VT25, VT25A Push Pull rejuvenation Part 1

...SOLD...

This amp have been my music companion for more than two decades and the only one I kept among the few I built until I came across the Philips MC1/60 triode.
It had all the qualities expected from the 10 family of triodes and I especially enjoyed the VT25 Visseaux in this configuration for its rich texture and refined sound.

Both amps remained for a while in my system but listening after listening it became evident that the big triode SE was better in terms of speed and impactness. The VT25 amp finally went to the attic as I have no room for two amps.

I never totally gave up the idea of a better VT25 amplifier but was not in the necessity to ask myself "what makes the difference between these two amps ?" Moreover reflection time being always beneficial to find solutions to a given problem, even unconsciously, I let things slowly growing.


During the past year the few electronic devices I built made me realize the essential contribution of the PSU in the final result, and naturally came to me this question : Is my VT25 amp lack of punch a supply issue ?
So, I put it back on the bench for major modifications including revisited supply, better implementation, different bias point....and new driver.

In my very first prototype the bias point was quite in the middle of the load line and the amp worked almost A₁, current flowing all the time at 22mA. This is not the best for triodes push pull operation, neither for power efficiency (even if I am not really concerned with 104dB speakers) nor distortion and I decided to move AB₁ with a setting current in the 8/10mA range. The biasing resistor was just increased to a 3,5 Kohm value. On the new amp the resistors are 3K, 1% sorted, for a 10/12mA flowing current. For some years now I use the vintage Sprague Koolohm non inductive resistors or Kiwame carbon in tubes cathodes with good results.

I also totally reconsidered my work which was not the most pertinent in term of good implementation: long path from the decoupling caps to the active devices, potentiometer to set input level (excellent for unwanted hum and noise), multiple wiring points (the best for ground loops), poor filament filtering and bad supply time constant. Despite these negative points the amp was performing very decently and I can expect some clear improvements with the right modifications.

Prior to change the power supply and the amp being completely stripped, I modified the driver on one amp. I wanted to take advantage of the very good results of the SRPP stage in the MC1/60 amp and performed some blind test to see (...hear) which driver was the best. The 5687 transformer loaded SRPP is the winner, no discussion. Excellent image and tempo, deeper sound stage, extended low end (no current through IT primary means wider bandwidth). Better linearity and lower distortion are this setup assets when coupled to the Tango NC14. Furthermore the good PSRR of this circuit will help to simplify HV supply. A good point and first step to an improved amplifier.

5687 µ & ρ



5687 SRPP resistor, gain and current calculation



5687 DC load line. Tube will work in a more linear region of characteristics than the previous stage



One minor drawback is a more demanding stage. I need to provide a 2V_rms input voltage when 0,9 where sufficient with the former one to swing the amp full power. However I can switch to the E182CC / 7044 for higher gain if necessary.

Good 5687 and E182CC chosen from my stock for this test

Amplifier revisited.

A 5µF coupling cap will insure wide frequency response with a -6dB cut off @ 6Hz for a 5 Kohm AC load but like for a parafeed load the coupling cap is dependent of the transformer primary inductance and is calculated large enough to avoid resonance at an audible frequency. In this case, considering a 50H primary inductance for the NC14, the resonance frequency f₀ = 1 / 2π √LC is about 10Hz. Lowering this value can give some bass boost to the circuit. Interesting for those who feel the VT25 family of tubes lacks of solid bass extension.



To be continued

E80CC SRPP Line preamplifier Part 3

Power supply

Unlike the 6J5 line preamp I used a CLC filtered supply. I usually prefer the LC for its very smooth ripple behind choke but listening tests demonstrated better dynamic with SRPP. A good calculation of ripple rejection and time constant helps to achieve a fast recovery supply with a minimum of cells. To minimize inter modulation the supply is splitted in two after the second capacitor



Ripple calculation

On C₁ the ripple is estimated by formula     V_C1~ = 10 I/C     where I is the current through circuit (in mA, here 7mA) and C the filtering capacitor (in µF).
V_C1~ about 2.2V_pp or 0.77V_RMS with a time constant of about 7 millisecond.

On C₂ the ripple is calculated by the voltage divider formula    Vout = Vin Z₂/Z₂+Z₁    where Z_L1 is 37.6K @ 100 hz and Z_C2 50 ohm at the same frequency.
V_C2~ is about 1.05 mV_RMS with a time constant of 24 millisecond. Thanks to the 60H choke that smooths out the ripple with great efficiency. A 20H one would have been enough but I had this one on hand.

