A while ago browsing the web to enrich my knowledge, I stumbled on a phono schematic around the C3g. Well known in the hobbyist world and a must try for many, this tube just average in triode mode is a real champion as a pentode.

I discovered his great qualities while seeking a first stage tube for my LCR phono stage. Sound was really involving, but gain was too high for comfortable listening in my setup and I did not persevere in that direction until today.

The scheme I found is very simple, but I noticed a few design mistakes that will surely affect the final result. A weird RIAA network and a rather high output impedance. Both easily fixed. 

Network is passive and when properly calculated can give a very accurate correction.   In that way the best is to follow Lipschitz 1979 publication. To make it short we have to keep in mind that network is dependent upon first stage Z_out, second stage input resistance Z_in and Miller’s input capacitance C_in. This last point is often neglected and a reason why you cannot change tubes without modifying network time constants.

Now, components calculation is easier. We just have to write down all we need.

 1 – Time constants

                                            R’₁C₁ = 2187 µS 

                                            R’₁C’₂ = 750 µS 

                                            R₂C₁ = 318 µS 

                                            C₁ // C’₂ = 2,916 

 2 – Components value 

                                            R’₁ = ( Z_out + R₁ ) // Z_in 

                                            Z_in ~ 50 Z_out 

                                           C’₂ = C₂ + C_in 

First stage gain, Zout and bandwidth

A pentode gain is merely equal to tube Gm by load. If we want to keep second stage input resistance in a reasonable value (~ 500K), C3g load have to be something like 10/15K to respect Lipschitz recommendations.

A graphical study will help find the best operating point and I finally stopped on 12K@10mA. Gain will be very close to 150 and Z_out 11,6K assuming a 400K internal resistance. Rg₂ (Z_in) can either be 510K or 560K depending upon RIAA resistors availability. These have to be 1% or closer tolerance.

To keep gain high G2 is fully decoupled to the lowest frequencies. As a bonus, decoupling dramatically reduces the grid-anode capacitance. Effect on bandwidth is a huge improvement, with high frequencies well above 100Mhz. To prevent any unwanted oscillation it is necessary to use a grid stopper close to G1 pin.

Second stage gain and Z_out

I did not go with a 5687 like in the original design. Amplification factor is too low and it needs a lot of mA to sound right.

Because gain of a standard phono preamp is about 300/350 or +50dB, I have to find a tube with at least a gain of 20 to compensate RIAA network insertion loss (-20dB). After much thought to decide if I should use a high µ triode with degenerative feedback or a classic medium µ anode follower, I stopped on the great sounding E182CC fully bypassed.

With a µ of 24, a low internal resistance and quite a high Gm (15mA/V), it is a good G3g companion. Moreover, the second triode will make a nice constant current coupled stage. I used this topology in the E88C/E86CC LCR phono preamp with great satisfaction.

Output impedance will be about ρ/µ+1 (~65Ω). This is far better than the 4K of the original design and will allow good frequencies transmission even with long cables. 

preamp schematic 

more to come...

Supply and more 

Classic C input supply with two LC cells for excellent ripple attenuation.  

Two other RC cells will floor the rectifier residual to negligible value. Especially important for the first stage where signal is in the mV range.

Construction with high quality parts.

I used both electrolytics and paper capacitors. I found the combination well balanced and use it with satisfaction for years now. Paper and aluminum foil are very reliable with a super fast discharge time. These old caps are a must in power supplies along with paper in oil. Unfortunately these are usually too big to fit the chassis I use

I built the preamp on a soldering board, old-fashioned, but very practical and compact. Therefore, it is possible to change a component very easily if necessary.

RIAA capacitors are Philips KS 1% polystyrene and Stéafix silvered mica. These are in parallel to give C5 a correct value and soldered on the opposite side. Value chosen taking in account the E182CC Miller capacitance. About 100pF.

Resistors 1% MRS 25. I use these for a long time and are perfectly suited for this task.

Pay attention to solder load resistors away from the board, they become hot in operation. I chose Kiwame 5W ones.

Grounding needs special care. To avoid hum and ground loops I used a single wire for each stage connected to a central point on the ground bus. Heaters are well filtered and polarized to prevent cathode leaks from V3.

C3g are Siemens, the only ones I stock. Other makers are TFK, Valvo, Lorenz, ITT.

The E182CC is a Philips design and sold under many brand names. Most come from Heerlen plant. Mine are RTC . These professional tubes were sorted at the factory to ensure a long time use and excellent reliability.

The making of

Listening report

Untill today I thought RC network was inferior in all respects to LCR one. Nothing is more untrue.

I am playing this preamp daily for more than a month now and it is just amazing. The most obvious quality is its dynamic capability, but it is not the only one.

Music is vivid, colourful and alive with excellent imaging. Soundstage is wide, not as deep as the LCR. Low end is very tight, ample, well controlled, but slightly rearward. The most remarquable is the medium / high that deserve an A+ note. Clear, precise, well focused and detailed to the extreme. No harsh or aggressivity even on loud music playing trumpet, guitar or drums.

On the very demanding Isao Suzuki trio record, snare & cymbals are rendered with incredible realism and I can feel cello's attacks like never before.

If I had to give on overall appreciation I would say the LCR is sensual and lush while the RC is more Rock n Roll and appealing. These are great qualities considering the music I listen to and it is very addictive. Today my LCR preamp is mainly used for human voice and orchestral music.

