The Rocket phono amplifier, Dec 2005.

Edited 2011.
The Rocket still goes on rocking!
Contents :-
Picture of a very good system,
Picture of the Rocket.
Schematic Sheet A and B of the Rocket, Full schematic explanation.
Basic cascode explained with 2SK369 and 6DJ8.
Schematic of reverse RIAA Eq testing circuit.
Picture of chassis underside of Rocket with Wimas and Auricaps
For Rocket power supply, see the PSU details at

One of my clients recently asked me if he could get better music with a
tubed preamp as an alternative to his USA made Sutherland  phono amp.
The Sutherland is all solid state using op-amps, battery powered, and cost
$3,300 in about 2004.

I thought I could do better with tubes, but I would have to use just the
ONE solid state device in the signal path, a single 2SK369 j-fet as the
phono input device.
My client had enough quality elsewhere in his system to allow the benefits
of a better phono amp to be discernible.
He has a nicely sized room which is well carpeted and furnished to reduce
reverb and glare from too much treble. His speakers in 2005 were
Vienna Acoustic Mozart, with Townsend super tweeters, and a custom
built subwoofer which I built. Sub amp is Musical Fidelity A3CR.
The power amps are 22 watt mono bloc SEUL 13E1 that I built in 1997,
upgraded for 2003, and the line stage preamp is also one of mine.
The line stage was upgraded in 2005, and its power supply re-built to power
both line stage and the Rocket phono stage.
His vinyl replay uses a Michel Orb TT with an extensive sand and wire
damping kit from a dude in Denmark, the cartridge is Helicon Lyra.
Cabling is thin wire home brew plaited for interconnects. Speaker cables
are Nordost. All up, this system is one of the few I am very happy to spend
hours in front of, and despite my client's pair of good quality silver disc
players for CD, SACD, DVD, even including a Sony player with
Allen Wright's special upgrade.
We both find that vinyl sounds better, and is more emotionally engaging,
unless the recording isn't good.
It is remarkable how often they got it right in the studio/venue in the analog
1970s, and they cannot do much better now when surrounded by digital
gadgets, and even with 96/24bits, and now, hard drives and goodness knows

Gary's system was extremely listenable.....

Directly below the Michel TT is a Marantz tuner, then the "Rocket" phono
amp I built.
The line stage preamp and power supply for both line and phono stages are to
the left of the silver colored Marantz tuner.

phono amplifie, dec 2005.
The "Rocket" six tube phono amp, Dec 2005. It rocks!

Sheet A amp schematic, 2005, edited 2011.

General features :-
The phono amp has
j-fet + triode cascode input stage followed by triode
-follower output stage.

Passive RIAA eq filter in one network is used between the two gain stages.

This circuit gives low noise & distortion, wide bandwidth, and good music.
The output -follower stage gives Rout < 1k0 and will drive any tubed or
solid state preamp or sound card for recording vinyl onto a CD.

The circuit can be matched to suit the type of phono cart used, MM or MC.

For MM :-
The link "MC link" is not used.
R1 is not needed, so nothing is connected where R1 is shown. C3 may
need adjustment to comply with cartridge specifications and turn table C.

Without the link, gain at 1kHz = 45dB, and will suit all MM carts including those
with rated 1kHz Vo < 1.5mV, giving amp Vo at 1kHz < 270mV.

For MC :-
The "MC link" is used.

R1 should be used if MC cartridge manufacturer recommends it.
Usual values of R1 can be from 100r to 1k0. MC carts have source R < 50r,
and their output level is not much altered by R1. With MC link amp gain
is up to +57dB at 1kHz, so an
MC cart rated for a low 0.2mV at 1kHz will
give 140mV at amp Vo.
I have used a Denon DL-103 MC with rated Vo = 0.4mV and I get 280mV
at amp Vo.