The same applies on C₃ with R = 2.7K and Z_C3 133 ohm. It leaves a 0.05mV_RMS ripple which represent a -130dB attenuation at a time constant of 33 millisecond.

Parts

Good sounding Philips/RTC EZ81 rectifier along with old Tango Hirata choke and some Siemens MP/JS and F&T capacitors are the parts of choice for this power supply.

Tube choice and tests

Not a long or tedious quest for different brands of tubes to be tested, just two makers for this fine tube; Philips (can bear Valvo, TFK, Siemens and so on names) and Tungsram. I attentively auditioned these two competitors and the result is disconcerting. In this setup the Philips that bears a reputation of excellence is not the winner of the test. The musical rendering appears fuzzy despite some great qualities in terms of tone and speed. It gives a blurry image that leaves the observer perplexed. On the other hand, the Tungsram brings music to a level of clarity and enjoyment seldom heard. It has incredible precision and gives a density to the sound that makes it perfect on any kind of material. Very detailed, not to say clinical sound but on the good side of absolute neutrality. You like it or not but it never leaves indifferent and if you are after audio perfection you get very close to it. Furthermore this tube gives an almost holographic image of the sound stage. Great, great tube !
As expected distortion is very low. I get 0.9% @ 1KHz and 2V_RMS output, mainly second order. Noise floor is also very low and hard to measure on my FFT analyzer, thanks to the very good CMRR (or PSRR, it's the same) of this circuit. In facts we have the same power supply ripple rejection qualities than in parafeed setup and some will consider the SRPP a parafeed cousin with an active load in place of the cumbersome anode choke.

SRPP line preamp completed

...in its industrial blue gray and black look. All front and rear plates anodized aluminium, CNC machined by Schaeffer AG. Outstanding workmanship for a clean, professional appearance.

same connecting organization than the 6J5 line preamp. Outputs are doubled by Lemo 0 coax outlets and this feature will certainly disapear in future works. The little male plug is a nightmare to solder considering the 0.6 mm central pin !

Last minute update...

I made a transformer connection mod. Although I did not notice any sound change, it is intellectually more interesting not to have the C₁ electrolytic decoupling cap in the signal path.



some CD's and LP's I appreciated a lot, and there are many others ...

E80CC SRPP Line preamplifier Part 2

SRPP calculation

The heart of the SRPP is the 3 resistors used in this circuit (the upper one is generally omitted but it is bad practice if you want a perfectly balanced push pull). Easy to calculate they just depend upon tube parameters and AC load.
With a fully decoupled cathode all have the same value which is :

                                               R = Rload + ρ / µ -1,5

Current through tubes can be determined by the following formula

                                          Io = 1/2 [ Vht / 2 ρ + R ( µ +1,5 ) ]             Ohm's law...

Gain calculation is way more complicated and is an extrapolation of the simple triode gain mathematical relation

                  G = µ Rload [ ρ + R ( µ +1 ) ] / ( ρ + R )² + Rload [ 2 ρ + R ( µ +1 ) ]

An Excel file makes a very handy tool to calculate all these parameters. Just has to be filled with tubes characteristics and AC load.



I also use it to determine the DC load RaV1. Drawing this load line on tube curves helps a lot to check linearity and allows to precisely set the working point (NB: This load to not take in consideration voltage drop in the 2 upper resistors, it is necessary to add it to anode voltage found on curves to have the exact Vak value).
This load is Io formula's denominator

                                                 RaV1 = 2 ρ + R ( µ +1,5 )



It is almost like a 47K loaded single tube and data sheet shows that we can expect very low distortion. In the case of a classic anode follower we have 4.1% mainly second order at 50Vrms!. In this preamp we deal with smaller voltages and the even harmonics will be greatly canceled by the SRPP configuration. Furthermore the 1.3K value has been selected using a HP 3561A dynamic analyzer to get the best distortion figure.
The SRPP operation will be very close to the Philips values, with a 57K load we have a 3.07 mA current for a 4V cathode voltage and a calculated gain of 18,11.



Line preamp schematic

Nothing special except an unusual input voltage divider. Gives some headroom on high level sources. Three positions, 0 -3 -6 dB permit a very fine gain control in association with the T pad. I used the excellent GRAYHILL 44 series rotary switches for both input/output selector and input attenuator. The quality is outstanding.
Like in the 6J5 line preamplifier just a few parts. Resistors are Holco H2, Philips (not Vishay) MRS25 and coupling capacitors ITT PMT/2R.