Of course this is very subjective and depends greatly upon room and overall setup. For this listening tets I used my old Garrard fitted with a FR24 arm, a Entré EC 30 cart (a rarity by Matsudaira-san also renowned for some of the best carts ever: Kiseki, Acuphase, ZYX and now MySonicLab) and its dedicated ET 100 transformer. The C3g phono is coupled to the 6J5 line preamp driving my MC1/60 SE amplifier.

Speakers are heavily modified Klipsch La Scala with upgraded network by Alk (all original plastic capacitors changed by waxed paper & foil, teflon and some excellent sounding Aerovox V161), Martinelli's sanded wood horn + JBL 2470 and Beyma CP25 tweeter.

Great jazz vinyls for this report

Among the numerous triodes suitable for an audio project few are indirectly heated ones. Saying they are sonically inferior to the directly heated ones or less audiophile would be restrictive and unfounded. Main reason is that they are not numerous and nowadays scarce.

One of the most notorious is the French R120. A PTT tube beloved by Japanese who bought almost all the existing stocks decades ago. French where not the only ones to develop such tubes.

German RL12T15 is another good example of a very linear triode who prove to be a great sounding tube. Its somewhat limited use comes from a proprietary socket a nightmare to find. American industry was not left out by offering the 6A5G created on the basis of a 6B4G, itself derived from the famous 2A3 / 6A3.

The 41MXP comes with or without carbonized anode, same sound.

Original datasheet does not give much information and I couldn’t find the usual Ip/Up curves. Not a real problem, Cossor gives the best working point, load and voltages. We can see that the 41MXP is not a very demanding valve, 200V on the anode at -12,5 V grid. Unusual for a triode that generally bias much deeper, it eases construction a lot.

Remarkable fact, internal assemblage reveals a second grid factory tied to the anode. This feature supposedly intended to increase HF extension by reducing Miller’s effect is seldom seen. The above cited R120 and the DA30 are other examples of such arrangement.

Presented as a super power tube, a well-founded superlative, this cute triode possesses a lot of delicacy and definition paired with a loud and authoritative voice despite a modest 1.4W output power.

To match such talent, I needed a driver of equal quality. My choice stopped on the ML4, another great British valve completely underrated and left aside by the audio community without reason. During the golden age of vacuum tubes, it was the power horse of many amplifiers. It’s low mu and great linearity where useful to drive tubes like PX25 & PX4. With an internal resistance of 2.8K this tube is suitable for transformer or choke coupling. I chose the second option because at about 280V, high voltage is too low to properly load and bias the ML4 with a resistor.

I have on my shelves a pair Tango TC 160-15 audio chokes. Nice old Hirata iron bought decades ago and a perfect choice for the task.

Nice looking and great sounding Osram ML4.

Another GEC CV1732, smoked bottle, black plate in typical military boxe.

schematic is very straightforward and the parts choice will determine the final result, especially the coupling capacitor.

Supply

Choke input is the way to go for most SE A1 amplifier with moderate current (below 100mA). Two good reasons:

1- Ripple is very low behind a choke, equal to 

V _Ripple = V_out / 6 π² f²LC√2

almost a perfect sin wave with very little transients in power transformer unlike input capacitor filtering.

2- Rectifier choice is greatly facilitated as we don’t have to deal with complex calculations mandatory for operation with capacitor input supply. Usually, tubes makers provide all needed information for choke input in their datasheet. You will always be safe if you stay well within the max peak inverse voltage and peak plate current.

For basic ripple and time constant calculation refer to LCR Phono preamplifier Part 3

Step by step construction

Holco, Kiwame and Allen Bradley resistors. Aerovox V161 coupling capacitors (IMHO best for the price with ITT PMT/2R when you are looking for a neutral well balanced sound).

100Hz & 10KHz squares

Listening report

Setup: Garrard 301, FR24/Entré EC30, UA7082/AT33 Mono, C3g phono preamp& 6J5 line preamp

I built several amps until today and I can affirm that this one is a top contender along with the Philips MC1/60 and the Visseaux A710

Vinyls, vynils & vynils encore

New tube, new sound. A while ago I stumbled on a little batch of NOS Visseaux 46. This tiny valve has the same size and shape than the world famous 45.

A few aficionados have used it, but despite very good results it stays a marginal tube. Main reason is the flea power you can expect in SE setup. It is in the same league as the 10, about 1,2 W. Not an issue with efficient speakers.

Those who have built an amp with are unanimous in saying that this little bottle contains a treasure of refinement and delicacy. It may even be better sounding than the 45 with a deeper and tighter low end. From this point of view it can be considered a 45 + 1 because it has an extra grid. However, it is not a tetrode and the data sheet confirms its use in triode only. The characteristics and operating point depends on how this additional grid is used. Low mu triode with G2 connected to the anode, high mu with both grids tied together. Only the low mu setup is practicable in hifi, especially in single ended operation.

A parenthesis to say that some rare tubes have the particularity of an additional grid connected to the anode. Best known are the French R120 or the British DA30 and the 41MXP. All sold as real triodes.

The 46 is a piece of history, an old tube developed in the late 20s and sold by all major American companies. It saw long service and use in radio receivers and was made in France under license by Visseaux, Radiotechnique and Mazda. Even if the 45 and the 46 are very similar, they are not interchangeable mainly because of the UY5 socket of the second one, but not only. Anode voltage, bias point and filament current are different.

Mine were probably made before WWII. A set comes in the usual cart board wrapping typical of military storage, the other one in Visseaux commercial blue and red box.

