 7199 was better than the 6U8A/6GH8A. My belief is that if you replace the main board and preamp tubes with different versions its not really a ST-70 anymore. So my plan was to take the original amp and enhance the circuit through better parts and retain as much of the original circuit character as I could.

  • Above is the ST-70 which I bought used in 1994. I have the invoice from the previous owner...they purchased it used in Alexandria, Va...on December 31, 1965 for $69.50. Instructions on the invoice read "deliver after 3pm". Sounds like it was intended for a New Years Eve party that night(!)

    It has been said that only 5% of the 300,000 ST-70s sold over the years were factory wired, the rest being kits. At some point in their production Dynaco began supplying the factory wired versions with a smaller sized manual, rivets instead of screws in places, spanner-screws on the bottom plate, a cage over the circuit board, and a sticker on the back. My bias pots are dated 1963, my manual is a large 8x10 stamped "Factory Wired" and I have none of those extra features. From what I have read these features didn't begin appearing on the amps until later than 1965 or so, when Dynaco began calling the factory wired version the ST-70A(assembled).

  • My amp came with the original Dynaco 7199s, Mullard EL-34s, and a GE 5AR4 rectifier. It powered up fine and the tubes biased correctly. But when I looked at the original circuit board with ancient film caps and grungy resistors I decided those parts had to go. I had stumbled across the July 1982 issue of Audio Basics. which detailed some really interesting mods. It retained the original design but made some simple changes that made the amp much more linear and musical without altering the ST-70 topology. All I needed was a new board to experiment on.

    But in 1994 no one was selling 7199 replacement boards that I could find. So I came up with my own solution. I bought an etching kit and by using the old-school method I traced the original board and etched a "new" ST70 board to stock dimensions. I drilled holes for the parts and ended up with a nice quality board. This let me keep the board "stock" and retained the 7199 tubes. While the traces may not have been audiophile grade copper, it has been troublefree since 1995...23 years(!).

    For resistors I selected Holco H4s which were considered the best sonically. Four large 1mf "Wondercaps" were installed to replace the original .1uf versions which restricted bandwidth between stages. A yellow .047uf MIT Musicap and 390/82pf mica rounded things out. The original Mullard EL-34s showed as tired on a tube tester so I replaced them with a set of Tesla EL-34s. But after only a few hours one Tesla started sparking so I replaced them a matched quad of Electro-Harmonix EL-34s. Because Sovtex 7199s were still available I bought a pair.

  • As you can see above this amp is not a concours restoration. While I spent some time fabricating the new circuit board and Cardas signal path wiring I kept the chassis layout pretty much stock and left the wiring a bit messy...even using plain-screw terminals for the speaker wires. I like the idea of taking a vintage amp and using the original circuit with minor enhancements, rough edges included.

    The 1982 Audio Basics article mentioned using a cap/resistor combo on the RCA input to filter out the extreme treble and extreme bass. This prevents the output transformers from becoming oversaturated with frequencies they can't handle...reducing distortion and increasing clarity in one step. This filter and the larger interstage coupling caps were the only changes made to the signal path...a mod I consider non-intrusive.

    Later I installed the large SDS Labs power supply board on the left. While the original year old quad-cap had worked OK in 1994, I was certain it was dried out and a performance bottleneck. However, new quad caps sold in the 1990s were the wrong dimensions and didn't seem to offer any performance advantage. Then around 2003...I stumbled across the SDS Labs drop-in power supply board. I immediately bought a bare board and selected my own parts....82/150/150/100uf 400V caps and bypassed them with .1 polypropylene 400V caps underneath. I disconnected the old quad cap but left it in place...I also installed a .01 film cap across the power switch to remove any snap at powerup. I left the original choke, dripping wax and all.

    This new power supply with the stock rectifier tube made a big difference. The sound took on a more focused character with a quieter soundstage. Dynamics were more punchy at higher volume, especially the bass. Transients were more musical, open, and accurate and popped out of the speakers. The best description is that the amp picked up the dynamics of a transistor amp while retaining the tubey smoothness of the old design. Another benefit is that I can detect no hum even with my ear pressed to the tweeter.

    NOTE: I had replaced the power cord with a 3-prong but found that connecting the ground wire to the amp chassis created a ground loop. When my preamp was connected I could hear a hum from the speakers. Clipping the ground wire from the power cord removed the hum. I have read that others also experienced hum from using a polarized AC cord.


    In 2017 I started looking for something new to upgrade. I wanted to keep the original chassis, transformers and 7199 circuit...but considered the bias circuit the weak spot. I had noticed that Dynakit made an interesting kit that allows the bias for each tube to be dialed in individually. It replaces the coarse 15.6 ohm circuit with a trimpot into the center of each preamp socket in the front...a great use of space. The concept is that while tubes get matched for plate voltage and transconductance each one may require different voltages to bias properly. To use it, you balance/zero out the voltage between tubes from the front trimpot, then set the master bias from the rear trimpot, easy.