The noise produced by the j-fet input is at least -20dB below what one gets with a
well selected 12AX7 or 6DJ8 etc. Hence there is no need for a step up transformer for
MC use. The j-fet noise is a soft hiss with mainly HF content which is much reduced
by the RIAA filter. Tube noise has much more LF content which is not filtered as much
by RIAA filter.

Technical details for those with technical education and experience :-
The cascode input.
The input 2SK369 j-fet has an unusual R network with R3,4,5,6,7,8,9,10,11 arranged
between j-fet source and shunt regulated -6.2Vdc rail, and all smoothed with C1,2,11.
The 6.2V zener diode noise is suppressed well enough by the RC filtering.

The R network allows for a minimum effect of electrolytic bypass caps in the j-fet
input circuit, and the gate of the j-fet is biased at 0V so no high value input C
are needed between input and gate. The R network stabilizes Idc flow through V1
triode and j-fet. There is an earlier page about evolution of the cascode input stage
at 10 Tube Preamp.
There is much at the earlier page about basic cascode principles and use of 2SK369.
The schematic above for Rocket input stage is set up so operation is for high gain

to suit MC with link soldered in. The j-fet source resistance comprises the resulting total
R formed by R3, 4, 5, 6 , 7 and 8 which
is 43.6 ohms.

If the link is removed, Rs becomes higher to increase local current FB which
reduces gain becomes lower to suit MM cart.

The load seen by the V1 tube varies from a maximum of 23.5k (ohms) at very
low F to about 15k at 1kHz  due to the nature of the impedance of the RIAA
filter which varies with F.

The second phono amp gain stage is -follower. Its basic function compared to
SRPP is more fully described at
10 Tube Preamp.
In this amp I have used paralleled 6CG7 upper triode and 12AT7 lower triode which
are happy with their equal Idc and the result combine the high gain of 12AT7
with ideal Rout character of 6CG7. 

Sheet B regulator schematic, 2005, edited 2011.
The phono amp power supply is from a separate chassis which provides power to
two chassis, one for line level gain and other for Phono amp.

Cascode explained.
Here is some detail for those able to understand more than basics
of cascode operation.
This phono input amp schematic includes all necessary calculations.
In this case V1 paralleled 6DJ8 has its two triodes paralleled. Each
has 2.6mAdc giving Ra = 5k0, = 33 and gm 6.6mA/V.
For the paralleled 6DJ8, it acts like a single triode with = 33, Ra = 2k5
Gm = 13.2mA/V.
( Ra and Gm vary considerably with the anode current and = Gm x Ra,
and is the least variable parameter; please examine the Ra data curves
for 6DJ8 / 6922. )

The 6DJ8 is working in common grid aka grounded grid mode and gain
formula is slightly different to common cathode. The common grid gives
gives slightly higher gain than common cathode mode.

The 6DJ8 cathode Rin is low and = RLa / A = 15k0 / 29.14 = 515r. Unlike a
a gate input with Rin = megohms, the cathode input current = anode signal
current. The 2SK369 drain RLd = 515r. The Rs = 44r, so total load of
2SK369 = 515r + 44r = 559r. For 2SK369, internal Rd = 80k and is so much
more than RLd it may be neglected for gain calcs so gain 2SK369 = Gm x total
fet RL = 0.04 x 559 =
22.36. The Rs is unbypassed so it gives local current FB.
The j-fet gain with LCFB = A' = / ( 1+ [ A x (Rs / [Rs + Rd] ) = 8.1014.
Thus fet gain is reduced by factor A' / A = 8.10 / 22.36 = 0.36 = -10dB.
Vg-0V gate input with LCFB = Vds / A' = 0.559V / 8.1 = 0.069V.