The 46 is pretty easy to drive, a favorable point to pick a good sounding driver among the numerous medium/high mu triodes available. For this purpose I chose a 7 pin miniature double triode primarily intended for computer use. The E92CC and it's lower mu relative the E90CC are inexpensive tubes with excellent tonal qualities and reliability. Paralleling both units gives a rather high transconductance tube, means low noise. Linearity is good and when properly loaded distortion is obviously very low.

I used a shared current coupled stage as driver. It is the same arrangement implemented in my phono preamplifiers. This setup widely used in professional equipment is to be considered a single unit providing high gain under low impedance. This helps achieve a more solid and defined low end from the final tube. Calculation is pretty simple as long as both tubes are Gm matched. From that point of view the TFK E92CC were made with tight tolerances, it helps a lot to sort two good pairs.

I first calculated the anode follower load for gain, bandwidth and distortion paying attention to keep enough grid headroom. Then used the same resistor in the second tube cathode. Stage is self balanced, there is nothing else to do, just check that voltage swing stays in the linear area of both unit. Additionally measured the CF bias which should be equal to the AF stage. This setup works great with high Gm triodes running some current, but can be tricky to implement with tubes like ECC81, ECC83...

Setting anode resistance to 22K gives a very similar AC load due to the CF high input impedance. Gain in the first stage will be   G = µR_L / R_L + ρ = 34  and overall one about 33 with the cathode follower.

The 46 is biased according to manufacturer specifications : 250V, -33V, 22mA. Preferred load 7K.

A 1.5K/10W resistor will make the job. I personally favor non inductive ones and used a Sprague Koolohm along with Philips vintage LL capacitor. These electrolytics are super reliable with great sonic qualities. Other components are Allen Bradley, Kiwame & Beyschlag resistors, Aerovox V161 coupling capacitors. Supply uses SEL PMP and ICAR Polypropylene capacitors.

During the European Triode Festival 20in21 in Bellême I bought a gorgeous pair 46 globe by RCA Cunningham. A dream for any tube lover.

Like the 41MXP SE amplifier, a choke input filter is the way to go. Excellent hum rejection, almost a perfect sin wave easy to smooth out and by far a less stressing load for the power transformer than a capacitor. The only concern is a higher AC voltage to get the desired DC supply.

Although first cell time constant is usually higher with a choke input in conjunction with a large capacitor (the total series resistance before the first filtering capacitor includes the choke resistance *), the transients are rendered with better dynamic.

* A reason why transformer makers pay great attention to get the lowest one when winding coils. Main drawback, chokes of that type are usually bigger and more expensive.

Construction is kept as simple as possible. One point wiring on a heavy ground bar and short wiring, whenever possible, are the basics of trouble free building.

As usually, just a few components of the highest quality. The heart being the output transformer plus a special input choke and oversized custom made power transformer.

Sub chassis with 46 heaters assemblies

On the bench for sin and squares. 3,26V_rms /8 ohm or 1.32W at clipping

Family grows up...

Sumptuous Cunningham globes and Telefunken baby bottles

Small Visseaux 46, great sound too

Listening report setup

Various vinyl's on my Garrard 301 + FR24 and Entré EC30 cart or JVC 7082 and AT33 Mono, C3g phono preamp, 6J5 line preamp and homebrewed Klipsch/Altec multicell speakers.

The 46 is a very nice little tube that sings on any kind of music. He knows how to be enchanting or even glamorous, especially on female voices. Despite a limited power, its voice is authoritative and can be capable of great energy on dedicated speakers. The most surprising is the low end reproduction, firm, powerful and ample, comparable to the big MC1/60 amp. Very different from its almost twin sister, the 45.

I understand enthusiasm for this triode which combines finesse and firmness. This tube can delight music lovers whatever the material except, perhaps, high level rock fans.

A few month ago I sold my E88C/E86C LCR Phono preamp, playing the recently build C3g/E182CC RC one. I soon realized I couldn’t live without. Not to say the C3g is inferior, it is the most enjoyable preamp I have today but I miss the absolute delicacy and sweetness the LCR had. My wish would be a preamp combining all qualities, even if everyone knows that the search for the ultimate phono, the one that brings pure enjoyment on any kind of music, is a never ending story.

Whatsoever, it didn’t take a long time before I decide to undertake a new one. New project, new interrogation. Is the final result a matter of network or tube choice, what has the most significant impact on sound reproduction ?

I personally think that active components, the way they are used, the dynamic behavior and the distortion figure under use conditions, have a much bigger influence upon sound than the EQ itself and decided to keep the same approach than in the RC phono, a first stage pentode instead of a triode.

Fortunately, when cost was in par with my wallet, I stocked a few transformers and can build a complete unit with the same irons previously used in the LCR phono. Just had to buy new EQ set.

Temptation was great to use the same tubes with the benefit of an A/B comparison, but the C3g gain would have been too high especially on a 20K load. I had to find a substitute with equal qualities and not too exotic. The search for the (almost) perfect equivalent took some time and I finally stopped on the E83F. A high Gm pentode, easy to source, apparently low distortion, low noise and primarily developed for class A audio applications in telephone repeaters. Surfing the Web I discovered it has been used in the last stage of the iconic TAB Telefunken V76 microphone amplifier. Choke loaded and capacitor coupled to the output transformer (full shematic here). This preamp has a reputation of excellence. A serious reference and a great encouragement for my project.

First stage

Beyond the basic DC study from datasheets, my old HP dynamic analyzer was of great help to set the best operating point and get a similar harmonics distribution than the C3g. Once done, gain is close to 160 with very, very low distortion.