    I found adding the Bias Balance Kit did make an audible difference. Not sure what to expect I found that the soundstage was cleaner and more focused. Details seemed cleaner as though fuzz around the transients had been lifted. It had a sense of being "dialed in" which of course it now was. It makes sense...precisely setting the bias on each tube will lower crossover distortion more than using a single setting for each pair. I decided I didn't like the aluminum shafts peeking out of the center of each preamp I used a black marker on each to mask them (below).

    I did notice something interesting however: the mod had me remove all the old, unused voltage wiring that runs from the EL34 sockets to the front-panel preamp sockets. I realized that having all this inductive wiring sitting idle but wired into the main loom likely adds noise and who-knows-what-else to the circuit. In other words, unless you still use the original preamp sockets does anyone?) I believe removing the associated preamp socket/tube wiring is beneficial, even if you don't use a BBCU-1.

    NOTE: Be careful cleaning the metal with chrome polish as many have ammonia in them which can add haziness the finish. Other metal polishes are abrasive which is how they polish oxidized metal..fine on a car wheel but not on thin nickel plating. I found that a 'magic eraser' safely removed grime and some discoloration and when followed with a microfibre cloth produced a nice luster. Armor-All style protectant on my transformers gave them a like-new look.


    When it comes to a listening session with a vintage tube amp...I have to admit that sitting down with a nice drink...powering it up and seeing the tubes glow...and hearing music from a 53 year old device that "makes music from light-bulbs" is a sublime experience. My listening 'salon' uses Denon DVD-2900 and Sony DVP-9000ES disc players and a modified Hafler 110 preamp. Because of the limited bass response of a ST-70 I am using a smaller set of Tannoy 603s on speaker stands.

    What is apparent is that with the new power supply and signal-path mods this vintage tube amp really shines with the right material. Music with lots of dynamics, percussion or loud passages, can overtax the circuit and produce a flat unrealistic soundstage. Turning it up louder increases the volume but not the dynamic range...resulting in a mildly congested sound. I suspect the dynamic range of the digital recordings exceeds the response of the input-section of the amp (TIM distortion). What this amp really likes is quality jazz, acoustic, and vocal music with lots of soundstage and harmonic overtones. Its as though the circuit enjoys portraying real instruments in real space.

    I tend to listen to newer recordings from Pat Metheny or the Alex Skolnick Trio and shy away from older analog recordings. But a CD I had ignored on my shelf jumped out at me: Homecoming by the 1970's band America. The acoustic tunes on this CD (tracks 2,4,6) all use acoustic guitar miked with a wide stereo chorus effect and the vocals have a slight reverb and beautiful presence. Check out "Moon Song"....this track must have been created with the ST-70 in mind.

    DYNACO 7199 vs SOVTEK 7199

    Rather than speculate whether original Dynaco tubes contained some sonic magic from's what my ears told me. My 1963 7199s have a flatter, steelier soundstage. Acoustic guitar is clean but zips lack the sparkling edge that makes it sound live. Music is pleasant but softer and some detail seems lost, I felt like the treble had been turned down slightly. The Sovteks were a different story. Very quiet and open with sharper transients I found them more dynamic with real treble. Details tended to pop-out more, with midrange details more open and accurately placed. Much more musical. While some purists have blasted the Sovteks I'm not sure why, as in my system they trounce the old Dynacos. My old 7199s could have been worn out, but that's more of a reason to use the Sovteks.

    GE 5AR4 vs SOVTEK 5AR4
    A while back I bought a Sovtek 5AR4 as a backup rectifier. I forget the exact numbers my voltmeter showed the Sovtek put out 20 or so less volts than my old GE 5AR4. The sound was slightly less dynamic as well. So nothing is wrong with the Tesla but I prefer the feng-shui of my vintage rectifier tube.

    One mod which has been around for decades is to convert the amp from ultralinear pentode to triode by a simple wiring change. This is said to increase sound quality but also lowers the output of the amp from 35 watts to about 10. When I converted mine it took on a smoother sound with a deeper soundstage...a pleasant sound. But I found that the dynamics seemed more restricted. So while triode mode may be nice to listen to music at low levels, I prefer the higher pentode output and more dynamic sound.

    ABOVE: this is the original 1963 circuit board with carbon resistors of questionable tolerance and noise levels...and ancient-dielectric capacitors. I'm as sentimental as anyone else for vintage parts but leaving these decrepit items in a modern signal path doesn't make sense. To the right are the original Dynaco 7199s and a Mullard EL34. Ā© 1997-2017
All mods are illustrative only, perform at your own risk.
Datsun is a registered trademark of NissanĀ®