The -10dB A reduction reduces j-fet THD by factor 0.36. THD in 6DJ8 is
very much reduced because j-fet Rd acts as very high cathode Rk
= Rd + ( Gm x Rd x Rs ) = 80,000 + ( 0.04 x 80,000 x 44r ) = 220k.
Thus the 6DJ8 produces little THD, the j-fet produces most of the measured

The cascode stage is very effective where we wish to avoid the anode to grid
Miller C. With statically measured C, 2SK369 data says gate input
Cgs = 75pF
and reverse transfer C ie, Cgd = 15pF. Without LCFB the Rocket j-fet gain
Vg-d = 0.515 / 0.025 = 20.6 so Cin at gate = Cgs + A x Cgd = 75pF + 309pF
= 384pF. This is much to high for MM, and minimum Rs = 44r is used which
suits nearly all MC carts or low Vo MM.

With LCFB gain Vg-d = 7.3 so Cgd becomes 7.3 x 15pF = 110pF and Cgs
becomes lower = static Cgs x Vgs / Vin = 75pF x 0.025 / 0.069 = 27pF so
total Cin at gate = 27pF + 110pF = 137pF.
This input C has negligible effect on HF bandwidth where MC cart Zin < 50r.
But MM carts mostly have unknown Zout and must rely on a cart maker's
specified Cin for a flat HF response.

If a 2SK369 is used for a simple SE common source gain stage with with
RLd = 3k0, then A = 120, and without local current FB the Cin = 1,875pF
which has unforgivable effects on audio HF response.

There is a huge amount of non precise info on cartridge properties including their
loading. But a good start to your study may be at
Fig 3 at this page deals with measuring a typical sample MM cartridge Z out.
I calculated 1.88k at 1kHz, and the use of 47k standard loading R for MM
will not cause much reduction of cart output at 1kHz. I recall measuring MM cart
Vo drop at 1kHz when 47k was connected and concluded cart output Z was
between 2k0 and 5k0.
The Input C for most MM carts is specified by the maker and is between 50pF
at 500pF, with many at 100pF to 220pF which includes turntable cable C + amp
input C.
The total shunt C across MM input has large effects on HF response above 8kHz.
Too much C will cause a resonant peak in MM response and in theory, if I have a
1.5M input cable with C = 100pF, then the input C in Rocket circuit will total about
237pF which suits a Shure V15, but not suit other carts, where less C is desirable,
so it is better to try to use the lowest C cables you can find, and trim C3 at Sheet A
Rocket schematic above.

To best suit MM use with the Rocket, the MC link is disconnected at one end and
this much increases Rs, and reduces gain and reduces Cin to allow higher C in
turntable cabling.
The Rocket has only two gain settings for gain which I think is enough variation.
Most who want real hi-fi with a Rocket will only use MC.

Data curves for 2SK369

The Rd curves for 2SK369 show useful regions for fairly linear Id vs Vgs
when the idle Q point is Vds = +10Vdc, Id = 5.2mAdc.

For RLd = 530r, equal +/- 0.1V Vgs change gives Id swing = +4.1mA,  - 3.1mA.
Total Id change = 9.3 - 2.1 = 7.2mA pk-pk = 2.5mArms.

The 2H % =  100% x 0.5 x ( difference Id change / sum of Id change )
= 100% x 0.5 ( [4.1 - 3.1] / [4.1 + 3.1] ) = 13.89%.

This seems a lot but the 2H% reduces in proportion to Id change and if Vo =
0.72mA pk-pk change 2H < 1.39%. In the Rocket, if 2SK369 gain with LCFB = 8.1,
and at 2.5mArms with 559r load the THD is reduced x0.36 from 13.9% to 5%.
If Vg-0V gate input = 0.5mVrms, Vds = 4.05mVrms and with RL = 559r the Id =
0.0072mArms so expect THD < 0.0072 / 2.5 x 5% = 0.014%, low enough to
never have to worry about it.

When the amp is used for MM cartridges with a typical rated output of 5mV
at 1kHz and without the gain link, Rs value is 259 ohms, the amount of current
LCFB is much increased, and gain & THD is reduced.