Second stage

RIAA network plus transformers (A8713 20K/600 and TKS50 600/50K) has a 0.158 overall gain which commands an amplification of 15 to get +45/+50dB. What immediately comes to mind is the 6SN7/12AU7 family of tubes. Having a good supply of miniature double triodes, it was obvious to give them a try.

From the dozen references tested I only kept six. All have outstanding performance, with a personal preference for Mullard M8136/CV4003 and Tungsram E80CC. These two fine tubes are just followed by Siemens 5814A triple mica and RTC 5814A. The others (Sylvania 6072A & GE 12BH7A) being slightly less involving while keeping all other qualities like great tonal balance, wide soundstage and lifelike restitution. Distortion was a bit higher with the12BH7A and 6072A had too high gain resulting in a forward presentation.

DC point is set according to manufacturer recommendations. Below the E80CC setup, identical for M8136 and 5814. The 12BH7A needs a lower bias resistor

Schematic is very similar to the full triodes preamp

Supply, parts and making of

Wishing to keep this phono as close as possible to the first one, at least in design, I calculated a double choke C input supply. For basics refer to the LCR supply post.

R7 serves two purposes: draws the necessary current to set DC voltage and acts like a crude shunt regulator.

Just one capacitor in signal path and another in G₂, its choice will greatly influence final balance. Not really fan of mellow or colored restitution, I prefer a neutral sound, less appealing but more realistic. A few, very few caps have such ability and they are not super exotic, fancy, expensive…. ITT PMT/2R and relatives like LMT, Aerovox V161 (polyester, used in original Pultech equalizer) are close to that ideal. Finally choose the PMT/2R, a long time favorite. Thanks to Hiraga San.

During the first listening test I noticed some hum, about 2mV, a bit annoying. This is a recurrent problem with pentodes (partition noise & HV supply ripple) and to fix this issue I replaced Rg2 carbon resistor by a metal film one and added a hum cancellation capacitor in parallel. Calculation is tedious (RDH 4th Ed. p539) for precise cancellation and usually not a standard value. I used 0.15uF for 0.192 calculated and got 0.3mV residual noise. Inaudible at listening distance and barely discernible close to the speaker, I can now run intensive listening tests. Also added a 22K grid stopper.

inside PSU

Mullard 5Z4GY rectifier is the best performer in my system, great presence with deep bass.

Phono completed

E83F soft glow

Measurements

First observation, the preamp have a great dynamic potential and huge headroom. With 80mV input signal I got more than 30V peak/peak with no sign of clipping. Couldn’t go beyond because my scope was on the lowest selectable sensitivity.

Next step was a THD test. Just have a look at the pictures to see how good this preamp is. Considering signal generator + dynamic analyzer total harmonic distortion (~ 0,014%) this phono exhibit 0,021%@1V rms, mainly H2 (H3 being almost at background noise level).

At 5 Vrms distortion is only 0,18% with the same harmonics distribution. Not that bad…

Sound is well focused, tonally correct and ample. Less punchy than the C3g but still offering a very detailed and vivid sound. Most obvious qualities are clear (could be dry depending upon second stage tube - Sylvania 6072A) and clean, different of the E88C/E86C combo which, in my memory, was softer and less analytical. This is probably due to a more euphonic sound in a full triode setup. Pentodes are by far more neutral and I understand Telefunken choice for the V76 mic preamp when it is crucial to catch the sound during a performance with a total neutrality. The E83F is an uncompromising tube of exceptional transparency. When I compare to the C3g, the E83F appears slightly less seductive but certainly the most truthful.

After several hours of attentive listening this preamp is an excellent performer on philharmonic music, big band, human voice, cello solo and organ, but not only. It is the only phono that makes me hear or feel the recording studio acoustic. The 2nd stage tube that remains the best (balanced, detailed, neutral) a on any kind of music is the Tungsram E80CC.

One drawback, you need top-notch drivers otherwise listening tends to be boring or tiring. Had the not so good idea to set the JBL 2470 in place of the Altec 288C in my homebrewed speakers. Will never do it again. Every improvement highlights mismatches and weaknesses.

vinyls, vinyls forever.....

Marconi Osram company, a GEC subsidiary, made some of the finest and thought after valves, from the scarce and super expensive PX family to the world famous KT one. These magnificent tubes were implemented for decades in high quality audio gears and remain today a standard in quality and reliability.

Almost every diy’er knows the KT66 and KT88, but very few the industrial DA tubes intended for high power public address amplifiers. Among these one deserves a special attention because of availability (at least for some time…), price (compared to a DA30, 60 or 100) and stunning sound quality. The DA41/CV1076.

A very fine directly heated audio triode with thoriated tungsten cathode and stamped anode (unlike the graphite one of its American relative TZ40). A pair of this powerful tube could deliver up to 175W with 5% distortion in class B push pull.

Of course such a power can’t be reached without some amount of grid current and this is probably why we don’t find much literature or schematic from experimenters. Class A2 appears to be a major obstacle for many amateurs. Implementation is not more complex than A1 if one understands the driver basics requirements.

When a grid becomes positive it acts like an anode and a few electrons emitted by the cathode are attracted by the grid. These electrons return to ground through driver impedance Zout and create a voltage drop in this load. If Zout is too high the driver is unable to raise grid to the desired voltage. Imagine a tube with 1K Zout intended to provide 10mA at +10V, it will only be able of 10V – (1K x 10mA) = 0V ! Ohms law.