These are to my humble opinion (and to my friends ears...) some of the best caps for the price along with AEROVOX V161 and ERO MKC 1860 / KP1832. They give a very neutral and accurate restitution without spending hundreds on exotic parts for a somewhat slight improvement. I made tests with well-known pure copper or silver film and foil ones. There was an improvement in transparency or finesse on certain parts of the musical message but each time to the detriment of the overall tonal balance .....went back to the ITT's.

Looking inside...

Can't be a more compact wiring

Next step: power supply, fully assembled unit and tests

E80CC SRPP Line preamplifier Part 1

This project was initiated by a request of a line preamp with selectable input sensitivity to match sources with different voltage outputs. This preamp should also have a low output impedance to drive long connecting cables without any loss.
With such goals in mind it became evident that a simple 6J5 preamp would not fit these requirements.

1- The use of two gain control stacked together would greatly impact music transcription.
2- The gain of this preamp would have been too low to accommodate an input voltage divider up to -6dB.

A high gain stage with a voltage divider at input and a volume attenuator at output (which is the best way to control the gain from my humble opinion) was the way to go.

For that purpose there are a lot of good circuits with good linearity and gain, cascode, µ follower, totem pole, push pull. Push pull circuit, seldom used in preamplifier, is very interesting because it cancels all even harmonics. With such a circuit you can expect very low distortion using triodes but a phase splitter (transformer or tube) add complexity to the circuit, increases final cost and is hard to implement in a standard chassis.

One circuit shares this low distortion, high gain capability without the necessity of a splitter stage, the SRPP.

I have been working with for a long time and I know how it works (well), its limitations and drawbacks (very few when correctly understood and used).
Some would argue that it have a special sonic signature, especially on voices, with a kind of emphasis that could be appealing but do not reflects reality.
Believe me, when correctly used and adjusted (a FFT analyzer is of great help) it can be very neutral without being clinical or cold with huge dynamic and stunning resolution and accuracy on complex music.

SRPP (very) short history and use

Originally intended for video (1943) it was widely used to drive low impedance capacitive loads with high gain and wide bandwidth.

SRPP as it should be set up.



As indicated by its name, the SRPP is a push pull even if not evident at first glance.
I won't bother you with any mathematical demonstration because it is not the main topic and there are very good publications that help to understand this circuit, Audio Xpress The Optimized SRPP and Tube Cad SRPP deconstructed.
Never forget it works like a push pull only when loaded and gain, distortion, bandwidth greatly depend upon this load. Moreover its linearity is a direct relation with the flowing current which itself depends upon load.
The good thing is if you have a well defined load that won't vary with the amplifier connected behind, you can fine tune the circuit to get the lowest distortion playing with only 3 resistors. Reason why I choose to load my SRPP with transformer.

Finding a good one was easy because I had some Tamura on hand.
The TKS20 used in this preamp is a 600/10K line input transformer but can be reversed used as there is no current flowing through primary.
Tamura are among the best transformers I had to work with. Unfortunately most are discontinued and prices skyrocket on the second hand market.

Previously I said that the best way to control gain was the use of an attenuator behind the transformer. I was lucky enough to have some Daven 600 ohm dual T pad, perfect for the purpose.
These pieces of engineering are unbelievably well made with plain silver contacts and sorted resistors for perfect tracking. They were built to last a lifetime. The only disadvantage is a 6dB insertion loss that have to be taken in account when calculating overall gain.
These attenuators are very hard to find today, especially NOS, and it is interesting to build its own T or bridged T attenuator. See Pots & Attenuators tutorial.

Tube choice

To reach my goals I needed a tube that gives a minimum overall gain of 4 (attenuator will divide it by 2). This means a SRPP gain of 16,30 (transformer voltage ratio seen by tubes is 4,08) and a tube with a µ of 20/30, a ρ of 7/12 K to correctly match the 10K dynamic (AC) load.
Double triodes like 12BH7A,13D3,E80CC would be good candidates. In this study it is stated that V1 tubes are identicals.
I finally used the E80CC because I had excellent results in the E140 amplifier and it is easy to source compared to the British 13D3. I made some simulations with the 12BH7A but gain was too low.



More to come...

MC1/60, E60M, TM100 SE amplifier - Part 1 & 2

...SOLD...