The "equivalent input noise resistance", EINR, of a j-fet approximately
= 0.7 / gm, where 0.7 is a constant given in reputable literature such as the
British radio amateur handbooks. Gm is in Amps per Volt.
So EINR for 2SK369 = 0.7 / 0.04 = 17.5 ohms.

The EINR is considered Equivalent Input Noise Resistance which is an R in series
with input to a perfect noiseless amplifier. Noise is difficult to measure inside the
device and is usually measured at the output, then input noise = output noise / gain.
Input noise is mainly Johnson noise in EINR, and other noise adds to total
at output such as shot noise in triodes and partition noise in pentodes.
The noise varies in proportion to the square root of the EINR, so that if the EINR
is increased 4 times, then noise doubles. Or, EINR must be reduced by 1/4 to
halve the noise.

A triode EINR = 2.5 / gm, so if the triode is a 12AX7, EINR = 2.5 / 0.0016
= 1,562 ohms.
The j-fet EINR = 0.0112 times the 12AX7's EINR. The input noise varies in
proportion to the square root of EINR, so that if the EINR difference is 0.0112,
then the noise difference = sq.rt. of 0.0112 = 0.105.
Thus we could expect the j-fet to produce 1/10 of the noise of a good 12AX7,
ie, the j-fet will be 20dB quieter than the triode.

In practice using a high transconductance j-fet reduces noise more than theory
predicts because j-fet noise has less LF content than tubes and is far more benign,
especially with phono amps where the RIAA filter has low filtering at LF.
The j-fet
noise is a very high pitched hiss, and at a level way below the triode
"barbecue noise".

People say j-fets are fragile devices and I would agree, but used in phono
stages at such low levels they seem to last well, but a cap between
gate and 0V, C3, probably does well to shunt any static V discharges. 

I found that among 4 randomly chosen samples of 2SK369, the Gm
matching was within 0.5%. This results in almost identical gains of the two
stereo channels of the Rocket in another amp I built in 2004.
The Rocket does have adjustment for channel gains with 20k linear pot in
series with R20 130k on Sheet A above. These pots do cause some loss of
gain as do all balance adjusting circuits do but the gain loss is only 1dB in mid
pot position and it is unlikely more than 2 dB balance difference will exist.

I found it extremely difficult to calculate the exact RIAA filter values but I did
get approximate values which I could start with. Then I used a reverse RIAA
filter between my signal gene and the preamp, and I adjusted the network
R values to those I have using series/parallel arrangements of standard R.

Providing my reverse RIAA filter is correct, then trimming R in amp RIAA filter
for the flattest sine wave response and best 1kHz square wave allows
for slight tolerance deviations in R or C values. With persistence, the response
was brought within +/- 0.25dB of dead flat along the AF band.

My RIAA filter differs slightly to most other folks "single filter network because
it has an additional R17 22k added in after the R16 + R18 + C7 3180uS and
318uS filter formed by R39, R18, and C7.
The R17 + R19 + C8 act to give 75uS and about 3uS time constant filtering,
and the R 17 builds the 75uS filter out from the R19 + C8 to help isolate the
interactive effects. This arrangement shown was easier to get to within 0.25dB
of the wanted flat response than when omitting R17 and using just a lower
value cap from the top of R18 to 0V with a smaller value R in series for the
50kHz or 3.18uS time constant. Some people use a 3 stage amp with RIAA
3180uS and 318uS attenuation done after stage 1, then 75uS done after stage 2,
and before buffer stage 3, to get the most non interactive and accurate
RIAA eq, but here I have done it by simply adding some resistance between
the 318uS and 75uS eq. it is simpler, and the V2 gain tube isn't subject to
large HF signals; its input has a flat F response so there is less distortion.


The results of testing the accuracy of a phono amp will depend on the flat
response of the signal generator sine waves and its low distortion.
Its output resistance must be less than 100 ohms lest you get some errors
at above 10kHz. The Reverse RIAA filter has R3 = 3k3 to limit the amount
of boost to HF above about 50kHz.
Most practical cutting head amps have this extra HF time constant of
about 3uS.