It’s simple, smart and works flawless. To determine operating point all we need to know is the grid current at the desired grid voltage. This is usually found in tubes datasheet.

GEC DA41 datasheet don’t give much information about grid current operation. I used the DA42 curves, a very close relative with indirect heated cathode. At Vg +25V/Va +350V, grid draws about 8mA which gives a 3K input impedance.

At that bias point, a 5K anode load appears to be the best power/distortion compromise and I can expect about 8W of great quality.

From the above statement the driving impedance must be at least 10 time lower than Zin, 20 time better. Only a cathode follower achieves this. Additionally it must be capable of some power. The simplest way could be a resistor loaded stage, this commands to raise the final tube cathode voltage with a resistor too, fully decoupled to prevent degenerative feedback, introducing a time constant in circuit and inevitably a phase shift.

The Yamanaka setup avoid this problem and collects electrons directly from the grid. Input impedance of power tube is at the same time the CF load and bias. The DA41 is directly tied to ground, no phase shift in such circuit. The main difficulty is to find a suitable tube that accommodates a 3K load, rather low, while supplying the necessary current to the final stage.

I spent some time consulting my books to find a good contender. To my disappointment very few triodes are usable and the only one that could match, a 6CK4, is hard to find in quantity (I usually pair two tubes from a batch of twelve). Best solution is a triode wired pentode. From half a dozen tested I only kept two, 6V6 and 6K6. The last one is almost unknown, darn cheap and can be considered a scale down 6F6, much better sounding. Despite its small size it’s a linear and powerful tube. Perfect for the task with 150 ohm Zout in triode mode.

Tung Sol and Visseaux, both excellent

Drawing the operating lines on Ip_Up curves seems to place the working point in a non linear region. Not the case. A cathode follower works under 100% feedback and computed characteristics are almost verticals with little curvature. Distortion remains very low. AC swing about +/-25V, enough to completely swing DA41 grid.

Really easy to implement, final bias is adjusted through anode voltage.

Such setup has a gain inferior to 1 and have to be be paired to a high gain, low distortion stage. Low distortion means high DC voltage and high load. I used the power horse 6CG7 double triode. A 150K anode resistor sets gain and a 0,025uF/200K coupling network insures good frequency transmission to the CF stage. Capacitor value can be small with a CF as input impedance is very high. In fact the grid resistor appears to be boostrapped, it’s value being multiplicated by μ+1.

Once working points, loads and DC voltages calculated the amplifier appears very simple. Three tubes, two capacitors and a few resistors.

More to come...

Power supply

My long time favorite for smooth filtering is a LC supply. Not only the mains residual is almost a perfect sine wave, but transients response is better than with a capacitor input. With a large choke value, ripple can be very low. However, to keep first cell time constant around 15mS, the total series resistance must remains low. A well wound power transformer and equally good choke are mandatory. For that purpose, I use top quality Hashimoto’s irons for more than a decade with total satisfaction.

A first cell (25H/50µF) will reduce ripple to Vripple = Vout / 6π² f² LC √2, a second one (180 ohm/100µF) will floor down residual to negligible value.

Some voltage adjustment being necessary on CF anode to precisely set DA41 grid current, I splitted supply in two after the second cell. One arm to the 6K6/6V6, the second to the 6CG7.

Making of

Aluminum CNC chassis, epoxy coated, and a few components

DA41 filaments supply on sub chassis

Amplifier completed, ready for measurements and listening test. It's a beefy unit weighing almost 40 pounds.

DA41 bright light is a pure enjoyment in the dark

Test setup. Max power, 100Hz_10KHz squares and THD

7,8 Vrms/8 ohm at clipping, 7,6W as expected

Despite a 18/20Kohm internal resistance the DA41 adapts quite well to the output transformer and low end roll off is not so obvious during listening test

Short reminder: transformer primary inductance has a direct effect on the low frequency response. The -3dB low frequency cutoff is determined by

F_Low = Z/2πL

Where Z is the primary impedance (reflected impedance in parallel with the tube internal resistance) and L the primary inductance. For a 5Kohm reflected load and 23H@70mA inductance (Hashimoto H20-7U)

F_Low = 28Hz

THD: 2,76% @ 1Watt, 6,06% @ 5 Watt. Nice harmonics distribution, mainly H2

I tried a close loop feedback between DA41 anode and 6K6/6V6 grid to get a flatter 100Hz square signal and it was disappointing. Sound became muddy, loosing all the life music could bring. Regardless of apparent poor scope results I prefer, by far, the amplifier without feedback. We must not lose sight that there is little to none relation between music, which is more in the transients domain, and squares.

Reason why it’s of prime importance to calculate accurately all time constants including power supply to respect signal ADSR.

Listening report

Garrard 301, FR24, Entré EC30, C3g phono preamplifier, 6J5 line preamplifier

On my homebrewed Klipsch Altec system the 6V6 gives the best balance with a great sense of refinement and a silky smooth sound. Each record I spin on my Garrard plays with ease and life. Dynamic is excellent, nervous or soft depending upon music. I love the way a trumpet or a sax is rendered and human voices are really addictive. Low end is firm, ample with a convincing restitution even if it doesn't extend deeply. Music is played without the somewhat arshness heard with transmitting triodes like 809 or 811. It was easy to make a comparison, the PT260 transformer can supply 6,3 or 7,5V. Is it the fact the DA41 was developed as a pure audio tube or the materials used to build it, I couldn’t say. Anyway the GEC sounds much more natural and involving than its counterparts and it blends very well with a 6V6 (Mazda or Westinghouse). The 6K6 is surprisingly more punchy with deeper bass extension, but restitution appears less natural with a forward presentation and could be a better choice on Rock music.