MC1/60, E60M, TM100 SE amplifier - Part 1

In the DIY world of triodes amplifier lovers, most people know the best of the large power tubes like the 211/VT4C or the 845. These tubes are easy to source, still manufactured by Chinese, Czech or German. It's even possible to find some NOS ones from GE, RCA…
There are plenty of circuits to try for the hobbyist and parts like output transformers are easy to find.
This topic will make you discover some of the best European triodes and especially french ones. For audio use everyone, almost everyone, knows the tubes from Mullard, Philips or Mazda, but very few have had the opportunity to get on Neotron, Visseaux or SFR.

These manufacturers made tubes mostly for administrations and are rarely present on the tubes vendors stalls still active today. In fact, most of these tubes were bought by the Japanese in the 80's and are now selling at indecent prices. Understandable when you see manufacturing quality that may turn you pale. Philips MC1/60 or SFR E60M where built to a very very high standard even for the tube industry of that time.

Let me introduce our competitors.




Philips MC1/60

This Dutch power triode was specifically developed for audio, rated at 75W plate dissipation,
very close in specification to the 211/VT4C, but easier to use due to its lower trans-conductance.
The MC1/60 have a second advantage with its 4 V oxide coated filament rated at 3,3A (to compare with the 10 V/ 3,25A of the 211). We will talk latter on an other advantage of this oxide coated cathode in term of music sweetness.
Sweetness and naturalness shared with the very best of the 4 V triodes like the PX4 or the AD1.






SFR E60M

French military triode made by Société Française Radioélectrique in the 40's. Incredibly well made with cost no object materials, twice the weight of its Philips relative, these tubes can bear the 3X75B military code. Very close in specification to the MC1/60 which is very helpful due to the lack of information concerning the SFR company products.
Like the Philips it's a 4 V / 3,3 A oxide coated cathode, anode power rating 75W.






MAZDA TM100

French military triode made by Mazda in the 30'/40',
also can bear the 3X75B military code.
Very different construction with glass rods to support filament springs.
Despite the old fashion and less impressive construction it is probably the best sounding triode of this group, and probably one of the best sounding ever.
Oxide coating cathode, 4 V / 3,3 A, power dissipation 75W (better not to go beyond 60W).
All these triodes can be switched as they have the same filament ratings, use the same socket and are very close in specifications.


Top view shows the very different construction.

The 6 watt monster amp

My first intention when I decided to build an amp around these tubes was to get a reasonable output power with the best possible sound. You will say this is a very personal and very subjective goal but during the past 20 years I built a lot of amp, preamp but only kept 3 ones after intensive listening test with golden ears friends (10 / VT25 SE and VT25 / VT25A AB1 PP amps, 6J5 / NP206 Tango transformer coupled line preamplifier). These materials serve as references to my future constructions.

My requirements are small in term of power. I mainly listen to Klipsch La Scala speakers upgraded with JBL 2470 drivers and Alk filters and at 105 dB I can shake hell with just 1 watt.
So I decided not to push my tubes too much and make them sing at 5/6 W output.
Looking at the curves I found a satisfying operating point at 600V / 80 mA with a 3.5K load,it may be weird at first sight but at that point internal resistance is mainly 1.6K (S=8mA/V and μ=12.5). This gives the maximum output with a 3.5K load (and I had the transformers on the shelf for a long time, it helped a bit to make a decision).



The driver was a more difficult choice. I needed very low distortion, large RMS output voltage, low output impedance, large bandwidth. Not easy to combine. After several tests with the most commonly used drivers and tubes I stuck to a 6SN7 SRPP feeding a transformer. In that way I had a well defined load for the SRPP, a very good bandwidth despite a coupling cap (no current through the transformer) and the possibility of less final distortion using a 10K/40K step-up transformer that double the driver voltage. Moreover I could use a negative voltage bias and the possibility to set the current depending upon tubes to be used.

The 6SN7 choice was not a choice. This tube is well known for its linearity, its sound qualities and I have plenty on hand.
Long time ago I have modelized the SRPP based on a very interesting thread by Merlin Blencowe from AudioXpress ( The optimized SRPP ) and I just have to fill some fields in a Excel file to have my operating points, resistor values, gain and output impedance.

Now some math considerations to see the different AC voltages involved in this project.