Rocket Regulation. 
In my own 10 tube 2004 preamp I didn't use any B+ regulation.
However in the Rocket I used an emitter follower style basic regulator at
least for the B+ commonly supplied to both channels.
This is in addition to the large value RC filter caps. The emitter follower with
BU208A plus MJE340 in a darlington pair was a simple way to achieve good
ripple rejection and LF stability, and prevent cross channel talk.
The floating 12.6 Vdc applied to the heaters is shunt regulated by the two
low voltage npn and pnp power transistors. The heater voltage is also biased
and held steady with a cap to 0V.

V2 Gain.
V2, 12AT7, is the second gain tube with gain approx = 45, and its anode DC
supply comes from V2 cathode with R26 8k2. R25 330r is used for a small amount
of LCFB. This -follower can generate Vo > 50Vrms but at 10Vrms THD < 0.2%.
At Vo = 0.2Vrms, THD < 0.005%, mainly all 2H, and below the noise floor.

The use of parallel twin triodes for each triode element in the circuit means that
there is minimal interaction between channels or different parts of the circuit.
The parallel halves of a twin creates a single triode with same but Ra = 1/2
of one triode, and double the gm. The parallel triodes give best sound IMHO.

12AT7 has copped huge criticism from "hi-fi cognescenti" over past years
but I find that most precious prejudices against many some tubes are just BS.
Too many ppl have told me 12AT7 sounds just fine.

If you don't try different brands of tubes, you won't know which are the best.
But yet I see no reason to try every tube in the universe; it is difficult
to make a bad sounding triode amp.

I heard ppl say they preferred NOS Mullards, RCA, Siemens, Telefunken, Sylvania
et all, but unless there's something wrong with the tube, it is aged, gassy, noisy, or
microphonic, or bad samples recently made in Russia or China, then the tube sound
differences will be minimal, so I suggest you just enjoy the music rather than be
obsessive about the equipment. Constant change of tubes will make pin grippers
in tube sockets looser, and you can get unwanted intermittencies.

Internal links away from tube sockets for filament heater wiring allows heater
voltages from 12.6Vdc or 6.3Vdc depending on the tubes chosen without
needing to alter soldered wires at lugs of tube sockets.  Concerning 9pin
miniature tubes and sockets, 12XXX tubes need 12.6V applied across socket
pins 4&5, While many 6XXX tubes use 6.3V across 4&5.
12XXX twin triodes can have  6.3V applied across 4&5 connected together, and pin 9.
If you used a 6DJ8 instead of 12AT7, you need to make the change to heater circuit
and at least that is possible. Of course ANY change to ANYTHING by DIYers
with limited knowledge can cause serious malfunction, poor sound, and smoke.

The Rocket was supplied with Wima coupling capacitors initially. After some
months with Wimas a trial of Auricaps was carried out with the line
level preamp. One channel of the line stage had Wimas as originally supplied,
one with Auricaps. A mono source of signal from a CD player was used as a
source and switched from one channel to the other. The mono output signal
from either channel was then taken to both power amps and a mono signal
played from both speakers. Both myself or my client could not pick which
channel had the Auricaps more than 50% of the time and I decided there
was no difference in the sonic performance using either Wimas or Auricaps.
Rocket underside

Rocket underside auricaps.

Only the coupling caps in the signal path were changed to Auricaps.
Bypassing of PSU electro-caps and parts of phono circuit prone
to RF oscillations was done with polyester caps. I believe such rail bypassing
caps have negligible effect on the great sonic character of such tube amps.
I remain doubtful Auricaps improve the sound and heard no reason to change
my own amps from red-box Wimas. But the customer was always king,
and when asked for Auricaps, he got them.

For power supply details see Nemo line preamp and power supply.

Back to index page.