I am somewhat confused and surprised. To this day I don’t know which of the Visseaux A710 or the GEC DA41 is the best of my amplifiers. In any case I listen to both with great pleasure and that’s the most important. Music, music toujours…

The story begins with a FR1/Mk3 cartridge bought from a Japanese seller. This Ikeda San masterpiece brought me to heaven when I heard it for the first time many decades ago. At that time, the poor student I was couldn’t afford such a piece of art.

Years later, I didn’t forget the magic of those moments and decided to find one in decent shape if possible. I was lucky to find one, apparently in good condition, in Akihabara and received a small packet a week later. Despite a worn box due to long-term storage, the cart appears to be mint, not to say unused, with a perfect diamond and suspension

it didn’t take a long time to set it on my Garrard for a ride, fitted on a new Jelco headshell. What to say... music is gorgeous, like in my memory, and the FR1 is still one of the finest cartridges I ever listened to. But I was disappointed.

At the time I heard this cart for the first time, I did not pay attention to the output level. And God, it is very, very low (about half my Entré EC30). It seems to be the way air coil cartridges behave and the price to pay for naturalness and absolute transparency.

It is not a good fit for my C3g phono preamp, which is suddenly short in gain. There is no other choice than a new one dedicated to the FR cartridge. Immediate problem: which tube could give a higher gain while keeping all the C3g qualities? Kind of catch-22.

I spent weeks digging my books and searching the web to finally exhume a totally obscure (at least to me) tube that appears to be the perfect contender in such a project, the British STC 5A/180M beam tetrode. Sourcing a complete datasheet was a nightmare. But it was worth trying.

With a max Gm of 32 mA/V and an internal resistance as low as 25K, it is a dream for a very high gain, full bandwidth, and low noise stage. Much better than a D3a with such a low Rp and on par with an E55L, but at one-third the current.

I was half my way. I found the good specs, now I had to find the tube. After a while, I sourced a supplier in Great Britain and bought a little batch at the correct price. The rest of the story is math and parts. The design is very similar to the C3g phono. Two stages, including the shared current setup I use in preamps as well as a driver in amplifiers. A setup that brought excellent results each time I implemented it.

To meet Lipschitz RIAA recommendations while keeping gain high with reasonable heat dissipation, the tetrode load is calculated accurately. This load is of prime importance to determine the output impedance that will determine the RIAA network component values.

A problem occurred with the grid current, as the datasheet does not provide any information about it. To overcome this, I used the same trick as for the alternate driver tested for the E88C LCR phono preamp. A VR tube is used in conjunction with a resistor to set the G2 voltage. This resistor Rx is calculated to pass a determined current that will split between regulator and grid.

For this purpose, I bought for cheap a few Russian SG5B, tiny little 150V glow discharge tubes.

The current through G2 is really stable as long as the voltage remains constant. This is mandatory with very high-slope tubes where any voltage change induces a great gain variation. One could object to some noise due to gas ionization, but it is far less than a Zener and easily reduced using a small capacitor in parallel. Anyway, it’s the counterpart for a great, dynamic, stable, wide soundstage. It is rarely annoying, the signal being way above VR tube noise. I personally hear nothing on my 105 dB speakers.

To determine the best load, I made simulations from 7.5 to 15K and finally stopped at 10K. It sets gain at 200/220 under 10 mA anode current (180V anode voltage, 150V G2 voltage). Rx is calculated to draw about 10 mA through G2 and VR. Means 13k with a HV supply at 280V. The closest standard value is 12K. The Rk total current will be close to 20 mA, and for VG -1,7 V, its value is 85 ohms.

The second-stage tube is chosen to meet 55 dB of overall gain. A double triode with a mu of 30/35 will do the job. The first idea was the world-famous E80CC, but the gain was too low. This put me in the E180CC family of tubes, but after long listening tests, I found these tubes too clinical-sounding (Amperex 7062), not to say unpleasant (Sylvania 6414). Even the highly prized 12AY7/6072 did not satisfy me. A solution came with another British valve, the 13D3. This nice STC Brimar double triode intended for audio applications has a clear, detailed, and mild voice that blends well with the 5A/180M.

With its medium mu, the gain is close to 24 (47K load) for an overall one of 530, or 54,5 dB. Perfect.

Zout is about 430 ohms, low enough to permit long interconnect cables if necessary.

A few pics

More to come...

Once in a while doesn’t harm, I built a tube-regulated supply. The reason was to precisely set the tetrode anode voltage and see if it changes or not sonic signature compared to a classic CLCLC supply.

Based on a classic 3-tubes schematic, a pass element is coupled to an error amplifier and paired to a constant voltage source. Many tubes are usable, so I dug my stock and picked a beefy 6336A, a tiny 6AU6, and a 0A2 glow discharge tube. Calculation is a bit tedious, but there is abundant literature to help design

Prior to regulation, HV is filtered by a basic CLC network. Time constants have been taken into account to ensure a fast recovery, even on drums, piano or guitar attacks. Never forget it is the heart of any amp or preamp.

The 6336A can draw a large current, and this supply is just limited by power transformer capabilities, here 200 mA. Way enough for this phono! Another very important feature is the very low output impedance.

Making of

On the bench, ready for tests.

Preamp has a very large swing capability without clipping.