Power output should be 6W/8 ohm or 6.9 Vrms.
Output transformer have an impedance ratio of 2.28 10-3, a voltage ratio 4.78 10-2 (-26,4dB).
It means for the power tube a plate voltage of 144 Vrms or 408 V peak to peak to get my 6.9Vrms.
Looking at the MC1/60 curves this can be done with 48 V peak to peak on the grid (17Vrms).
Gain of power triode is 8.5 (+18.6dB), confirmed by calculation where

                                                  G = µ x Rload/Rload + ρ.



Now to get 17 Vrms for MC1/60 complete anode swing I just need 8.5 Vrms (24 V peak to peak) out of my 6SN7 SRPP due to a 10K/40K interstage transformer used in that circuit. Easily achieved, under 300V anode supply the SRPP circuit give a gain of 13.4 (+22.5dB),means 0,63Vrms at input (1.80 V peak to peak) and very low distortion.





The amplifier overall gain is :
+22.5dB (6SN7) + 6dB (Transformer) + 18.6dB (MC1/60) – 26.4dB (Transformer) = +20.7dB


To be continued......
full amplifier circuit, construction details, listening test, tubes to be used, mods and tweaks.



------------------------------------------------------------------------------------------------------------------



MC1/60, E60M, TM100 SE amplifier - Part 2

Back on the air with amplifier circuit

Voltage requirements.

We need 600V for the final stage, 300V for driver, negative bias voltage, heater voltage.

High voltage is not very high, I can take advantage of vacuum tube rectifier (helps for soft start) and polypropylene capacitor for main supply. I prefer the sound of these caps over the usual electrolytic.

For driver stage I choose choke input filter with bleeder resistor. Choke input is less disturbing for the power transformer and ripple is very smooth.

Main power transformer windings.

1- 500_0_500 V will give me the proper 600V high voltage with GZ34 rectifier.
2- 350_0_350 V will feed the SRPP with 30H choke input filter and ultra fast soft recovery diodes.
3- 50V for bias.
4- 7.5V for MC1/60 to provide 4V choke input filtered too.
5- 6.3V for 6SN7.
6- 5V for GZ34.



On the bench

Sub chassis ease construction a lot.........

Amp finished





Tamura TN351 inter-stage transformer with it's 1:2 voltage ratio helps a lot to drive the final tube easily with minimum distortion from the driver.
The little black switch next to the 6SN7 socket is for a local NFB loop. I can choose between 0 / 3.5dB feedback. Personally I prefer no feedback.
Screw speaker terminal, grey paint and AEI silver plated socket give a retro fashion look to the amplifier.



The second black switch next to AC connector give me the choice to earth or not the amplifier, helps to get the lowest noise floor.



Screwdriver potentiometer and large panel meter for easy current setting.





The mighty E60M tube @ 600V / 75mA




Listening test

Amplifier is dead silent on my Klipsch, I have to put my ear against tweeter to hear a slight hiss.

First impressions .... punchy, very clear and detailed without being tiring. Beautiful and delicate mid and highs (as fine as the 10/VT25 amp !! I can make comparison on mono recordings VT25 amp on one side, MC1/60 on the other), voices are natural, very articulated and living like (Madeleine Peyroux_Careless Love).
Bass are impressive, this is the first time I can physically feel bass with my Klipsch (Bach_ Cello suites, Pierre Fournier). Excellent soundstage with lot of depth, very airy and detailed, I ear the very, very slight shift of the bottle neck on the guitar, the slight fingers taps on strings ( Pedro Soler_Sombras or Selmer 607 Gypsy guitar CD) never heard before! It's good, very good in fact even on very complex music ( Verdi_Messa da Requiem_Giulini).
I also listened to E60M and I have great difficulty in distinguishing the tubes. More body may be? It depends a lot upon input tube. Best with MC1 / 60 are the VT231 Ken Rad (so punchy in the low end), Hytron VT231 or Brimar 6SN7 GTY with E60M or TM100.

Amazing to see after so many years that the 4 V tubes have always been killers compared to the other ones (2.5, 6.3, 7.5V …..). I have been playing with plenty of these classics (2A3, 45, 50,811,211, 845 and so on) and found each time they were inferior to the 4 V ones *. Is it a matter of cathode coating, is 4 V the best voltage for regular emission ? I have no answer, just listening facts that aficionados of the PX4, PX25 and AD1 perfectly know.

* Just one exception, the Visseaux A710, a 10 with BIG plate (as big as a 50) using a 7.5 V oxide coated filament. It's to my knowledge the only 10 of that kind and the only one that sounds like a PX4.

Have to wait for some friends to come and make a very serious listening.
At that time this amp is the very best I have ever built.

Some CD's for this test