THD setup. 0,038% @ 1 Vrms/1KHz (deducted 0,02% from signal generator), mainly H2, other harmonics are at background noise level. Not that bad.

Once the supply was completed, I carried out a comparative study.

I must confess that the regulated supply vs. just a filtered one brought me to a very different perception of music restitution. Soundstage is wider and deeper, no contest, but what hit me the most was the more focused, deeper bass. Music gained weight and intelligibility while being more suave and distinguished. Dynamic is excellent, probably due to a very fast recovery on attacks.

I decided to compare this new phono to the C3g one, which I know quite well. I fed the 5A/180M preamp with Mullard's M8136/CV4003 to be on the same amplification level as the C3g and used the Entré EC30 cartridge.

Have to say that the M8136, old production, deserves its reputation. Probably the best of all the 12AU7/ECC82, along with the 5814A triple mica by Siemens, warm tone, clear and ample voice, and wide soundstage. A must-try. To my ears, no other tube comes close, not even the very desirable TFK ECC802S.

I replaced the Russian SG5B by Raytheon 6542, which is less noisy by a slight margin. Probably because they can withstand more than twice the current.

With very similar gain for both units, A/B tests can be performed easily.

I thought the C3g was the ultimate step in micro-detail restitution. The 5A/180M goes beyond. I never heard voices I perfectly knew with such accuracy and presence (Claire Austin_when your lover has gone). This is due to subtle information like very small breaths, lip movements, or tongue noises I hadn’t heard before. This also happens with instruments and brings music to a new level of authenticity. Unfortunately, this is also true for surface noises, which sometimes bother with poor records. However, this preamp has such a resolution that I can’t stop listening to music encore and encore.

In its dedicated environment (6J5 line preamp, DA41 A2 amplifier, Klipsch_Altec speakers ) this phono does everything with ease. Low end is just amazing with the FR1Mk3 and 13D3 (Ludwig Streitcher_Plays Bottesini), but not only. I played a variety of music, from jazz (for Duke_Bill Berry and Don Wilkerson's_Preach Brother) to classical (Vivaldi Concerto Köln_Shunske Sato, Beethoven sonatas_Paul Badura-Skoda), with pure enjoyment.

A few months ago, I built a 41MXP amp with the intention of using it with my tweeters along with an active crossover. The results were not as expected. Not to say this amp sounds bad, the 41MXP is a clean and clear triode with an enlightening treble. It’s more about the sonic results I had with multi-amplification. The stage appeared thinner, and the overall presentation was less convincing and natural than with my passive Xover.

After careful consideration, I decided to take it apart and use transformers, chokes, and outputs for another project.

It didn’t take long for me to get a nice pair of 6A5G triodes out of my stash. These tubes are said to be fine-sounding triodes, sometimes compared to a 2A3, with a unique internal construction. It’s an indirect-heated tube with a cathode tied around the heaters. What an oddity! I believe it was very hard to build. The tube has been engineered that way with the goal of reducing hum while keeping all the characteristics of its relative, the 6B4G. And it works. According to some people, it’s a delicate and detailed tube. It is a good reason to give it a try, especially since I wonder why I did not use it sooner. In fact, I remember buying them after reading good reports about the Peerless A100A amplifier. But this is another story.

I have completely redesigned the driver. Replaced the ML4 with a 6SN7 in a shared current stage that proved to be an excellent solution, providing high gain under low impedance.

By experience, I found driving a power triode under low impedance to be a good solution for better bass restitution and a more relaxed sound. At least when CF is properly designed to prevent overdrive by the preceding stage.

I kept the Tango anode choke and added a resistor to the circuit. This combined load ensures proper DC voltage while keeping AC gain higher than with a resistor alone. Transients are much better, too. Current transfer in such a complex time constant depends upon the choke/resistor ratio.

τ = L/R

To make a long story short, the smaller τ, the faster the steady state of the circuit of the circuit will be established. Means that changing R will affect the sonic behavior of the amplifier.

It took a lot of time to find the best value, spending hours listening to different setups with loads ranking from 27K to 68K. I finally stopped at 40.2K. Any 39K will do the job, but I had the 40.2K on hand, and these vintage metal films sound very good.

To get the necessary voltages, I replaced the PT220 power transformer with a PT260 and recalculated a few resistors in the HV line to keep time constants as low as usual.

I didn’t exactly bias the 6A5G according to the datasheet. I chose another point with a lower anode voltage and a 750 ohm bias resistor.

The reason is that I wanted to test the French R120, as I have a few samples on the shelf. A similar triode that bears an excellent reputation among tube lovers. Despite a completely different construction, the characteristics are quite similar. A simple bias change will allow me to enjoy these superb triodes, originally intended to fit professional equipment. The recommended value is 600 ohm. I implemented a switch to connect a 3K resistor in parallel with the 750 ohm one.

Filaments are floating due to a center tap connection on the 6A5G.

Wiring is kept as compact as possible using parts I had on hand: Aerovox V161, a longtime favorite; Sprague Koolohm; Allen Bradley's; and Sic Safco low ESR electrolytics. The power supply is fitted with ICAR polypropylene capacitors along with SEL metalized paper.

The 6A5G is renowned for its noise-free restitution, but it is a somewhat intriguing tube, depending on the manufacturer. I have a few Visseaux on hand, and there is a persistent hum, apparently for no reason. Different construction? It’s a pity; they sound better than Sylvania. Mysterious issue.

I made some measurements to see how these tubes perform. The squares are good, and the slight overshoot at 10 KHz is due to transformer ringing. The vintage Tektronix 453 I just bought (what an incredible construction) has a very fine trace and is very pleasant to use despite its small cathode ray tube. Up to the task for this use and much more compact than my imposing 7834.

Maximum power is 3.5W and 3.2W, respectively (6A5G_R120). The harmonic distortion versus power is exactly as expected.

At first glance, they both share good sonic qualities, but after a long time of listening, they show quite different personalities. It depends on music and mood. The R120 is crisp and punchy compared to the 6A5G, which appears a bit mellow. Thus, despite a magnificent medium, the 6A5G proves to be less attractive than the R120 on human voices; the message is slightly confused, and I prefer to keep it for piano or violin, where it adds softness and romanticism. In my opinion, the R120 is better suited for complex music with its clear and detailed reproduction. The 6A5G plays in a more relaxed atmosphere, ideal for ambient music, easy to listen to for hours, but globally lacks punch and vivacity. Conversely, the R120 is more nervous and lifelike, an attention-grabber whatever I spin on my Garrard. This is the tube to go on small jazz formations, vocals, big band, orchestral music...in fact on any kinf of music.

Despite higher harmonic distortion, the R120 sounds better to my ears and will remain the tube of choice in this amplifier.

Driver choice

I tried half a dozen 6SN7/VT231 tubes to finally end on their predecessor. A friend of mine gave me a nice NOS pair of Tung Sol VT99 with adapters. Once plugged in, it became evident that the R120 had found its counterpart, and the search was over. These double triodes excel in any aspect of music restitution: balance, delicacy, precision, naturalness, etc. Only the Ken Rad VT231 outperforms the 6F8G in the low register by a slight margin.

I guess one day I'll have to rewire the sockets to accommodate the different pinout, but I feel a bit lazy. I prefer spending time with music for pure enjoyment.

There was no objective reason to modify the good-working RLC E83F phono preamp except this insatiable curiosity, which is mine. In fact, there is one, and it is a corollary to the development of the 5A/180M phono preamp. When I bought this tube, I also, unreasonably, bought some 5A/152M because I found these ones nice, were manufactured by STC, and at a very affordable price. Once the 5A/180M preamplifier was completed, the results were so satisfying that I wondered if all the STC tubes were as interesting, and I immediately thought of the ones lying dormant in my stock.

It took me about two minutes to find the 5A/152M datasheet and start drawing several load lines. And about the same time to understand that at equivalent load the gain was close to that of the E83F. I also realized it adapts better to the transformer I use. There you go; the damage was done.

A short digression to say that the RLC correction is the only one that can restore with integrity the musical message as it was engraved. More complicated to implement, it has the enormous advantage of a constant impedance vs. frequency, which is not the case with an RC network. The main difficulty is the 600 ohm load, a real challenge for the driver. I choose the classical transformer coupling solution; although expensive, it insures the RIAA network an almost perfect loading. The only drawback is the 300 ohm load seen by the second transformer. This has to be taken into account in the gain calculation.

No compromise during this complete overhaul; I completely stripped the board to leave a clean and empty place. It allowed me to punch a larger hole for the STC valve socket and therefore redesign the whole wiring.

I spent some time finding the best implementation to overcome the hum/noise inherent to pentodes. Must note that this tube has a lower noise equivalent resistance than the E83F (670 ohm vs. 750), which places it in the same league as a C3g. Good point.

I used shielded wires in the signal path, and it effectively reduced hiss. Another necessity is a VERY low ripple HV supply because 5A/152M gain is high and any unwanted hum will be greatly amplified. Grounding is critical and probably the most important point to achieve a quiet preamplifier. Each stage has its own reference point, and the whole preamp is wired using multi-star grounds.

I constructed a regulated supply that is nearly identical to the one used in the 5A/180M RC phono. Just added a second network to floor ripple at about 500 uV prior to regulation.

Partition noise, in fact the only source of noise in this preamp, is only audible with potentiometers set beyond a comfortable level, almost fully clockwise. Will never go that far; music is really loud and painful. At realistic volume, this preamp is totally silent.

There are different ways to build this preamp, and after long listening tests, I personally prefer to set volume with a potentiometer next to the second transformer. Music gains clarity, and the mids/heights appear more refined. As a bonus, noise is reduced.

Good: Classic 600 ohm loading.

Better: Reflected impedance by the transformer's secondary loading.

Regulated supply. Robust device with a 150 mA capacity.

This phono has such a gain it can be directly connected to the power amp. No need of a line preamp. Output impedance stays very low; this ensures very good adaptation with a wide bandwidth and great dynamic. Distortion is really good, 0,1% @ 1Vrms, 0,18% @ 3Vrms, mainly H2.

Of course, my memory is unreliable, but I sincerely think this one sounds better. The Philips/Tungsram E83F was on the clinical, dry side of music reproduction, while the STC 5A/152M appears more relaxed and realistic with a wide soundstage. This tube is full of life and air, never tiring with a clear and clean voice. Apart from the pentode noise that could be objectionable for some people, it’s a very enjoyable setup, and the final restitution can be set upon each taste with some tube rolling. From E80CC for a neutral and authoritative sound to 12BH7A for more mellow, fleshy music. I personaly use the Mullard M8136 with great pleasure.

In order to achieve a marginally better overall distortion (0.09% 1KHz 1Vrms), I swapped out the R12_R14 47K resistors in the second stage with 51K resistors.

My advice is to snag this affordable and easy-to-find gem while supplies exist. Good bottles tend to disappear quickly these days.

Would be nothing without good vinyls....